Система Orphus
Главная > Раздел "Биология" > Полная версия


The Structure Of
Evolutionary Theory







Copyright © 2002 by the President and Fellows of Harvard College

All rights reserved

Printed in the United States of America

Library of Congress Cataloging-in-Publication Data

Gould, Stephen Jay.

The structure of evolutionary theory / Stephen Jay Gould.

p.    cm.

Includes bibliographical references (p. )

ISBN 0-674-00613-5 (alk. paper)

1. Evolution (Biology)    2. Punctuated equilibrium (Evolution)    I. Title.

QH366.2.G663   2002

576.8—dc21                   2001043556

Sixth printing, 2002



For Niles Eldredge and Elisabeth Vrba

May we always be the Three Musketeers

Prevailing with panache

From our manic and scrappy inception at Dijon

To our nonsatanic and happy reception at Doomsday

All For One and One For All




Chapter 1:
Defining and Revising the Structure of Evolutionary Theory   1

Part I, Chapters 2-7
The History of Darwinian Logic and Debate   91

Segue to Part II

Part II, Chapters 8-12
Towards a Revised and Expanded Evolutionary Theory   593


Illustration Credits



Expanded Contents

Chapter 1: Defining and Revising the Structure of Evolutionary Theory
• Theories Need Both Essences and Histories   1
• The Structure of Evolutionary Theory: Revising the Three Central Features of Darwinian Logic   12
Apologia Pro Vita Sua   24
A Time to Keep   24
A Personal Odyssey   33
• Epitomes for a Long Development   48
Levels of Potential Originality   48
An Abstract of One Long Argument   53

Part I: The History of Darwinian Logic and Debate

Chapter 2: The Essence of Darwinism and the Basis of Modern Orthodoxy: An Exegesis of the Origin of Species
• A Revolution in the Small   93
• Darwin as a Historical Methodologist   97
One Long Argument   97
The Problem of History   99
A Fourfold Continuum of Methods for the Inference of History   103
• Darwin as a Philosophical Revolutionary   116
The Causes of Nature's Harmony   116
Darwin and William Paley   116
Darwin and Adam Smith   121
The First Theme: The Organism as the Agent of Selection   125

The Second Theme: Natural Selection as a Creative Force   137
The Requirements for Variation   141
Copious   141
Small   143
Undirected   144
Gradualism   146
The Adaptationist Program   155
The Third Theme: The Uniformitarian Need to Extrapolate: Environment as Enabler of Change   159
• Judgments of Importance   163

Chapter 3: Seeds of Hierarchy
• Lamarck and the Birth of Modern Evolutionism in Two-Factor Theories   170
The Myths of Lamarck   170
Lamarck as a Source   174
Lamarck's Two-Factor Theory: Sources for the Two Parts   175
The First Set: Environment and Adaptation   176
The Second Set: Progress and Taxonomy   179
Distinctness of the Two Sets   181
Lamarck's Two-Factor Theory: The Hierarchy of Progress and Deviation   175
Antinomies of the Two-Factor Theory   189

• An Interlude on Darwin's Reaction

• No Allmacht without Hierarchy: Weissman on Germinal Selection
The Allmacht of Selection   197
Weismann's Argument on Lamarck and the Allmacht of Selection   201
The Problem of Degeneration and Weismann's Impetus for Germinal Selection   203
Some Antecedents to Hierarchy in German Evolutionary Thought   208
Haeckel's Descriptive Hierarchy in Levels of Organization   208
Roux's Theory of Intracorporeal Struggle   210
Germinal Selection as a Helpmate to Personal Selection   214
Germinal Selection as a Full Theory of Hierarchy   219

• Hints of Hierarchy in Supraorganismal Selection: Darwin on the Principle of Divergence
Divergence and the Completion of Darwin's System   224
The Genesis of Divergence   232

Divergence as a Consequence of Natural Selection   234
The Failure of Darwin's Argument and the Need for Species Selection   236
The Calculus of Individual Success   238
The Causes of Trends   240
Species Selection Based on Propensity for Extinction   246
Postscript: Solution to the Problem of the “Delicate Arrangement”   248

• Coda

Chapter 4: Internalism and Laws of Form: Pre-Darwinian Alternatives to Functionalism

• Prologue: Darwin's Fateful Decision

• Two Ways to Glorify God in Nature
William Paley and British Functionalism: Praising God in the Details of Design   262
Louis Agassiz and Continental Formalism: Praising God in the Grandeur of Taxonomic Order   271
An Epilog on the Dichotomy   278

• Unity of Plan as the Strongest Version of Formalism: The Pre-Darwinian Debate
Mehr Licht on Goethe's Leaf   281
Geoffroy and Cuvier   291
Cuvier and Conditions of Existence   291
Geoffroy's Formalist Vision   298
The Debate of 1830: Foreplay and Aftermath   304
Richard Owen and English Formalism: The Archetype of Vertebrates   312
No Formalism Please, We're British   312
The Vertebrate Archetype: Constraint and Nonadaptation   316
Owen and Darwin   326

• Darwin's Strong but Limited Interest in Structural Constraint
Darwin's Debt to Both Poles of the Dichotomy   330
Darwin on Correlation of Parts   332
The “Quite Subordinate Position” of Constraint to Selection   339

Chapter 5: The Fruitful Facets of Galton's Polyhedron: Channels and Saltations in Post-Darwinian Formalism

• Galton's Polyhedron

• Orthogenesis as a Theory of Channels and One-Way Streets: the Marginalization of Darwinism   351
Misconceptions and Relative Frequencies   351
Theodor Eimer and the Ohnmacht of Selection   355
Alpheus Hyatt: An Orthogenetic Hard Line from the World of Mollusks   365
C.O. Whitman: An Orthogenetic Dove in Darwin's World of Pigeons   383

• Saltation as a Theory of Internal Impetus: A Second Formalist Strategy for Pushing Darwinism to a Causal Periphery
William Bateson: The Documentation of Inherent Discontinuity   396
Hugo de Vries: A Most Reluctant Non-Darwinian   415
Dousing the Great Party of 1909   415
The (Not So Contradictory) Sources of the Mutation Theory   418
The Mutation Theory: Origin and Central Tenets   425
Darwinism and the Mutation Theory   439
Confusing Rhetoric and the Personal Factor   439
The Logic of Darwinism and Its Different Place in de Vries' System   443
De Vries on Macroevolution   446
Richard Goldschmidt's Appropriate Role as a Formalist Embodiment of All that Pure Darwinism Must Oppose   451

Chapter 6: Pattern and Progress on the Geological Stage

• Darwin and the Fruits of Biotic Competition
A Geological License for Progress   467
The Predominance of Biotic Competition and Its Sequelae   470

• Uniformity on the Geological Stage
Lyell's Victory in Fact and Rhetoric   479
Catastrophism as Good Science: Cuvier's Essay   484
Darwin's Geological Need and Kelvin's Odious Spectre   492
A Question of Time (Too Little Geology)   496
A Question of Direction (Too Much Geology)   497

Chapter 7: The Modern Synthesis as a Limited Consensus

• Why Synthesis?

• Synthesis as Restriction
The Initial Goal of Rejecting Old Alternatives   505

R. A. Fisher and the Darwinian Core   508
J. B. S. Haldane and the Initial Pluralism of the Synthesis   514
J. S. Huxley: Pluralism of the Type   516

• Synthesis as Hardening
The Later Goal of Exalting Selection's Power   518
Increasing Emphasis on Selection and Adaptation between the First (1937) and Last (1951) Edition of Dobzhansky's Genetics and the Origin of Species   524
The Shift in G. G. Simpson's Explanation of “Quantum Evolution” from Drift and Nonadaptation (1944) to the Embodiment of Strict Adaptation (1953)   528
Mayr at the Inception (1942) and Codification (1963): Shifting from the “Genetic Consistency” to the “Adaptationist” Paradigm   531
Why Hardening?   541

• Hardening on the Other Two Legs of the Darwinian Tripod
Levels of Selection   544
Extrapolation into Geological Time   556

• From Overstressed Doubt to Overextended Certainty
A Tale of Two Centennials   566
All Quiet on the Textbook Front   576
Adaptation and Natural Selection   577
Reduction and Trivialization of Macroevolution   579

Segue to Part II

Part II: Towards a Revised and Expanded Evolutionary Theory

Chapter 8: Species as Individuals in the Hierarchical Theory of Selection

• The Evolutionary Definition of Individuality
An Individualistic Prolegomenon   595
The Meaning of Individuality and the Expansion of the Darwinian Research Program   597
Criteria for Vernacular Individuality   602
Criteria for Evolutionary Individuality   608

• The Evolutionary Definition of Selective Agency and the Fallacy of Selfish Genes

A Fruitful Error of Logic   613
Hierarchical vs. Genie Selectionism   614
The Distinction of Replicators and Interactors as a Framework for Discussion   615
Faithful Replication as the Central Criterion for the Gene-Centered View of Evolution   616
Sieves, Plurifiers, and the Nature of Selection: The Rejection of Replication as a Criterion of Agency   619
Interaction as the Proper Criterion for Identifying Units of Selection   622
The Internal Incoherence of Gene Selectionism   625
Bookkeeping and Causality: The Fundamental Error of
Gene Selectionism   632
Gambits of Reform and Retreat by Gene Selectionists   637

• Logical and Empirical Foundations for the Theory of Hierarchical Selection
Logical Validation and Empirical Challenges   644
R. A. Fisher and the Compelling Logic of Species Selection   644
The Classical Arguments against Efficacy of Higher-Level Selection   646
Overcoming These Classical Arguments, in Practice for Interdemic Selection, but in Principle for Species Selection   648
Emergence and the Proper Criterion for Species Selection   652
Differential Proliferation or Downward Effect?   652
Shall Emergent Characters or Emergent Fitnesses Define the Operation of Species Selection?   656
Hierarchy and the Sixfold Way   673
A Literary Prologue for the Two Major Properties of Hierarchies   673
Redressing the Tyranny of the Organism: Comments on Characteristic Features and Differences among Six Primary Levels   681
The Gene-Individual   683
Motoo Kimura and the “Neutral Theory of Molecular Evolution”   684
True Genie Selection   689
The Cell-Individual   695
The Organism-Individual   700
The Deme-Individual   701
The Species-Individual   703

Species as Individuals   703
Species as Interactors   704
Species Selection as Potent   709
The Clade-Individual   712

• The Grand Analogy: A Speciational Basis for Macroevolution
Presentation of the Chart for Macroevolutionary Distinctiveness   714
The Particulars of Macroevolutionary Explanation   716
The Structural Basis   716
Criteria for Individuality   720
Contrasting Modalities of Change: The Basic Categories   721
Ontogenetic Drive: The Analogy of Lamarckism and Anagenesis   722
Reproductive Drive: Directional Speciation as an Important and Irreducible Macroevolutionary Mode Separate from Species Selection   724
Species Selection, Wright's Rule, and the Power of Interaction with Directional Speciation   731
Species-level Drifts as More Powerful than the Analogous Phenomena in Microevolution   735
The Scaling of External and Internal Environments   738
Summary Comments on the Strengths of Species Selection and its Interaction with Other Macroevolutionary Causes of Change   741

Chapter 9: Punctuated Equilibrium and the Validation of Macroevolutionary Theory

• What Every Paleontologist Knows
An Introductory Example   745
Testimonials to Common Knowledge   749
Darwinian Solutions and Paradoxes   755
The Paradox of Insulation from Disproof   758
The Paradox of Stymied Practice   761

• The Primary Claims of Punctuated Equilibrium
Data and Definitions   765
Microevolutionary Links   774
Macroevolutionary Implications   781
Tempo and the Significance of Stasis   782
Mode and the Speciational Foundation of Macroevolution   783

• The Scientific Debate on Punctuated Equilibrium: Critiques and Responses   784
Critiques Based on the Definability of Paleontological Species   784
Empirical Affirmation   784
Reasons for a Potential Systematic Underestimation of Biospecies by Paleospecies   789
Reasons for a Potential Systematic Overestimation of Biospecies by Paleospecies   792
Reasons Why an Observed Punctuational Pattern Might Not Represent Speciation   793
Critiques Based on Denying Events of Speciation as the Primary Locus of Change   796
Critiques Based on Supposed Failures of Empirical Results to Affirm Predictions of Punctuated Equilibrium   802
Claims for Empirical Refutation by Cases   802
Phenotypes   802
Genotypes   810
Empirical Tests of Conformity with Models   812

• Sources of Data for Testing Punctuated Equilibrium
Preamble   822
The Equilibrium in Punctuated Equilibrium: Quantitatively Documented Patterns of Stasis in Unbranched Segments of Lineages   824
The Punctuations of Punctuated Equilibrium: Tempo and Mode in the Origin of Paleospecies   839
The Inference of Cladogenesis by the Criterion of Ancestral Survival   840
The “Dissection” of Punctuations to Infer Both Existence and Modality   850
Time   851
Geography   852
Morphometric Mode   852
Proper and Adequate Tests of Relative Frequencies: The Strong Empirical Validation of Punctuated Equilibrium   854
The Indispensability of Data on Relative Frequencies   854
Relative Frequencies for Higher Taxa in Entire Biotas   856
Relative Frequencies for Entire Clades   866
Causal Clues from Differential Patterns of Relative Frequencies   870

• The Broader Implications of Punctuated Equilibrium for Evolutionary Theory and General Notions of Change   874
What Changes May Punctuated Equilibrium Instigate in Our Views about Evolutionary Mechanisms and the History of Life?   874
The Explanation and Broader Meaning of Stasis   874
Frequency   875
Generality   876
Causality   877
Punctuation, the Origin of New Macroevolutionary Individuals, and Resulting Implications for Evolutionary Theory   885
Trends   886
The Speciational Reformulation of Macroevolution   893
Life Itself   897
General Rules   901
Particular Cases   905
Horses as the Exemplar of “Life's Little Joke”   905
Rethinking Human Evolution   908
Ecological and Higher-Level Extensions   916
Punctuation All the Way Up and Down? The Generalization and Broader Utility of Punctuated Equilibrium (in More Than a Metaphorical Sense) at Other Levels of Evolution, and for Other Disciplines In and Outside the Natural Sciences   922
General Models for Punctuated Equilibrium   922
Punctuational Change at Other Levels and Scales of Evolution   928
A Preliminary Note on Homology and Analogy in the Conceptual Realm   928
Punctuation Below the Species Level   931
Punctuation Above the Species Level   936
Stasis Analogs: Trending and Non-Trending in the Geological History of Clades   936
Punctuational Analogs in Lineages: The Pace of Morphological Innovation   939
Punctuational Analogs in Faunas and Ecosystems   946
Punctuational Models in Other Disciplines: Towards a General Theory of Change   952
Principles for a Choice of Examples   952
Examples from the History of Human Artifacts and Cultures   952
Examples from Human Institutions and Theories about the Natural World   957

Two Concluding Examples, a General Statement, and a Coda   962

• Appendix: A Largely Sociological (and Fully Partisan) History of the Impact and Critique of Punctuated Equilibrium
The Entrance of Punctuated Equilibrium into Common Language and General Culture   972
An Episodic History of Punctuated Equilibrium   979
Early Stages and Future Contexts   979
Creationist Misappropriation of Punctuated Equilibrium   986
Punctuated Equilibrium in Journalism and Textbooks   990
The Personal Aspect of Professional Reaction   999
The Case Ad Hominem against Punctuated Equilibrium   1000
An Interlude on Sources of Error   1010
The Wages of Jealousy   1014
The Descent to Nastiness   1014
The Most Unkindest Cut of All   1019
The Wisdom of Agassiz's and von Baer's Threefold History of Scientific Ideas   1021
A Coda on the Kindness and Generosity of Most Colleagues   1022

Chapter 10: The Integration of Constraint and Adaptation (Structure and Function) in Ontogeny and Phylogeny: Historical Constraints and the Evolution of Development

• Constraint as a Positive Concept
Two Kinds of Positivity   1025
An Etymological Introduction   1025
The First (Empirical) Positive Meaning of Channeling   1027
The Second (Definitional) Positive Meaning of Causes outside Accepted Mechanisms   1032
Heterochrony and Allometry as the Locus Classicus of the First Positive (Empirical) Meaning. Channeled Directionality by Constraint.   1037
The Two Structural Themes of Internally Set Channels and Ease of Transformation as Potentially Synergistic with Functional Causality by Natural Selection: Increasing Shell Stability in the Gryphaea Heterochronocline   1040
Ontogenetically Channeled Allometric Constraint as a Primary Basis of Expressed Evolutionary Variation: The Full Geographic and Morphological Range of Cerion uva   1045

The Aptive Triangle and the Second Positive Meaning: Constraint as a Theory-Bound Term for Patterns and Directions Not Built Exclusively (Or Sometimes Even at All) by Natural Selection   1051
The Model of the Aptive Triangle   1051
Distinguishing and Sharpening the Two Great Questions   1053
The Structural Vertex   1053
The Historical Vertex   1055
An Epitome for the Theory-Bound Nature of Constraint Terminology   1057

• Deep Homology and Pervasive Parallelism: Historical Constraint as the Primary Gatekeeper and Guardian of Morphospace
A Historical and Conceptual Analysis of the Underappreciated Importance of Parallelism for Evolutionary Theory   1061
A Context for Excitement   1061
A Terminological Excursus on the Meaning of Parallelism   1069
The Nine Fateful Little Words of E. Ray Lankester   1069
The Terminological Origin and Debate about the Meaning and Utility of Parallelism   1076
A Symphony in Four Movements on the Role of Historical Constraint in Evolution: Towards the Harmonious Rebalancing of Form and Function in Evolutionary Theory   1089
Movement One, Statement: Deep Homology across Phyla: Mayr's Functional Certainty and Geoffroy's Structural Vindication   1089
Deep Homology, Archetypal Theories, and Historical Constraint   1089
Mehr Licht (More Light) on Goethe's Angiosperm Archetype   1092
Hoxology and Geoffroy's First Archetypal Theory of Segmental Homology   1095
An Epitome and Capsule History of Hoxology   1095
Vertebrate Homologs in Structure and Action   1101
Segmental Homologies of Arthropods and Vertebrates: Geoffroy's Vindication   1106
Rediscovering the Vertebrate Rhombomeres   1107
More Extensive Homologies throughout the Developing Somites   1109
Some Caveats and Tentative Conclusions   1112
Geoffrey's Second Archetypal Theory of Dorso-Ventral Inversion in the Common Bilaterian Groundplan   1117

Movement Two, Elaboration: Parallelism of Underlying Generators: Deep Homology Builds Positive Channels of Constraint   1122
Parallelism All the Way Down: Shining a Light and Feeding the Walk   1122
Parallelism in the Large: Pax-6 and the Homology of Developmental Pathways in Homoplastic Eyes of Several Phyla   1123
Data and Discovery   1123
Theoretical Issues   1127
A Question of Priority   1130
Parallelism in the Small: The Origin of Crustacean Feeding Organs   1132
Pharaonic Bricks and Corinthian Columns   1134
Movement Three, Scherzo: Does Evolutionary Change Often Proceed by Saltation Down Channels of Historical Constraint?   1142
Movement Four, Recapitulation and Summary: Early Establishment of Rules and the Inhomogenous Population of Morphospace: Dobzhansky's Landscape as Primarily Structural and Historical, Not Functional and Immediate   1147
Bilaterian History as Top-Down by Tinkering of an Initial Set of Rules, Not Bottom-Up by Adding Increments of Complexity   1147
Setting of Historical Constraints in the Cambrian Explosion   1155
Channeling the Subsequent Directions of Bilaterian History from the Inside   1161
An Epilog on Dobzhansky's Landscape and the Dominant Role of Historical Constraint in the Clumped Population of Morphospace   1173

Chapter 11: The Integration of Constraint and Adaptation (Structure and Function) in Ontogeny and Phylogeny: Structural Constraints, Spandrels, and the Centrality of Exaptation in Macroevolution

• The Timeless Physics of Evolved Function
Structuralism's Odd Man Outside   1179
D'Arcy Thompson's Science of Form   1182
The Structure of an Argument   1182
The Tactic and Application of an Argument   1189
The Admitted Limitation and Ultimate Failure of an Argument   1196

Odd Man In (D'Arcy Thompson's Structuralist Critique of Darwinism) and Odd Man Out (His Disparagement of Historicism)   1200
An Epilog to an Argument   1207
Order for Free and Realms of Relevance for Thompsonian Structuralism   1208

• Exapting the Rich and Inevitable Spandrels of History
Nietzsche's Most Important Proposition of Historical Method   1214
Exaptation and the Principle of Quirky Functional Shift: The Restricted Darwinian Version as the Ground of Contingency   1218
How Darwin Resolved Mivart's Challenge of Incipient Stages   1218
The Two Great Historical and Structural Implications of Quirky Functional Shift   1224
How Exaptation Completes and Rationalizes the Terminology of Evolutionary Change by Functional Shifting   1229
Key Criteria and Examples of Exaptation   1234
The Complete Version, Replete with Spandrels: Exaptation and the Terminology of Nonadaptative Origin   1246
The More Radical Category of Exapted Features with Truly Nonadaptive Origins as Structural Constraints   1246
Defining and Defending Spandrels: A Revisit to San Marco   1249
Three Major Reasons for the Centrality of Spandrels, and Therefore of Nonadaptation, in Evolutionary Theory   1258

• The Exaptive Pool: The Proper Conceptual Formula and Ground of  Evolvability
Resolving the Paradox of Evolvability and Defining the Exaptive Pool   1270
The Taxonomy of the Exaptive Pool   1277
Franklins and Miltons, or Inherent Potentials vs. Available Things   1277
Choosing a Fundamentum Divisionis for a Taxonomy: An Apparently Arcane and Linguistic Matter That Actually Embodies a Central Scientific Decision   1280
Cross-Level Effects as Miltonic Spandrels, Not Franklinian Potentials: The Nub of Integration and Radical Importance   1286
A Closing Comment to Resolve the Macroevolutionary Paradox that Constraint Ensures Flexibility Whereas Selection Crafts Restriction   1294

Chapter 12: Tiers of Time and Trials of Extrapolationism, With an Epilog on the Interaction of General Theory and Contingent History   1296

• Failure of Extrapolationism in the Non-Isotropy of Time and Geology
The Specter of Catastrophic Mass Extinction: Darwin to Chicxulub   1296
The Paradox of the First Tier: Towards a General Theory of Tiers of Time   1320

• An Epilog on Theory and History in Creating the Grandeur of This View of Life


Illustration Credits



Defining and Revising
the Structure of Evolutionary Theory

Theories Need Both Essences and Histories

In a famous passage added to later editions of the Origin of Species, Charles Darwin (1872, p. 134) generalized his opening statement on the apparent ab­surdity of evolving a complex eye through a long series of gradual steps by re­minding his readers that they should always treat “obvious” truths with skepticism. In so doing, Darwin also challenged the celebrated definition of science as “organized common sense,” as championed by his dear friend Thomas Henry Huxley. Darwin wrote: “When it was first said that the sun stood still and world turned round, the common sense of mankind declared the doctrine false; but the old saying of Vox populi, vox Dei [the voice of the people is the voice of God], as every philosopher knows, cannot be trusted in science.”

Despite his firm residence within England's higher social classes, Darwin took a fully egalitarian approach towards sources of expertise, knowing full well that the most dependable data on behavior and breeding of domesticated and cultivated organisms would be obtained from active farmers and hus­bandmen, not from lords of their manors or authors of theoretical treatises. As Ghiselin (1969) so cogently stated, Darwin maintained an uncompromis­ingly “aristocratic” set of values towards judgment of his work — that is, he cared not a whit for the outpourings of vox populi, but fretted endlessly and fearfully about the opinions of a very few key people blessed with the rare mix of intelligence, zeal, and attentive practice that we call expertise (a demo­cratic human property, respecting only the requisite mental skills and emo­tional toughness, and bearing no intrinsic correlation to class, profession or any other fortuity of social circumstance).

Darwin ranked Hugh Falconer, the Scottish surgeon, paleontologist, and Indian tea grower, within this most discriminating of all his social groups, a panel that included Hooker, Huxley and Lyell as the most prominent mem­bers. Thus, when Falconer wrote his important 1863 paper on American fos­sil elephants (see Chapter 9, pages 745749, for full discussion of this inci­dent), Darwin flooded himself with anticipatory fear, but then rejoiced in his critic's generally favorable reception of evolution, as embodied in the closing  {2}  sentence of Falconer's key section: “Darwin has, beyond all his cotemporaries [sic], given an impulse to the philosophical investigation of the most back­ward and obscure branch of the Biological Sciences of his day; he has laid the foundations of a great edifice; but he need not be surprised if, in the progress of erection, the superstructure is altered by his successors, like the Duomo of Milan, from the roman to a different style of architecture.”

In a letter to Falconer on October 1, 1862 (in F. Darwin, 1903, volume 1, p. 206), Darwin explicitly addressed this passage in Falconer's text. (Darwin had received an advance copy of the manuscript, along with Falconer's re­quest for review and criticism — hence Darwin's reply, in 1862, to a text not printed until the following year): “To return to your concluding sentence: far from being surprised, I look at it as absolutely certain that very much in the Origin will be proved rubbish; but I expect and hope that the framework will stand.”

The statement that God (or the Devil, in some versions) dwells in the de­tails must rank among the most widely cited intellectual witticisms of our time. As with many clever epigrams that spark the reaction “I wish I'd said that!”, attribution of authorship tends to drift towards appropriate famous sources. (Virtually any nifty evolutionary saying eventually migrates to Т. Н. Huxley, just as vernacular commentary about modern America moves to­wards Mr. Berra.) The apostle of modernism in architecture, Ludwig Mies van der Rohe, may, or may not, have said that “God dwells in the details,” but the plethora of tiny and subtle choices that distinguish the elegance of his great buildings from the utter drabness of superficially similar glass boxes throughout the world surely validates his candidacy for an optimal linkage of word and deed.

Architecture may assert a more concrete claim, but nothing beats the extraordinary subtlety of language as a medium for expressing the importance of apparently trivial details. The architectural metaphors of Milan's cathe­dral, used by both Falconer and Darwin, may strike us as effectively identical at first read. Falconer says that the foundations will persist as Darwin's leg­acy, but that the superstructure will probably be reconstructed in a quite dif­ferent style. Darwin responds by acknowledging Falconer's conjecture that the theory of natural selection will undergo substantial change; indeed, in his characteristically diffident way, Darwin even professes himself “absolutely certain” that much of the Origin's content will be exposed as “rubbish.” But he then states not only a hope, but also an expectation, that the “framework” will stand.

We might easily read this correspondence too casually as a polite dialogue between friends, airing a few unimportant disagreements amidst a commit­ment to mutual support. But I think that this exchange between Falconer and Darwin includes a far more “edgy” quality beneath its diplomacy. Consider the different predictions that flow from the disparate metaphors chosen by each author for the Duomo of Milan — Falconer's “foundation” vs. Darwin's “framework.” After all, a foundation is an invisible system of support, sunk into the ground, and intended as protection against sinking or toppling of the  {3}  overlying public structure. A framework, on the other hand, defines the basic form and outline of the public structure itself. Thus, the two men conjure up very different pictures in their crystal balls. Falconer expects that the underly­ing evolutionary principle of descent with modification will persist as a fac­tual foundation for forthcoming theories devised to explain the genealogical tree of life. Darwin counters that the theory of natural selection will persist as a basic explanation of evolution, — even though many details, and even some subsidiary generalities, cited within the Origin will later be rejected as false, or even illogical.

I stress this distinction, so verbally and disarmingly trivial at a first and superficial skim through Falconer's and Darwin's words, but so incisive and portentous as contrasting predictions about the history of evolutionary the­ory, because my own position — closer to Falconer than to Darwin, but in ac­cord with Darwin on one key point — led me to write this book, while also supplying the organizing principle for the “one long argument” of its entirety. I do believe that the Darwinian framework, and not just the foundation, per­sists in the emerging structure of a more adequate evolutionary theory. But I also hold, with Falconer, that substantial changes, introduced during the last half of the 20th century, have built a structure so expanded beyond the origi­nal Darwinian core, and so enlarged by new principles of macroevolutionary explanation, that the full exposition, while remaining within the domain of Darwinian logic, must be construed as basically different from the canonical theory of natural selection, rather than simply extended.

A closer study of the material basis for Falconer and Darwin's metaphors — the Duomo (or Cathedral) of Milan — might help to clarify this important distinction. As with so many buildings of such size, expense, and centrality (both geographically and spiritually), the construction of the Duomo occupied sev­eral centuries and included an amalgam of radically changing styles and pur­poses. Construction began at the chevet, or eastern end, of the cathedral in the late 14th century. The tall windows of the chevet, with their glorious flamboyant tracery, strike me as the finest achievement of the entire structure, and as the greatest artistic expression of this highly ornamented latest Gothic style. (The term “flamboyant” literally refers to the flame-shaped element so extensively used in the tracery, but the word then came to mean “richly deco­rated” and “showy,” initially as an apt description of the overall style, but then extended to the more general meaning used today.)

Coming now to the main point, construction then slowed considerably, and the main western facade and entrance way (Fig. 1-1) dates from the late 16th century, when stylistic preferences had changed drastically from the points, curves and traceries of Gothic to the orthogonal, low-angled or gently rounded lintels and pediments of classical Baroque preferences. Thus, the first two tiers of the main (western) entrance to the Duomo display a style that, in one sense, could not be more formally discordant with Gothic elements of de­sign, but that somehow became integrated into an interesting coherence. (The third tier of the western facade, built much later, returned to a “retro” Gothic style, thus suggesting a metaphorical reversal of phylogenetic conventions, as


1-1. The west facade (main entrance) of Milan Cathedral, built in baroque style in the 16th century, with a retro-gothic third tier added later.

up leads to older — in style if not in actual time of emplacement!) Finally, in a distinctive and controversial icing upon the entire structure (Fig. 1-2), the “wedding cake,” or row-upon-row of Gothic pinnacles festooning the tops of all walls and arches with their purely ornamental forms, did not crown the edifice until the beginning of the 19th century, when Napoleon conquered the city and ordered their construction to complete the Duomo after so many centuries of work. (These pinnacle forests may amuse or disgust architectural purists, but no one can deny their unintended role in making the Duomo so uniquely and immediately recognizable as the icon of the city.)

How, then, shall we state the most appropriate contrast between the Duomo of Milan and the building of evolutionary theory since Darwin's Ori­gin in 1859? If we grant continuity to the intellectual edifice (as implied by  {5}  comparison with a discrete building that continually grew but did not change its location or basic function), then how shall we conceive “the structure of evolutionary theory” (chosen, in large measure, as the title for this book be­cause I wanted to address, at least in practical terms, this central question in the history and content of science)? Shall we accept Darwin's triumphalist stance and hold that the framework remains basically fixed, with all visually substantial change analogous to the non-structural, and literally superficial, icing of topmost pinnacles? Or shall we embrace Falconer's richer and more critical, but still fully positive, concept of a structure that has changed in radical

1-2. The “wedding cake” pinnacles that festoon the top of Milan Cathedral, and that were not built until the first years of the 19th century after Napoleon conquered the city.


ways by incorporating entirely different styles into crucial parts of the building (even the front entrance!), while still managing to integrate all the differences into a coherent and functional whole, encompassing more and more territory in its continuing enlargement?

Darwin's version remains Gothic, and basically unchanged beyond the vi­sual equivalent of lip service. Falconer's version retains the Gothic base as a positive constraint and director, but then branches out into novel forms that mesh with the base but convert the growing structure into a new entity, largely defined by the outlines of its history. (Note that no one has suggested the third alternative, often the fate of cathedrals — destruction, either total or, partial, followed by a new building of contrary or oppositional form, erected over a different foundation.)

In order to enter such a discourse about “the structure of evolutionary the­ory” at all, we must accept the validity, or at least the intellectual coherence and potential definability, of some key postulates and assumptions that are often not spelled out at all (especially by scientists supposedly engaged in the work), and are, moreover, not always granted this form of intelligibility by philosophers and social critics who do engage such questions explicitly. Most importantly, I must be able to describe a construct like “evolutionary theory” as a genuine “thing” — an entity with discrete boundaries and a definable his­tory — especially if I want to “cash out,” as more than a confusingly poetic image, an analogy to the indubitable bricks and mortar of a cathedral.

In particular, and to formulate the general problem in terms of the specific example needed to justify the existence of this book, can “Darwinism” or “Darwinian theory” be treated as an entity with defining properties of “ana­tomical form” that permit us to specify a beginning and, most crucially for the analysis I wish to pursue, to judge the subsequent history of Darwinism with enough rigor to evaluate successes, failures and, especially, the degree and character of alterations? This book asserts, as its key premise and one long argument, that such an understanding of modern evolutionary theory places the subject in a particularly “happy” intellectual status — with the cen­tral core of Darwinian logic sufficiently intact to maintain continuity as the centerpiece of the entire field, but with enough important changes (to all ma­jor branches extending from this core) to alter the structure of evolutionary theory into something truly different by expansion, addition, and redefini­tion. In short, “The structure of evolutionary theory” combines enough sta­bility for coherence with enough change to keep any keen mind in a perpetual mode of search and challenge.

The distinction between Falconer's and Darwin's predictions, a key ingredi­ent in my analysis, rests upon our ability to define the central features of Darwinism (its autapomorphies, if you will), so that we may then discern whether the extent of alteration in our modern understanding of evolutionary mechanisms and causes remains within the central logic of this Darwinian foundation, or has now changed so profoundly that, by any fair criterion in vernacular understanding of language, or by any more formal account of de­parture from original premises, our current explanatory theory must be  {7}  described as a different kind of mental “thing.” How, in short, can such an in­tellectual entity be defined? And what degree of change can be tolerated or accommodated within the structure of such an entity before we must alter the name and declare the entity invalid or overthrown? Or do such questions just represent a fool's errand from the start, because intellectual positions can't be reified into sufficient equivalents of buildings or organisms to bear the weight of such an inquiry?           

As arrogant as I may be in general, I am not sufficiently doltish or vainglorious to imagine that I can meaningfully address the deep philosophical ques­tions embedded within this general inquiry of our intellectual ages — that is, fruitful modes of analysis for the history of human thought. I shall therefore take refuge in an escape route that has traditionally been granted to scien­tists: the liberty to act as a practical philistine. Instead of suggesting a princi­pled and general solution, I shall ask whether I can specify an operational way to define “Darwinism” (and other intellectual entities) in a manner spe­cific enough to win shared agreement and understanding among readers, but broad enough to avoid the doctrinal quarrels about membership and alle­giance that always seem to arise when we define intellectual commitments as pledges of fealty to lists of dogmata (not to mention initiation rites, secret handshakes and membership cards — in short, the intellectual paraphernalia that led Karl Marx to make his famous comment to a French journalist: “je ne suis pas marxiste”).

As a working proposal, and as so often in this book (and in human affairs in general), a “Goldilocks solution” embodies the blessedly practical kind of approach that permits contentious and self-serving human beings (God love us) to break intellectual bread together in pursuit of common goals rather than personal triumph. (For this reason, I have always preferred, as guides to human action, messy hypothetical imperatives like the Golden Rule, based on negotiation, compromise and general respect, to the Kantian categorical im­peratives of absolute righteousness, in whose name we so often murder and maim until we decide that we had followed the wrong instantiation of the right generality.) We must, in short and in this case, steer between the “too lit­tle” of refusing to grant any kind of “essence,” or hard anatomy of defining concepts, to a theory like Darwinism; and the “too much” of an identification so burdened with a long checklist of exigent criteria that we will either spend all our time debating the status of particular items (and never addressing the heart or central meaning of the theory), or we will waste our efforts, and poi­son our communities, with arguments about credentials and anathemata, ap­plied to individual applicants for membership.

In his brilliant attempt to write a “living” history and philosophy of science about the contemporary restructuring of taxonomic theory by phenetic and cladistic approaches, Hull (1988) presents the most cogent argument I have ever read for “too little” on Goldilocks's continuum, as embodied in his de­fense of theories as “conceptual lineages” (1988, pp. 15-18). I enthusiasti­cally support Hull's decision to treat theories as “things,” or individuals in the crucial sense of coherent historical entities — and in opposition to the  {8}  standard tactic, in conventional scholarship on the “history of ideas,” of tracing the chronology of expression for entirely abstract concepts defined only by formal similarity of content, and not at all by ties of historical continuity, or even of mutual awareness among defenders across centuries and varied cul­tures. (For example, Hull points out that such a conventional history of the “chain of being” would treat this notion as an invariant and disembodied Platonic archetype, independently “borrowed” from the eternal storehouse of potential models for natural reality, and then altered by scholars to fit local contexts across millennia and cultures.)

But I believe that Hull's laudable desire to recast the history of ideas as a narrative of entities in historical continuity, rather than as a disconnected chronology of tidbits admitted into a class only by sufficient formal similarity with an abstract ideological archetype, then leads him to an undervaluation of actual content. Hull exemplifies his basic approach (1988, p. 17): “A con­sistent application of what Mayr has termed 'population thinking' requires that species be treated as lineages, spatiotemporally localized particulars, in­dividuals. Hence, if conceptual change is to be viewed from an evolutionary perspective, concepts must be treated in the same way. In order to count as the 'same concept,' two term-tokens must be part of the same conceptual lin­eage. Population thinking must be applied to thinking itself.”

So far, so good. But Hull now extends this good argument for the necessity of historical connectivity into a claim for sufficiency as well — thus springing a logical trap that leads him to debase, or even to ignore, the “morphology” (or idea content) of these conceptual lineages. He states that he wishes to “organize term-tokens into lineages, not into classes of similar term-types” (pp. 16-17). I can accept the necessity of such historical continuity, but nei­ther I nor most scholars (including practicing scientists) will then follow Hull in his explicit and active rejection of similarity in content as an equally neces­sary criterion for continuing to apply the same name — Darwinian theory, for example — to a conceptual lineage.

At an extreme that generates a reductio ad absurdum for rejecting Hull's conclusion, but that Hull bravely owns as a logical entailment of his own prior decision, a pure criterion of continuity, imbued with no constraint of content, forces one to apply the same name to any conceptual lineage that has remained consciously intact and genealogically unbroken through several generations (of passage from teachers to students, for example), even if the current “morphology” of concepts directly inverts and contradicts the central arguments of the original theory. “A proposition can evolve into its contra­dictory,” Hull allows (1988, p. 18). Thus, on this account, if the living intel­lectual descendants of Darwin, as defined by an unbroken chain of teaching, now believed that each species had been independently created within six days of 24 hours, this theory of biological order would legitimately bear the name of “Darwinism.” And I guess that I may call myself kosher, even though I and all members of my household, by conscious choice and with great ideo­logical fervor, eat cheeseburgers for lunch every day — because we made this  {9}  dietary decision in a macromutational shift of content, but with no genealogi­cal break in continuity, from ten previous generations of strict observers of kashrut.

The objections that most of us would raise to Hull's interesting proposition include both intellectual and moral components. Certain kinds of systems are, and should be, defined purely by genealogy and not at all by content. I am my father's son no matter how we interact. But such genealogical defini­tions, as validated by historical continuity, simply cannot adequately charac­terize a broad range of human groupings properly designated by similarity in content. When Cain mocked God's inquiry about Abel's whereabouts by ex­claiming “Am I my brother's keeper” (Genesis 4:9), he illustrated the appro­priateness of either genealogy by historical connection or fealty by moral re­sponsibility as the proper criterion for “brotherhood” in different kinds of categories. Cain could not deny his genealogical status as brother in one sense, but he derided a conceptual meaning, generally accorded higher moral worth as a consequence of choice rather than necessity of birth, in disclaim­ing any responsibility as keeper. As a sign that we have generally privileged the conceptual meaning, and that Cain's story still haunts us, we need only re­member Claudius's lament that his murder of his own brother (and Hamlet's father) “hath the primal eldest curse upon't.”

Ordinary language, elementary logic, and a general sense of fairness all combine to favor such preeminence for a strong component of conceptual continuity in maintaining a name or label for a theory. Thus, if I wish to call myself a Darwinian in any just or generally accepted sense of such a claim, I do not qualify merely by documenting my residence within an unbroken lin­eage of teachers and students who have transmitted a set of changing ideas organized around a common core, and who have continued to study, aug­ment and improve the theory that bears such a longstanding and honorable label. I must also understand the content of this label myself, and I must agree with a set of basic precepts defining the broad ideas of a view of natural real­ity that I have freely chosen to embrace as my own. In calling myself a Dar­winian I accept these minimal obligations (from which I remain always and entirely free to extract myself should my opinions or judgments change); but I do not become a Darwinian by the mere default of accidental location within a familial or educational lineage.

Thus, if we agree that a purely historical, entirely content-free definition of allegiance to a theory represents “too little” commitment to qualify, and that we must buttress any genealogical criterion with a formal, logical, or anatom­ical definition framed in terms of a theory's intellectual content, then what kind or level of agreement shall we require as a criterion of allegiance for in­clusion? We now must face the opposite side of Goldilocks's dilemma — for once we advocate criteria of content, we do not wish to impose such strin­gency and uniformity that membership becomes more like a sworn obedience to an unchanging religious creed than a freely chosen decision based on per­sonal judgment and perception of intellectual merits. My allegiance to  {10}  Darwinian theory, and my willingness to call myself a Darwinian biologist, must not depend on my subscription to all 95 articles that Martin Luther nailed to the Wittenburg church door in 1517; or to all 80 items in the Syllabus of Er­rors that Pio Nono (Pope Pius IX) proclaimed in 1864, including the “fal­lacy,” so definitionally uncongenial to science, that “the Roman Pontiff can and should reconcile himself to and agree with progress, liberalism and mod­ern civilization”; or to all 39 articles of the Church of England, adopted by Queen Elizabeth in 1571 as a replacement for Archbishop Thomas Cranmer's 42 articles of 1553.

Goldilocks's “just right” position between these extremes will strike nearly all cooperatively minded intellectuals, committed to the operationality and advance of their disciplines, as eminently sensible: shared content, not only historical continuity, must define the structure of a scientific theory; but this shared content should be expressed as a minimal list of the few defining at­tributes of the theory's central logic — in other words, only the absolutely es­sential statements, absent which the theory would either collapse into fallacy or operate so differently that the mechanism would have to be granted an­other name.

Now such a minimal list of such maximal centrality and importance bears a description in ordinary language — but its proper designation requires that evolutionary biologists utter a word rigorously expunged from our profes­sional consciousness since day one of our preparatory course work: the con­cept that dare not speak its name — essence, essence, essence (say the word a few times out loud until the fear evaporates and the laughter recedes). It's high time that we repressed our aversion to this good and honorable word. Theories have essences. (So, by the way, and in a more restrictive and nuanced sense, do organisms — in their limitation and channeling by con­straints of structure and history, expressed as Bauplane of higher taxa. My critique of the second theme of Darwinian central logic, Chapters 45 and 1011, will treat this subject in depth. Moreover, my partial defense of or­ganic essences, expressed as support for structuralist versions of evolutionary causality as potential partners with the more conventional Darwinian functionalism that understandably denies intelligibility to any notion of an es­sence, also underlies the double entendre of this book's title, which honors the intellectual structure of evolutionary theory within Darwinian traditions and their alternatives, and which also urges support for a limited version of structuralist theory, in opposition to certain strict Darwinian verities.)

Our unthinking rejection of essences can be muted, or even reversed into propensity for a sympathetic hearing, when we understand that an invocation of this word need not call down the full apparatus of an entirely abstract and eternal Platonic eidos — a reading of “essence” admittedly outside the logic of evolutionary theory, and historical modes of analysis in general. But the solu­tion to a meaningful notion of essence in biology lies within an important epi­sode in the history of emerging evolutionary views, a subject treated exten­sively in Chapter 4 of this book, with Goethe, Etienne Geoffroy St. Hilaire, and Richard Owen as chief protagonists.  {11} 

After all, the notion of a general anatomical blueprint that contains all particular incarnations by acting as a fundamental building block (Goethe's leaf or Geoffroy's vertebra) moved long ago from conceptualization as a disem­bodied and nonmaterial archetype employed by a creator, to an actual struc­ture (or inherited developmental pathway) present in a flesh and blood ances­tor — a material basis for channeling, often in highly positive Ways, the future history of diversity within particular phyletic lineages. This switch from ar­chetype to ancestor permitted us to reformulate the idea of “essence” as broad and fruitful commonalities that unite a set of particulars into the most meaningful relationships of common causal structure and genesis. Our active use of this good word should not be hampered by a shyness and disquietude lacking any validity beyond the vestiges of suspicions originally set by battles won so long ago that no one can remember the original reasons for anathe­matization. Gracious (and confident) victors should always seek to revive the valid and important aspects of defeated but honorable systems. And the tran­scendental morphologists did understand the importance of designating a small but overarching set of defining architectural properties as legitimate es­sences of systems, both anatomical and conceptual.

Hull correctly defines theories as historical entities, properly subject to all the principles of narrative explanation — and I shall so treat Darwinian logic and its substantial improvements and changes throughout this book. But the­ories of range and power also feature inherent “essences,” implicit in their logical structure, and operationally definable as minimal sets of proposi­tions so crucial to the basic function of a system that their falsification must undermine the entire structure, and also so necessary as an ensemble of mutual implication that all essential components must work in concert to set the theory's mechanism into smooth operation as a generator and explana­tion of nature's order. In staking out this middle Goldilockean ground be­tween (1) the “too little” of Hull's genealogical continuity without commit­ment to a shared content of intellectual morphology and (2) the “too much” of long lists of ideological fealty, superficially imbibed or memorized, and then invoked to define membership in ossified cults rather than thoughtful al­legiance to developing theories, I will argue that a Darwinian essence can be minimally (and properly) defined by three central principles constituting a tri­pod of necessary support, and specifying the fundamental meaning of a pow­erful system that Darwin famously described as the “grandeur in this view of life.”

I shall then show that this formulation of Darwinian minimal commit­ments proves its mettle on the most vital ground of maximal utility. For not only do these three commitments build, in their ensemble, the full frame of a comprehensive evolutionary worldview, but they have also defined the chief objections and alternatives motivating all the most interesting debate within evolutionary theory during its initial codification in the 19th century. More­over, and continuing in our own time, these three themes continue to specify the major weaknesses, the places in need of expansion or shoring up, and the locus of unresolved issues that make evolutionary biology such a central and  {12}  exciting subject within the ever changing and ever expanding world of mod­ern science.

The Structure of Evolutionary Theory: Revising the Three Central Features of Darwinian Logic

In the opening sentence of the Origin's final chapter (1859, p. 459), Darwin famously wrote that “this whole volume is one long argument.” The pres­ent book, on “the structure of evolutionary theory,” despite its extravagant length, is also a brief for an explicit interpretation that may be portrayed as a single extended argument. Although I feel that our best current formulation of evolutionary theory includes modes of reasoning and a set of mechanisms substantially at variance with strict Darwinian natural selection, the logical structure of the Darwinian foundation remains remarkably intact — a fasci­nating historical observation in itself, and a stunning tribute to the intel­lectual power of our profession's founder. Thus, and not only to indulge my personal propensities for historical analysis, I believe that the best way to ex­emplify our modern understanding lies in an extensive analysis of Darwin's basic logical commitments, the reasons for his choices, and the subsequent manner in which these aspects of “the structure of evolutionary theory” have established and motivated all our major debates and substantial changes since Darwin's original publication in 1859.1 regard such analysis not as an antiquarian indulgence, but as an optimal path to proper understanding of our current commitments, and the underlying reasons for our decisions about them.

As a primary theme for this one long argument, I claim that an “essence” of Darwinian logic can be defined by the practical strategy defended in the first section of this chapter: by specifying a set of minimal commitments, or broad statements so essential to the central logic of the enterprise that disproof of any item will effectively destroy the theory, whereas a substantial change to any item will convert the theory into something still recognizable as within the Bauplan of descent from its forebear, but as something sufficiently differ­ent to identify, if I may use the obvious taxonomic metaphor, as a new subclade within the monophyletic group. Using this premise, the long argu­ment of this book then proceeds according to three sequential claims that set the structure and order of my subsequent chapters:

1. Darwin himself formulated his central argument under these three basic premises. He understood their necessity within his system, and the difficulty that he would experience in convincing his contemporaries about such unfa­miliar and radical notions. He therefore presented careful and explicit de­fenses of all three propositions in the Origin. I devote the first substantive chapter (number 2) to an exegesis of the Origin of Species as an embodiment of Darwin's defense for this central logic.

2. As evolutionary theory experienced its growing pains and pursued its founding arguments in the late 19th and early 20th centuries (and also in  {13}  its pre-Darwinian struggles with more inchoate formulations before 1859), these three principles of central logic defined the themes of deepest and most persistent debate — as, in a sense, they must because they constitute the most interesting intellectual questions that any theory for causes of descent with modification must address. The historical chapters of this book's first half then treat the history of evolutionary theory as responses to the three central issues of Darwinian logic (Chapters 37).

3. As the strict Darwinism of the Modern Synthesis prevailed and “hard­ened,” culminating in the overconfidences of the centennial celebrations of 1959, a new wave of discoveries and theoretical reformulations began to challenge aspects of the three central principles anew — thus leading to an­other fascinating round of development in basic evolutionary theory, extend­ing throughout the last three decades of the 20th century and continuing to­day. But this second round has been pursued in an entirely different and more fruitful manner than the 19th century debates. The earlier questioning of Darwin's three central principles tried to disprove natural selection by offer­ing alternative theories based on confutations of the three items of central logic. The modern versions accept the validity of the central logic as a foun­dation, and introduce their critiques as helpful auxiliaries or additions that enrich, or substantially alter, the original Darwinian formulation, but that leave the kernel of natural selection intact. Thus, the modern reformula­tions are helpful rather than destructive. For this reason, I regard our modern understanding of evolutionary theory as closer to Falconer's metaphor, than to Darwin's, for the Duomo of Milan — a structure with a firm foundation and a fascinatingly different superstructure. (Chapters 812, the second half of this book on modern developments in evolutionary theory, treat this third theme.)

Thus, one might say, this book cycles through the three central themes of Darwinian logic at three scales — by brief mention of a framework in this chapter, by full exegesis of Darwin's presentation in Chapter 2, and by lengthy analysis of the major differences and effects in historical (Part 1) and modern critiques (Part 2) of these three themes in the rest of the volume.

The basic formulation, or bare-bones mechanics, of natural selection is a disarmingly simple argument, based on three undeniable facts (overproduc­tion of offspring, variation, and heritability)1 and one syllogistic inference (natural selection, or the claim that organisms enjoying differential reproduc­tive success will, on average, be those variants that are fortuitously better adapted to changing local environments, and that these variants will then pass their favored traits to offspring by inheritance). As Huxley famously, and ruefully, remarked (in self-reproach for failing to devise the theory him­self), this argument must be deemed elementary (and had often been formulated  {14}  before, but in negative contexts, and with no appreciation of its power — see p. 137), and can only specify the guts of the operating machine, not the three principles that established the range and power of Darwin's rev­olution in human thought. Rather, these three larger principles, in defining the Darwinian essence, take the guts of the machine, and declare its simple operation sufficient to generate the entire history of life in a philosophical manner that could not have been more contrary to all previous, and cher­ished, assumptions of Western life and science.

The three principles that elevated natural selection from the guts of a working machine to a radical explanation of the mechanism of life's history can best be exemplified under the general categories of agency, efficacy, and scope. I treat them in this specific order because the logic of Darwin's own de­velopment so proceeds (as I shall illustrate in Chapter 2), for the most radical claim comes first, with assertions of complete power and full range of appli­cability then following.

Agency. The abstract mechanism requires a locus of action in a hierar­chical world, and Darwin insisted that the apparently intentional “benevo­lence” of nature (as embodied in the good design of organisms and the har­mony of ecosystems) flowed entirely as side-consequences of this single causal locus, the most “reductionistic” account available to the biology of Darwin's time. Darwin insisted upon a virtually exceptionless, single-level theory, with organisms acting as the locus of selection, and all “higher” order emerging, by the analog of Adam Smith's invisible hand, from the (unconscious) “strug­gles” of organisms for their own personal advantages as expressed in dif­ferential reproductive success. One can hardly imagine a more radical refor­mulation of a domain that had unhesitatingly been viewed as the primary manifestation for action of higher power in nature — and Darwin's brave and single-minded insistence on the exclusivity of the organismic level, although rarely appreciated by his contemporaries, ranks as the most radical and most distinctive feature of his theory.

Efficacy. Any reasonably honest and intelligent biologist could easily understand that Darwin had identified a vera causa (or true cause) in natural selection. Thus, the debate in his time (and, to some extent, in ours as well) never centered upon the existence of natural selection as a genuine causal force in nature. Virtually all anti-Darwinian biologists accepted the reality and action of natural selection, but branded Darwin's force as a minor and negative mechanism, capable only of the headsman's or executioner's role of removing the unfit, once the fit had arisen by some other route, as yet uniden­tified. This other route, they believed, would provide the centerpiece of a “real” evolutionary theory, capable of explaining the origin of novelties. Dar­win insisted that his admittedly weak and negative force of natural selection could, nonetheless, under certain assumptions (later proved valid) about the nature of variation, act as the positive mechanism of evolutionary novelty — that is, could “create the fit” as well as eliminate the unfit — by slowly accu­mulating the positive effects of favorable variations through innumerable generations.  {15} 

Scope. Even the most favorably minded of contemporaries often admit­ted that Darwin had developed a theory capable of building up small changes (of an admittedly and locally “positive” nature as adaptations to changing environments) within a “basic type” — the equivalent, for example, of making dogs from wolves or developing edible corn from teosinte. But these critics could not grasp how such a genuine microevolutionary process could be ex­tended to produce the full panoply of taxonomic diversity and apparent “progress” in complexification of morphology through geological time. Dar­win insisted on full sufficiency in extrapolation, arguing that his micro-evolutionary mechanism, extended through the immensity of geological time, would be fully capable of generating the entire pageant of life's history, both in anatomical complexity and taxonomic diversity — and that no further causal principles would be required.

Because primates are visual animals, complex arguments are best portrayed or epitomized in pictorial form. The search for an optimal icon to play such a role is therefore no trivial matter (although scholars rarely grant this issue the serious attention so richly merited) — especially since the dangers of confu­sion, misplaced metaphor, and replacement of rigor with misleading “intu­ition” stand so high. I knew from the beginning of this work that I needed a suitable image for conveying the central logic of Darwinian theory. As one of my humanistic conceits, I hoped to find a historically important scientific im­age, drawn for a different reason, that might fortuitously capture the argu­ment in pictorial form. But I had no expectation of success, and assumed that I would need to commission an expressly designed figure drawn to a long list of specifications.

The specific form of the image — its central metaphorical content, if you will — plays an important role in channeling or misdirecting our thoughts, and therefore also requires careful consideration. In the text of this book, I speak most often of a “tripod” since central Darwinian logic embodies three major propositions that I have always visualized as supports — perhaps be­cause I have never been utterly confident about this entire project, and I needed some pictorial encouragement to keep me going for twenty years. (And I much prefer tripods, which can hold up elegant objects, to buttresses, which may fly as they preserve great Gothic buildings, but which more often shore up crumbling edifices. Moreover, the image of a tripod suits my major claim particularly well — for I have argued, just above, that we should define the “essence” of a theory by an absolutely minimal set of truly necessary propositions. No structure, either of human building or of abstract form, captures this principle better than a tripod, based on its absolute minimum of three points for fully stable support in the dimensional world of our physical experience.)

But organic images have always appealed more strongly, and I preferred a biological icon. If the minimal logic can be represented by a tripod pointing downward, then the same topology can be inverted into a structure growing upward. Darwin's own favorite image of the tree of life immediately sug­gested itself, and I long assumed that I would eventually settle on a botanical  {16}  icon. But I also remembered Darwin's first choice for an organic metaphor or picture of branching to capture his developing views about descent with modification and the causes of life's diversity — the “coral of life” of his “B Note­book” on transmutation, kept during the 1830's as he became an evolutionist and struggled towards the theory of natural selection (see Barrett et al., 1987).

As I began to write this summary chapter, I therefore aimlessly searched through images of Cnidaria from my collection of antiquarian books in paleontology. I claim no general significance whatsoever for my good fortune, but after a lifetime of failure in similar quirky quests, I was simply stunned to find a preexisting image — not altered one iota from its original form, I promise you, to suit my metaphorical purposes — that so stunningly embodied my needs, not only for a general form (an easy task), but down to the smallest de­tails of placement and potential excision of branches (the feature that I had no right or expectation to discover and then to exapt from so different an original intent).

The following figure comes from the 1747 Latin version of one of the semi­nal works in the history of paleontology — the 1670 Italian treatise of the Sicilian savant and painter Agostino Scilla, ha vana speculazione disingan-nata dal senso (“Vain speculation undeceived by the senses” — Scilla's de­fense, at the outset of “the scientific revolution” of Newton's generation, for empirical methods in the study of nature, and specifically, in this treatise, for a scientific paleontology and the need to recognize fossils as remains of an­cient organisms, not as independent products of the mineral kingdom). This work, famous not only for an incisive text, but also for its beautiful plates (see Fig. 1-3), engraved by an author known primarily as an artist of substan­tial eminence, includes this figure, labeled Coralium articulatum quod copio-sissimum in rupibus et collibus Messanae reperitur (“Articulated coral, found in great abundance in the cliffs and hills of Messina”).

This model, and its organic features, work uncommonly well as a meta­phor for the Goldilockean position of definition by a barest minimum of truly fundamental postulates. For Scilla's coral, with its branching structure (see Fig. 1-4) — particularly as expressed in the lessening consequences of excising branches at ever higher levels nearer the top (the analogs of disconfirming theoretical features of ever more specialized and less fundamental import) — so beautifully captures the nature and operation of the intellectual structure that I defended above for specifying the essences of theories. The uncanny ap­propriateness of Scilla's coral lies in the fortuity that this particular specimen (accurately drawn from nature by Scilla, I assume, and not altered to as­sert any general point) just happens to include exactly the same number of branches (three) as my Darwinian essential structure. (They terminate at the same upper level, so I could even turn the specimen over into a tolerably unwobbly tripod!) Moreover, since this particular genus of corals grows in discrete segments, the joining points correspond ideally with my metaphor of chopping planes for excising parts of structures at various levels of impor­tance in an intellectual entity. But, most incredibly, the segmental junctions of


1-3. The famous frontispiece from Scilla's treatise of 1670 defending the organic nature of fossils. The solid young man, representing the truth of sensory experi­ence, shows a fossil sea urchin in his right hand to a wraithlike figure represent­ing the former style of speculative thinking. With his left hand, the solid figure points to other fossils found in Sicily. The text proclaims: “Vain speculation un­deceived by the senses.”


this particular specimen just happen to occupy the exact places that I needed a priori to make my central point about lower choppings that destroy theo­ries, middle choppings that change theories in a Falconerian way (major al­terations in structure upon a preserved foundation), and upper choppings that change theories in the lesser manner of Darwin's Milanese metaphor (smaller excisions that leave the framework intact as well).

The central trunk (the theory of natural selection) cannot be severed, or the creature (the theory) dies. (The roots, if you will, represent sources of evi­dence; any one may be excised, if recognized as incorrect by later study, so long as enough remain to anchor the structure). This central trunk then di­vides into a limited number of major branches. These basic struts — the three

1-4. Agostino Scilla was also a celebrated painter as well as a scien­tist. The plates of his 1670 trea­tise are therefore particularly well done. This figure, representing a fossil coral that Scilla found near Messina, fortuitously (and without any alteration whatsoever), pre­sents a detailed picture of the basic logic of Darwinian theory as recog­nized in this book. See text for details.


branches of the Darwinian essence in this particular picture — are also so es­sential that any severing of a complete branch either kills, or so seriously compromises, the entire theory that a new name and basic structure becomes essential.

We now reach the interesting point where excisions and regraftings pre­serve the essential nature of an intellectual structure, but with two quite dif­ferent levels of change and revision, as characterized by Falconer's and Dar­win's competing metaphors for the Duomo of Milan. I would argue that a severing low on any one of the three major branches corresponds with a revi­sion profound enough to validate the more interesting Falconerian version of major revision upon a conserved foundation. (The Falconerian model is, in this sense, a Goldilockean solution itself, between the “too much” of full de­struction and the “too little” of minor cosmetic revision.) On the other hand, the severing of a subbranch of one of the three branches symbolizes a less portentous change, closer to Darwinian models for the Milanese Duomo — an alteration of important visual elements, but without change in the basic framework.

My fascination with the current state of evolutionary theory, at least as I read current developments in both logic and empirics, lies in its close con­formity to the Falconerian model — with enough continuity to make the past history of the field so informative (and so persistently, even emotionally, compelling), but with enough deep difference and intellectual fascination to stim­ulate anyone with a thirst for the intriguing mode of novelty that jars previ­ous certainty, but does not throw a field into the total anarchy of complete rebuilding (not a bad thing either, but far from the actual circumstance in this case).

To summarize my views on the utility of such a model for the essence of Darwinian logic, I will designate three levels of potential cuts or excisions to this organic (and logical) coral of the structure of evolutionary theory, as originally formulated by Darwin in the Origin of Species, and as revised in a Falconerian way in recent decades. The most inclusive and most fundamental K-cuts (killing cuts) sever at least one of the three central principles of Dar­winian logic and thereby destroy the theory tout court. The second level of R-cuts (revision cuts) removes enough of the original form on one of the three central branches to ensure that the new (and stronger or more arborescent) branch, in regrowing from the cut, will build a theory with an intact Darwin­ian foundation, but with a general form sufficiently expanded, revised or re­constructed to present an interestingly different structure of general explana­tion — the Falconerian model for the Duomo of Milan. Finally, the third level of S-cuts (subsidiary cuts) affects only a subbranch of one of the three major branches, and therefore reformulates the general theory in interesting ways, while leaving the basic structure of explanation intact — the Darwinian model for the Duomo of Milan.

I wrote this book because I believe that all three pillars, branches, or tri­pod legs, representing the three fundamental principles of Darwinian central logic, have been subjected to fascinating R-cuts that have given us at least the  {20}  firm outlines — for the revised structure of evolutionary explanation remains a work vigorously in progress, as only befits the nature of its subject, after all! — of a far richer and fascinatingly different theory with a retained Dar­winian core rooted in the principles of natural selection. In short, we live in the midst of a Falconerian remodeling of our growing and multiform, yet co­herently grounded, intellectual mansion.

I will not, in this chapter, detail the nature of the K-cuts that failed (thus preserving the central logic of Darwinism), the R-cuts that have succeeded in changing the structure of evolutionary theory in such interesting ways, and the S-cuts that have refurbished major rooms in particular wings of the edi­fice — for these specifications set the subject matter of all following chapters. But to provide a better opening sense of this book's argument — and to clarify the nature of the three central claims of Darwinian logic — I shall at least dis­tinguish, for each branch, the K-cuts that never prevailed (and therefore did not fell the structure) from the R-cuts that have affected each branch, and have therefore provoked our current process of building an enriched struc­ture for evolutionary theory.

Returning to Scilla's coral (Fig. 1-4), consider the central branch as the first leg of the tripod (agency, or the claim for organismal selection as the causal locus of the basic mechanism), the left branch as the second leg (efficacy, or the claim that selection acts as the primary creative force in building evo­lutionary novelties), and the right branch as the third leg (scope, or the claim that these microevolutionary modes and processes can, by extrapola­tion through the vastness of geological time, explain the full panoply of life's changes in form and diversity).

The cut labeled K1 on Figure 1-4 would have severed the entire coral by disproving natural selection as an evolutionary force at all. The cut labeled K2 would have fully severed the second branch, leaving natural selection as a legitimate cause, but denying it any creative role, and thereby dethroning Darwinism as a major principle in explaining life's history. (We shall see, in Chapters 36, that such a denial of creativity underlay the most common anti-Darwinian argument in the first generations of debate.) The cut labeled К3 would have fully severed the third branch, allowing that natural selection might craft some minor changes legitimately called “creative” in a local sense, but denying that Darwin's mechanism could then be extended to explain the panoply of macroevolutionary processes, or the actual pageant of life's his­tory. The success of any one of these K-cuts would have destroyed Darwinian theory, plain and simple. None of them succeeded, and the foundation of Darwinian central logic remains intact and strong.

In striking, and most positive, contrast, I believe that higher R-cuts — leav­ing the base of each major branch intact, but requiring a substantial regrowth and regrafting of an enlarged structure upon the retained foundation — have been successfully wielded against all three branches of Darwinian logic, as the structure of evolutionary theory developed in the last third of the 20th cen­tury (following too rigid a calcification of the original structure, a good ad­umbration of the coral metaphor!, in the hardening of the Modern Synthesis  {21}  that culminated in the Darwinian centennial celebrations of 1959). On the first branch of agency, the cut labeled R1 (see Fig. 1-4) expanded Darwin's unilevel theory of organismal selection into a hierarchical model of selection acting simultaneously on several legitimate levels of Darwinian individuality (genes, cell-lineages, organisms, demes, species, and clades). I shall show in Chapters 3, 8, and 9 how the logic of this pronounced expansion builds a the­ory fascinatingly different from, and not just a smooth extension of, Darwin's single level mechanism of agency — my reason for portraying the hierarchical model as a deeply interesting R-cut rather than a more superficial S-cut.

On the second branch of efficacy, the cut labeled R2 accepts the validity of Darwin's argument for creativity (by leaving the base of the branch intact), but introduces a sufficient weight of formalist thinking — via renewed appre­ciation for the enormous importance of structural, historical, and develop­mental constraint in channeling the pathways of evolution, often in highly positive ways — that the pure functionalism of a strictly Darwinian (and ex­ternalist) approach to adaptation no longer suffices to explain the channeling of phyletic directions, and the clumping and inhomogeneous population of organic morphospace. The strict Darwinian form of explanation has thereby been greatly changed and enriched, but in no way defeated. I shall discuss the historical aspect of this branch in Chapters 4 and 5, and modern reformula­tions of this R2 cut in Chapters 10 and 11.

On the final branch of scope, the cut labeled R3 accepts the Darwinian contention that microevolutionary modes and principles can build grand pat­terns by cumulation through geological immensity, but rejects the argument that such extrapolations can render the entire panoply of phenomena in life's history without adding explicitly macroevolutionary modes for distinctive expression of these processes at higher tiers of time — as in the explanation of cladal trends by species sorting under punctuated equilibrium, rather than by extended adaptive anagenesis of purely organismal selection, and in the ne­cessity of titrating adaptive microevolutionary accumulation with occasional resetting of rules and patterns by catastrophically triggered mass extinctions at time's highest tier. Chapters 6 and 12 discuss historical and modern cri­tiques of Darwinian extrapolationism.

For now, I will say little about the even higher and more superficial S-cuts of subbranches, but I will at least indicate how I construe this category by stating a hypothetical example for each branch: an S1 cut, for example, might accept the selective basis of evolutionary change in a purely mechanical sense, but then deny full force to Darwin's deliciously radical philosophical claim that all apparent “higher level” harmony arises consequentially, through the invisible hand of lower levels acting for personal reproductive success. One might, in principle, propose such a revision by arguing that a higher force, operating by an overarching principle of order, “employs” natural selection as its mechanical agent. (I speak only hypothetically here, for no such defend­able scientific hypothesis now exists, although the concept certainly remains intelligible. Explicitly theological versions don't count as science, whatever their kind or form of potential validity.) An S2 cut might be assayed by a  {22}  developmental saltationist who accepted the selectionist basis of adaptive change but felt that, at a sufficient relative frequency to be counted as impor­tant, the initial steps of such changes may be larger than the pure continuationism of Darwinian selection can admit. And an S3 cut might accept the full validity of microevolutionary extrapolationism, but deny the subsidiary de­fense of progress that Darwin grafted onto this apparatus (see Chapter 6) with ecological arguments about plenitude and the priority of biotic over abiotic competition.

As a paleontologist and part-time historian of science by profession, my reading of these important R-cuts arose from a macroevolutionary perspec­tive framed largely in terms of longstanding difficulties faced by Darwinism in extending its successes for explaining small changes in palpable time into equally adequate causal accounts for broader patterns and processes in geo­logical history. I have, in this effort, also benefited from my personal study of Darwin's life and times, and especially the late 19th century debates on mech­anisms of evolution (as promulgated largely by professionals who could nei­ther fully understand nor accept the radical philosophical commitments un­derlying Darwin's view). This historical study allowed me to grasp the continuity in basic themes from Darwin's own formulation, through these foundational debates, right down to the major theoretical struggles of our own time. An appreciation of this continuity allowed me to discern and de­fine the distinctively Darwinian view of life.

But I recognize only too well that every strength comes paired with weaknesses. In my case, a paleontological focus leads me into relative ignorance for an equally important locus of reform in the structure of Darwinism — in­creasing knowledge of the nature of genomes and the mechanics of develop­ment. (I try to cover the outlines of important theoretical critiques from this “opposite” realm of the smallest, but the relative weightings in my text reflect my own varying competencies far more than the merits of the cases. For ex­ample, although I do discuss, and perhaps even adequately outline, the im­portance of Kimura and King's neutralist theory in questioning previous as­sumptions of adaptationist hegemony, I surely do not give an appropriate volume of attention to this enormously important subject.)

Nonetheless, I hope that I have managed to present an adequate account of the coordinating themes that grant such interest and coherence to modern reformulations of the structure of evolutionary theory. Such thematic consis­tency in revision becomes possible largely because Darwin himself, in his characteristically brilliant way, tied the diverse threads of his initiating argu­ment into an overall view with a similarly tight structure — thus granting clear definition to his own commitments, and also permitting their revision in the form of an equally coherent “package.” I would argue, moreover, and with­out wishing to become extravagantly hagiographical (for I wrote this book, after all, primarily to discuss a critique and revision of strict Darwinism), that our modern sense of limitations in the canonical version arises from decisions that Darwin made for tough and correct reasons in the context of his initiat­ing times — reasons that made his account the first operational theory of  {23}  evolution in modern science. In particular, as Chapter 2 will discuss in detail, Darwin converted evolution from untestable speculation to doable science by breaking through the old paradox (as embedded most prominently in Lamarck's system) of contrasting a palpable force of small-scale change that could do little in extension, with a basically nonoperational (and orthogonal) mechanism of large-scale change putatively responsible for all the interesting patterns of life's history, but imperceptible and untestable from the uniformitarian study of modern organisms.

By claiming that the small-scale mechanics of modern change could, by extension, explain all of evolution, Darwin opened the entire field to empirical study. And yet, as Hegel and so many other students of change have noted, progress in human (and other) affairs tends to spiral upwards in cycles of pro­posal (thesis), then countered by opposition (antithesis), and finally leading to a new formulation combining the best aspects of both competitors (synthe­sis). Darwin's thesis established evolution as a science, but his essential com­mitments, as expressed in the three legs of his necessary logical tripod (or the three branches of his conceptual tree or coral, as in the alternate metaphor of Fig. 1-4), eventually proved too narrow and confining, thus requiring an an­tithesis of extension and reformulation on each branch, and leading — or so this book maintains as a central thesis of its own — to a still newer and richer synthesis expressing our best current understanding of the structure of evolu­tionary theory.

In fact, and to repeat my summary in this different form, one might encapsulate the long argument of this book in such a Hegelian format. Pre-Darwinian concepts of evolution remained speculative and essentially nonoperational, largely because (see Chapter 3) they fell into the disabling para­dox of contrasting an effectively unknowable large-scale force of cosmic progress against an orthogonal, palpable and testable small-scale force that could generate local adaptation and diversity, but that couldn't, in principle, explain the macroevolutionary pattern of life. Then Darwin, in his thesis (also an antithesis to these earlier sterile constructions), brilliantly argued that the putative large-scale force did not exist, and that all evolution could be explained by upward extrapolation from the small-scale force, now prop­erly understood as natural selection. In a first stage of debate during the late 19th and early 20th centuries (Chapters 36), most critiques of Darwinism — one might designate them as a first round of ultimately destructive antithe­ses — simply denied sufficient agency, efficacy and range to natural selection, and reasserted the old claim of duality, with selection relegated to triviality, and some truly contrary force sought as the explanation for major features of evolution. Strict Darwinism eventually fended off these destructive critiques, reasserted itself in the triumphant, and initially (and generously) pluralistic form of the Modern Synthesis, but eventually calcified into a “hardened” ver­sion (Chapter 7).

Then, in a strikingly different, and ultimately fruitful, second round of antitheses, a renewed debate about central theoretical issues arose during the last three decades of the 20th century, and reshaped the field by recognizing  {24}  that selection needed to be amplified, reformulated and invigorated by other, noncontrary (and, at most, orthogonal) causes, not rejected as wrong, or scorned as trivial (Chapters 812). The one long argument of this book holds that a synthesis (still much in progress) has now sufficiently coagulated from this debate to designate our best current understanding of the structure of evolutionary theory as something rich and new, with a firmly retained basis in Darwinian logic — in other words, and following the organizing and opening metaphor of this chapter, as a validation of Falconer's, rather than Darwin's, concept of the historical growth and change of Milan's cathedral.

Ariel's telling verse in Shakespeare's The Tempest proclaims in dense metaphor:

Full fathom five thy father lies;

Of his bones are coral made;

Those are pearls that were his eyes:

Nothing of him that doth fade

But doth suffer a sea-change

Into something rich and strange.

With the exception of one possible (and originally unintended) modern reading of these images, this famous and haunting verse provides a beautiful description of both the priceless worth and intriguing modern transformation of Darwin's original theory. (For the exception, several connotations of deep burial in the sea — full fathom five — might be viewed negatively, as in “deep sixing” or going to Davy Jones's locker. But, for natural historians who read this book, and coming from an invertebrate paleontologist as author, the seafloor could not represent a more positive resting place or point of origin — and I intend to evoke only these upbeat images in citing Ariel's lines.) Other­wise, Darwin's original structure has only yielded greater treasure in cascad­ing implications and developments through the subsequent history of evolu­tionary thought — the conversion of the bones of an original outline into precious coral and pearls of current substance. Nothing of Darwin's central logic has faded or fully capsized, but his theory has been transformed, along his original lines, into something far different, far richer, and far more ade­quate to guide our understanding of nature.

The last three lines of Shakespeare's verse also appear on the tombstone of the great poet Percy Bysshe Shelley (also the author of the preface to his wife's novella, Frankenstein, which cites Erasmus Darwin in its first line of text). I believe that these words would suit, and honor, Charles Darwin just as well and just as rightly.

Apologia Pro Vita Sua


The Preacher spoke ever so truly in writing his famous words (Ecclesiastes 3:1-7): “For every thing there is a season, and a time to every purpose ... A  {25}  time to break down, and a time to build up ... A time to rend, and a time to sew: a time to keep silence, and a time to speak.” Evolutionary theory now stands in the happier second state of these genuine dichotomies (in part be­cause the first state had been mined to the limited extent of its utility): we live in a time for building up, for sewing together, and for speaking out.

Not all times are such good times, and not all scientists win the good for­tune to live within these times of motion. For theories grow as organisms do, with periods of Sturm und Drang, long latencies of youth or ossifications of age, and some happy times of optimally productive motion in between (an­other Goldilockean phenomenon). I recently studied the life and career of E. Ray Lankester (Gould, 1999a), clearly the most talented evolutionary morphologist of the generation just after Darwin. He did “good” work and had a “good” career (see Chapter 10, pages 10691076 for his best theoretical efforts), but he never transcended the ordinary. Perhaps the limitation lay largely within his own abilities. However, I rather suspect that he did possess both the temperamental gumption and the requisite intellectual might — but that the tools of major empirical advance just didn't emerge in his generation, for he remained stuck in a relatively unproductive middle, as Darwin had seized the first-fruits from traditional data of natural history, and the second plucking required a resolution of genetic mechanisms.

I felt a similar kind of frustration in 1977, after writing my first technical book, Ontogeny and Phylogeny (see Chapter 10, pages 10611063). I had spent the best years of a young career on a subject that I knew to be rele­vant (at a time when most of the profession had forgotten). But then defeat snatched my prize from the jaws of victory. I am proud of the book, and I do believe that it helped to focus interest on a subject that became doable soon thereafter. But I ran up against a wall right at the end — for the genetics of de­velopment clearly held the key to any rapprochement of embryology and evo­lution, and we knew effectively nothing about eukaryotic regulation. Indeed, as we could then only characterize structural genes by electrophoretic tech­niques, our major “arguments” for regulatory effects (if they even merited such a positive designation) invoked such negative evidence as the virtual identity in structural genes between chimps and humans, coupled with a fair visceral sense of extensive phenotypic disparity in anatomy and behavior — with the differences then attributed to regulatory genes that we could not, at the time, either study or even identify.

By sheer good fortune (abetted in minuscule ways by my own pushes and those of my paleontological colleagues), the field moved fast and I lived long enough to witness a sea change (if I may cite Ariel yet again) towards potentiation on all three major intellectual and social substrates for con­verting a subject from great promise combined with even more frustrating inoperability, into a discipline bursting with new (and often utterly surpris­ing) data that led directly to testable hypotheses about basic issues in the structure of evolutionary theory.

Empirics. During the last third of the 20th century, new techniques and conceptualizations opened up important sources of data that challenged  {26}  orthodox formulations for all three branches of essential Darwinian logic. To cite just one relevant example for each branch, theoretical development and accumulating data on punctuated equilibrium allowed us to reconceptualize species as genuine Darwinian individuals, fully competent to participate in processes of selection at their own supraorganismic (and suprademal) level — and then to rethink macroevolution as the differential success of species rather than the extended anagenesis of organismal adaptation (see Chapter 9). This validation of the species-individual aided the transformation of what had begun as a particular argument about group (or interdemic) selection into a fully generalized hierarchical theory, with good cases then documented from the genic to the cladal level (see Chapter 8).

On the second branch of full efficacy for natural selection as an externalist and functionalist process, the stunning discoveries of extensive deep homologies across phyla separated by more than 500 million years (particularly the vertebrate homologs of arthropod Hox genes) — against explicit statements by architects of the Modern Synthesis (see p. 539) that such homologies could not exist in principle, in a world dominated by their conception of nat­ural selection — forced a rebalancing or leavening of Darwinian functionalism with previously neglected, or even vilified, formalist perspectives based on the role of historical and structural constraints in channeling directions of evolu­tionary change, and causing the great dumpings and inhomogeneities of morphospace — phenomena that had previously been attributed almost exclu­sively to functionalist forces of natural selection.

On the third branch of extrapolation, the discovery and relatively quick validation, beginning in 1980, of a truly catastrophic trigger for at least one great mass extinction (the K-T event of 65 million years ago), fractured the uniformitarian consensus, embraced by a century of paleontological compla­cency, that all apparent faunal overturns could be “spread out” into sufficient time for explanation by ordinary causes under plausible intensifications that would not alter conventional modes of evolution and extinction.

Moreover, as we shall see, these three apparently rather different kinds of data and their attendant critiques cohere into a revised general structure for evolutionary theory — thus marking our age as a time for building up and not only as a time for breaking down.

Concepts. Following the Kantian dictum that percepts without con­cepts are blind, but concepts without percepts empty, these two categories interpenetrate as “pure” data suggest novel ideas (how can one not rethink the causes of mass extinction when evidence surfaces for a bolide, 7-10 km in diameter, and packing 104 the megatonnage of all the earth's nuclear weapons combined), whereas “abstract” concepts then taxonomize the natural world in different ways, often “creating” data that had never been granted enough previous intellectual space even to be conceived (as when punctuated equilib­rium made stasis a theoretically meaningful and interesting phenomenon, and not just an embarrassing failure to detect “evolution,” in its traditional definition of gradual change — and paleontologists then began active stud­ies of a subject that had previously been ignored as uninteresting, if conceptualized  {27}  at all). But, speaking parochially as a student of the fossil record, I can at least say that the conceptual revolution in macroevolutionary thinking re­vitalized the field of paleobiology (even creating the name as a subdiscipline of paleontological endeavor). Whatever the varied value of different individ­ual efforts in this burgeoning field, we may at least be confident that our pro­fession will no longer be humiliated as a synecdoche for ossified boredom among the natural sciences — as Nature did in 1969 when editorializing about the salutary value of plate tectonics in revitalizing the geological sciences: “Scientists in general might be excused for assuming that most geologists are paleontologists and most paleontologists have staked out a square mile as their life's work. A revamping of the geologist's image is badly needed” (Anonymous, 1969, p. 903).

The intricate and multifaceted concepts that have nuanced and altered the central logic on all three branches of Darwinism's essential postulates repre­sent ideas of broad ramification and often remarkably subtle complexity, as we vain scientists soon discovered in our fractured bubbles of burst pride — for we had been so accustomed to imagining that an evening in an armchair could conquer any merely conceptual issue, whereas we all acknowledge the substantial time and struggle that empirical problems, demanding collection and evaluation of data, often require. Yet, on these basic questions in formu­lating evolutionary theory, we often read and thought for months, and ended up more confused than when we began.

The general solution to such procedural dilemmas lies in a social and intellectual activity that scientists do tend to understand and practice better than colleagues in most other academic disciplines — collaboration. Far more than most colleagues, I have tended to work alone in my professional life and pub­lication. But for each of the conceptually difficult and intellectually manifold issues of reevaluation for the central logic of the three essential Darwinian postulates, I desperately needed advice, different skills, and the give and take of argument, from colleagues who complemented my limited expertise with their equally centered specialties and aptitudes for other aspects of these large and various problems. Thus, on the first leg or branch of hierarchy theory, I worked with Niles Eldredge on punctuated equilibrium, and with Elisabeth Vrba on levels of selection and sorting. On the second leg of structuralist al­ternatives to adaptationist argument, I worked with Dick Lewontin on span­drels, Elisabeth Vrba on exaptation, David Woodruff on the functional and structural morphology of Cerion, and with “the gang of four” (increased to five with the later inclusion of Jack Sepkoski) — Dave Raup, Tom Schopf, Dan Simberloff, and me — on trying to specify how many aspects of apparently or­dered phyletic patterns, heretofore confidently attributed to selection for little reason beyond the visual appearance of order itself, could plausibly be gener­ated within purely random systems. And on the third leg of extrapolationism, my earliest interests in the logic and justification of uniformitarianism in phi­losophy, and of Lyellian perspectives in the history of science, could not have developed without advice and substantial aid (but not collaborative publica­tion this time) with historians Martin Rudwick, Reijer Hooykaas, and Cecil  {28}  Schneer, and with philosophers Nelson Goodman, Bonnie Hubbard, and George Geiger. (Geiger, my mentor at Antioch College, was the last student of John Dewey and played with Lou Gehrig on the Columbia University base­ball team, thus embodying both my professional and avocational interests.)

In fact, and as a comment within the sociology of science, I would venture that future historians might judge the numerous seminal (and published) collaborations between evolutionary biologists and professional philosophers of science as the most unusual and informative operational aspect of the reconstruction of evolutionary theory in the late 20th century. Research scientists tend to be a philistine lot, with organismic biologists perhaps at the head of this particular pack (for we work with “big things” that we can see and un­derstand at our own scale. Thus, we suppose that we can afford to be more purely empirical in our reliance on “direct” observation, and less worried about admittedly conceptual problems of evaluating things too small or too fast to see). Most of us would scoff at the prospect of working with a profes­sional philosopher, regarding such an enterprise as, at best, a pleasant waste of time and, at worst, an admission that our own clarity of thought had be­come addled (or at least as a fear that our colleagues would so regard our in­terdisciplinary collaboration).

And yet, the conceptual problems presented by theories based on causes operating at several levels simultaneously, of effects propagated up and down, of properties emerging (or not) at higher levels, of the interaction of random and deterministic processes, and of predictable and contingent influ­ence, have proven to be so complex, and so unfamiliar to people trained in the simpler models of causal flow that have served us well for centuries (see the next section on Zeitgeist), that we have had to reach out to colleagues ex­plicitly trained in rigorous thinking about such issues. Thus, we learned, to our humbling benefit, that conceptual muddles do not necessarily resolve themselves “automatically” just because a smart person — namely one of us, trained as a scientist — finally decides to apply some raw, naive brain power to the problem. Professional training in philosophy does provide a set of tools, modes and approaches, not to mention a feeling for common dangers and fal­lacies, that few scientists (or few “smart folks” of any untrained persuasion) are likely to possess by the simple good fortune of superior raw brainpower. (We might analogize this silly and vainglorious, although regrettably com­mon, belief to the more popular idea that great athletes should be able to ex­cel at anything physical by reason of their general bodily virtue — a myth and chimaera that dramatically exploded several years ago when Michael Jordan discovered that he could not learn to hit a curve ball, just because he excelled so preeminently in basketball, and possessed the world's best athletic body in general — for he ended up barely hitting over 0.200 in a full season of minor league play. I do, however, honor and praise his persistence in staying the course and taking his lumps.)

Indeed, I know of no other substantial conceptual advance in recent science so abetted by the active collaboration of working scientists and professional philosophers (thus obviating, for once, the perennial, and justified, complaint  {29}  of philosophers of science that no scientists read their journals or even en­counter their analyses). Several key achievements in modern evolutionary theory, particularly the successful resolution of conceptual difficulties in for­mulating a workable theory of hierarchical selection (rooted in concepts like emergence and simultaneous selection at several levels that our minds, with their preferences for two-valued logics, don't handle either automatically, or well at all), have appeared as joint publications of biologists and philoso­phers, including the books of Sober and Wilson, 1998, and Eldredge and Grene, 1992; and articles of Sober and Lewontin, 1982, and Mayo and Gilinsky, 1987. My own understanding of how to formulate an operational theory of hierarchical selection, and my “rescue” from a crucial conceptual error that had stymied my previous thinking (see Chapter 8, pages 656673), emerged from joint work with Elisabeth Lloyd, a professional philosopher of science. I take great pride in our two joint articles (Lloyd and Gould, 1993; Gould and Lloyd, 1999), which, in my partisan judgment, resolve what may have been the last important impediment to the codification of a conceptually coherent and truly operational theory of hierarchical selection.

Zeitgeist. Although major revisions to the structure of evolutionary theory emerged mainly from the conventional substrates of novel data and clearer concepts, we should not neglect the admittedly fuzzier, but by no means unimportant, input from a distinctive social context, or intellectual “spirit of the times” (a literal meaning of Zeitgeist) that, at the dawn of a calendrical millennium, has suffused our general academic culture with a set of loosely coherent themes and concerns far more congenial with the broad revisions here proposed within evolutionary theory than any previous set of guiding concepts or presuppositions had been. Needless to say, Zeitgeists are two edged swords of special sharpness — for either they encourage sheeplike conformity with transient ghosts of time (another literal meaning of Zeitgeist) that will soon fade into oblivion, or they open up new paths to insights that previous ages could not even have conceptualized. Any intellectual would therefore be a fool to argue that conformity with a Zeitgeist manifests any preferential correlation with scientific veracity ipso facto. Zeitgeists can only suggest or facilitate.

Nonetheless, we would be equally foolish in our naive empiricism if we claimed that major advances in science must be entirely data driven, and that social contexts can only act as barriers to our vision of nature's factuality. Both the social and scientific world were “ready” for evolution in the mid 19th century. People of equal intelligence could neither have formulated nor owned such a concept in Newton's generation, even if some hypothetical Dar­win had then advanced such a claim (and probably ended up in Bedlam for his troubles). In Chapter 2, I shall document not only this general readiness of Western science within the Zeitgeist of Darwin's time, but also the specific so­cial impetus that Darwin gained from studying the distinctive theories (also a product of the earlier Enlightenment Zeitgeist, and not accessible before) of Adam Smith and the Scottish economists. Thus, and by analogy a century later, the altered Zeitgeist of our own time may also facilitate a fruitful  {30}  reconsideration of major evolutionary concepts that still bear the originating stamp of a Victorian scientific context strongly committed to unidirectional, single-level and deterministic views of natural causality — subtly controlling con­cepts that many scientists would now label as limiting and outmoded.

Although the next few paragraphs will be the most vague and impressionis­tic (I trust) of the entire book, I venture these ill-formulated statements about Zeitgeist because I feel that something important lurks behind my inability to express these inchoate thoughts with precision. I argue above (page 14) that the key concerns of the three essential branches of Darwinian logic might be identified as agency, efficacy and scope of natural selection. In each of these domains, I believe, the revised structure of evolutionary theory, as presented in this book, might be characterized as expansion and revision according to a set of coordinated principles, all consonant with our altered Zeitgeist vs. the scientific spirit of Darwin's own time. The modern revision seeks to replace Darwin's unifocal theory of organismic selection with a hierarchical account (leg one); his unidirectional theory of adaptational construction in the func­tionalist mode with a more balanced interaction of these external causes, treating internal (or structural) constraints primarily as positive channels, and not merely as limitations (leg two); and his unilevel theory of micro-evolutionary extrapolation with a model of distinctive but interacting modes of change, each characteristic for its tier of time. In short, a hierarchy of inter­acting levels, each important in a distinctive way, for Darwin's single locus; an interaction of environmental outsides with organic insides for Darwin's single direction of causal flow; and a set of distinctive temporal tiers for Dar­win's attempt to situate all causality in the single microevolutionary world of our own palpable moments.

I do sense a common underlying vision behind all these proposed reforms. Strict Darwinism, although triumphant within mid 20th century evolutionary theory, embodied several broad commitments (philosophical or metatheoretical, in the technical sense of these terms), more characteristic of 19th than of 20th century thought (and, obviously, not necessarily wrong, or even to be discounted, for this reason — as nothing can be more dangerous to the prog­ress of science than winds of fashion, and we do, after all, learn some things, and develop some fruitful approaches, with validity and staying power well beyond their time of origin and initial popularity). Some aspects of Darwin's formulation broke philosophical ground in a sense quite consonant with our modern Zeitgeist of emphasis upon complexity and interaction — particularly, Darwin's focus on the interplay of chance and necessity in sources of varia­tion vs. mode of selection. Indeed, Darwin paid the usual price for such inno­vation in the failure of nearly all his colleagues, even the most intellectually acute, to grasp such a radical underlying philosophy. But, in many command­ing respects, Darwinism follows the norms of favored scientific reasoning in his time.

The logic of Darwin's formulation rests upon several preferences in scien­tific reasoning more characteristic of his time than of ours — preferences that  {31}  many scientists would now view as unduly restrictive in their designation of a privileged locus of causality, a single direction of causal flow, and a smooth continuity in resulting effects. Classical Darwinism follows standard reduc­tionist preferences in designating the lowest level then available — the organ­ism, for Darwin — as an effectively unique locus of causality (the first leg of agency). In this sense, the efforts of Williams and Dawkins (see Chapter 8) to reduce the privileged locus even further to the genic level (perforce unavail­able to Darwin) should be read as a furthering and intensification of Darwin's intent — in other words, a basically conservative adumbration of Darwin's own spirit and arguments, and not the radical conceptual revision that some have imagined.

At this single level of causality, classical Darwinism then envisages a similarly privileged direction of causal flow, as information from the environment (broadly construed, of course, to include other organisms as well as physical surroundings) must impact the causal agent (organisms struggling for repro­ductive success) and be translated, by natural selection, into evolutionary change. The organism supplies raw material in the form of “random” varia­tion, but does not “push back” to direct the flow of its own alteration from inside. Darwinism, in this sense, is a functionalist theory, leading to local ad­aptation as the environment proposes and natural selection disposes. Finally, classical Darwinism completes a trio of privileged causal places and conse­quently directional flows by postulating strict continuity in results, as local selection scales smoothly through the immensity of geological time to engen­der life's history by pure extrapolation of lowest-level modes and causes.

By contrast, the common themes behind the reformulations defended in this book all follow from serious engagement with complexity, interaction, multiple levels of causation, multidirectional flows of influence, and pluralis­tic approaches to explanation in general — a set of integrated approaches that strongly contribute to the Zeitgeist of our moment. To anticipate and make a preemptive strike against the obvious counterattack from Darwinian tradi­tionalists, these alternative themes do not substitute a “laid back, laissez-faire, anything goes” kind of sloppy tolerance for contradiction and fuzziness in argument against the genuine rigor of old-line Darwinism. The social and psychological contributions of a Zeitgeist to the origin of hypotheses bear no logical relationship to any subsequent scientific defense and validation of the same hypotheses. Moreover, on this subject of test and confirmation, I es­pouse a rigorously conventional and rather old-fashioned “realist” view that an objective factual world exists “out there,” and that science can access its ways and modes. Whatever the contribution of a Victorian Zeitgeist to Darwin's thinking, or of a contemporary Zeitgeist to our revisions, the dif­ferences are testable and subject to validation or disproof by the usual arma­mentarium of scientific methods. That is, either Darwin is right and effec­tively all natural selection occurs at the organismic level (despite the logical conceivability of other levels), or the hierarchical theory is right and several levels make interestingly different and vitally important simultaneous  {32}  contributions to the overall pattern of evolution. The same ordinary form of testability can be applied to any other contrast between strict Darwinism and the revised and expanded formulations defended in this book.

As the most striking general contrast that might be illuminated by reference to the different Zeitgeists of Darwin's time and our own, modern revisions for each essential postulate of Darwinian logic substitute mechanics based on in­teraction for Darwin's single locus of causality and directional flow of effects. Thus, for Darwin's near exclusivity of organismic selection, we now propose a hierarchical theory with selection acting simultaneously on a rising set of levels, each characterized by distinctive, but equally well-defined, Darwinian individuals within a genealogical hierarchy of gene, cell-lineage, organism, deme, species, and clade. The results of evolution then emerge from complex, but eminently knowable, interactions among these potent levels, and do not simply flow out and up from a unique causal locus of organismal selection.

A similar substitution of interaction for directional flow then pervades the second branch of selection's efficacy, as Darwin's functionalist formulation — with unidirectional flow from an external environment to an isotropic or­ganic substrate that supplies “random” raw material but imposes no direc­tional vector of its own to “push back” from internal sources of constraint — yields to a truly interactive theory of balance between the functionalist Dar­winian “outside” of natural selection generated by environmental pressures, and a formalist “inside” of strong, interesting and positive constraints gen­erated by specific past histories and timeless structural principles. Finally, on the third and last branch of selection's range, the single and control­ling microevolutionary locus of Darwinian causality yields to a multileveled model of tiers of time, with a unified set of processes working in distinctive and characteristic ways at each scale, from allelic substitution in observable years to catastrophic decimation of global biotas. Thus, and in summary, for the unifocal and noninteractive Darwinian models of exclusively organismal selection, causal flow from an environmental outside to an organismal in­side, and a microevolutionary locus for mechanisms of change that smoothly extrapolate to all scales, we substitute a hierarchical selectionist theory of numerous interacting levels, a balanced and bidirectional flow of causal­ity between external selection and internal constraint (interaction of func­tionalist and structuralist perspectives), and causal interaction among tiers of time.

Among the many consequences of these interactionist reformulations, punctuational rather than continuationist models of change (with stronger structuralist components inevitably buttressing the punctuational versions) may emerge as the most prominent and most interesting. The Darwinian me­chanics of functionalism yield an expectation of continuously improving lo­cal adaptation, with longterm stability representing the achievement of an optimum. But interactionist and multi-leveled models of causality reconcep­tualize stasis as a balance, actively maintained among potentially competing forces at numerous levels, with change then regarded as exceptional rather than intrinsically ticking most of the time, and punctuational rather than  {33}  smoothly continuous when it does occur (representing the relatively quick transition that often accompanies a rebalancing of forces).

To end this admittedly vague section with the punch of paradox (and even with a soundbite), I would simply note the almost delicious irony that the for­mulation of a hierarchical theory of selection — the central concept of this book, and invoking a non-vernacular meaning of hierarchy in the purely structural sense of rising levels of inclusivity — engenders, as its most impor­tant consequence, the destruction of a different and more familiar meaning of hierarchy: that is, the hierarchy of relative value and importance embod­ied in Darwin's privileging of organismic selection as the ultimate source of evolutionary change at all scales. Thus, a structural and descriptive hierarchy of equally effective causal levels undermines a more conventional hierarchy of relative importance rooted in Darwin's exclusive emphasis on the micro-evolutionary mechanics of organismal selection. And so, this structuralist view of nature's order enriches the structure of evolutionary theory — carry­ing the difference between strict Darwinism and our current understanding through more than enough metatheoretical space to fashion a Falconerian, not merely a Darwinian, rebuilding and extension for our edifice of coherent explanation.


For reasons beyond mere self-indulgence or egotism, I believe that defenders of such general theories about large realms of nature owe their readers some explanation for the personal bases and ontogeny of their choices — for at this level of abstraction, no theory can claim derivation by simple logical or em­pirical necessity from observed results, and all commitments, however well defended among alternative possibilities, will also be influenced by authorial preferences of a more contingent nature that must then be narrated in order to be understood. Moreover, and in this particular case, the structure of this book includes a set of vigorously idiosyncratic features that, if not acknowl­edged and justified, might obscure the far more important raison d'etre for its composition: the presentation of a tight brief for substantial reformulation in the structure of evolutionary theory, with all threads of revision conceptually united into an argument of different thrust and form, but still sufficiently con­tinuous with its original Darwinian base to remain within the same intellec­tual lineage and logic.

Two aspects of my idiosyncratic procedures require explicit commentary here because, at least as my intention, they should reinforce this book's cen­tral argument for coherence (logical, historical and empirical) of the revised and general structure of evolutionary theory, and not further the opposite, al­beit customary, function of such “confessional” writing — namely, to slake authorial egos, fight old battles, and relate twice-told tales to one's own ad­vantage (although I claim no immunity from these all too human foibles).

This book will be published in the Spring of 2002, an auspicious and palindromic year just one step out of the starting gate for a new millennium.  {34}  At the same time, and fortuitously, my 10th and last volume of monthly es­says in Natural History Magazine, written without a single break from Janu­ary 1974 to January 2001, will also appear in print. In an eerie coincidence (with no meaning that I can discern), my first technical book, Ontogeny and Phylogeny, appeared exactly 25 years before, in 1977, also at the same time as my first book of Natural History essays, Ever Since Darwin. This odd and twofold simultaneous appearance, 25 years apart, of my best youthful efforts in the contrasting (but not really conceptually different) realms of technical and popular science, and then of my best shots from years of greater maturity in the same two realms, has forced me to think long and hard about the meaning of continuity, commitment and personal perspective.

My popular volumes fall into the explicit and well recognized category of essays, a literary genre defined, ever since Montaigne's initiating 16th century efforts, as the presentation of general material from an explicitly personal and opinionated point of view — although the best essays (literally meaning “attempts,” after all) tend to be forthright in their expression of opinions, generous (or at least fair) to other views, and honest in their effort to specify the basis of authorial preferences. On the other hand, technical treatises in science do not generally receive such a license for explicitly personal expres­sion. I believe that this convention in technical writing has been both harmful and more than a bit deceptive. Science, done perforce by ordinary human beings, expressing ordinary motives and foibles of the species, cannot be grasped as an enterprise without some acknowledgment of personal dimen­sions in preferences and decisions — for, although a final product may display logical coherence, other decisions, leading to other formulations of equally tight structure, could have been followed, and we do need to know why an author proceeded as he did if we wish to achieve our best understanding of his accomplishments, including the general worth of his conclusions.

Logical coherence may remain formally separate from ontogenetic con­struction, or psychological origin, but a full understanding of form does re­quire some insight into intention and working procedure. Perhaps some pre­sentations of broad theories in the history of science — Newton's Principia comes immediately to mind — remain virtually free of personal statement (sometimes making them, as in this case, virtually unreadable thereby). But most comprehensive works, in all fields of science, from Galileo's Dialogo to Darwin's Origin, gain stylistic strength and logical power by their suffu­sion with honorable statements about authorial intents, purposes, prejudices, and preferences. I cannot think of a single major book in natural history — from Buffon's Histoire naturelle and Cuvier's Ossemens fossiles to Simpson's Tempo and Mode, and Mayr's Animal Species — that does not include such extensive personal information, either in explicit sections, or inserted by-the-by throughout. (Even so abstract a presentation as R. A. Fisher's Genetical Theory of Natural Selection gains greatly in comprehension through its long and final, if in retrospect regrettable, section on the author's idiosyncratic eugenical views about human improvement.) I have included personal discus­sion throughout this text, but let me also devote a few explicit pages to the  {35}  two points that I regard as most crucial to understanding the general argu­ment through (or despite) conscious idiosyncrasies in my presentation.


Many technical treatises in science begin with a short section on previous his­tory of work in the field — usually written in the hagiographical mode to de­pict prior history as a march towards final truths revealed in the current vol­ume. Sometimes, authors get a bit carried away, and these historical sections expand into substantial parts of the final book. Lest anyone make the false in­ference that my full first half of history arose in this haphazard and initially unintended way, I hasten to assure readers that my final result was my inten­tion from the start.

For several reasons, I always conceived this book as a smooth joining of two halves, roughly equal in length and importance. First, and ontogenetically, I had written my earlier technical book, Ontogeny and Phylogeny, in this admittedly unusual manner — and I remain pleased with both the distinctiveness and the efficacy of the result. Second, I believe that the history of evo­lutionary thought, and probably of any other subject imbued with such im­portance to our lives and to our understanding of nature, constitutes an epic tale of fascinating, and mostly honorable, people engaged in a great struggle to comprehend something very deep and very difficult. Thus, such histories capture a bit of the best in us (also of the worst, but all human endeavors so conspire) — a bit, moreover, that cannot be expressed in any other way. We re­ally do need to honor the temporal substrate of our current understanding, not only as a guide to our continuing efforts, but also as a moral obligation to our forebears.

But a third and practical reason trumps all others. Although I would not state such a claim as a generality for all scientific analyses, in this particular case I do not see how the structure of evolutionary theory can be resolved and the appropriate weights of relative importance assigned to the different com­ponents thereof, absent such a historical perspective. (Would it not be odd to claim, in any case, that the quintessential science for resolving the nature of life's history can itself be understood as a pristine construction, a fully-formed conceptual entity drawn intact from some analog of Zeus's brow, rather than an “organic” structure of ideas with its own ontogeny and history?)

To give one example at the largest and at the smallest scales of my argu­ment, I don't know how I could have properly defended my identification and explication of the threefold essence of Darwinian logic without documenting the history of theoretical debate in order to tease out the components that have always been most troubling, most central, and most directive. A pure description of the theory's abstract logic simply will not suffice. To epitomize, I have identified these essential components on three basic grounds: that logic compels (Chapter 2), that history validates (Chapters 37), and that current debate reaffirms (Chapters 812). The middle term of this epitome unites the end members; I cannot present a coherent or compelling defense without this linkage. The three issues of agency, efficacy and scope build the Darwinian  {36}  essence both because the logical structure of the theory so dictates, and because the history and current utility of the theory so document.

To complement this most general statement with just one example of the utility of historical analysis at the smaller scale of details, I offer the following case as the strongest argument for my central claim that Darwin's brave at­tempt to construct a single-level, exclusively organismic theory of natural se­lection must fail in principle, and that all selectionists must eventually own a hierarchical model. What better evidence can we cite than the historical dem­onstration (see Chapters 3 and 5 for details) that each of the only three foundational thinkers who truly understood the logic of selectionism — August Weismann, Hugo de Vries, and Charles Darwin himself — tried mightily to make the single-level version work as a fully sufficient explanation for evolu­tion. And each failed, after intense intellectual struggle, and for fascinatingly different reasons documented later in the book — Darwin for explaining di­versity by reluctant resort to species selection; Weismann for a strongest ini­tial commitment to a single level, and an eventual recognition of full hierar­chy as the most important and distinctive conclusion of his later career (by his own judgment); and de Vries for reconciling his largely psychological fealty to Darwin as his intellectual hero, with his clearly non-Darwinian account of the origin of species and the explanation of trends (including an explicit coin­ing of the term “species selection” for explaining cladal patterns).

One might cite various truisms telling us that people ignorant of history will be condemned to repeat its errors. But I would rather re-express this ac­curate and rueful observation in a more positive manner by illustrating the power of historical analysis to aid both our current understanding and the depth of our appreciation for the intellectual importance of our enterprise. Finally, and to loosen the rein on personal bravado that I usually try to hold at least somewhat in check, no scholar should impose a project of this length upon his colleagues unless he believes that some quirk of special skill or expe­rience permits him to proceed in a unique manner that may offer some insight to others. In my case, and only by history's fortune of no immediate competi­tion in a small field, I may be able to combine two areas of professional com­petence not otherwise conjoined among current evolutionists. I am not a credentialed historian of science, but I believe that I have done sufficient work in this field (with sufficient understanding of the difference between the Whiggish dilettantism of most enthusiastic amateurs, and the rigorous meth­ods applied by serious scholars) to qualify as adequately knowledgeable. (At least I can read all the major works in their original languages, and I stay close to the “internalist” style of analysis that people who understand the logic and history of theories, but cannot claim truly professional expertise in the “externalist” factors of general social and historical context, can usefully pursue.) Meanwhile I am, for my sins, a lifelong and active professional pale­ontologist, a commitment that began at age five as love at first sight with a di­nosaur skeleton.

Many historians possess deeper knowledge and understanding of their immediate subject than I could ever hope to acquire, but none enjoy enough  {37}  intimacy with the world of science (knowing its norms in their bones, and its quirks and foibles in their daily experience) to link this expertise to contem­porary debates about causes of evolution. Many more scientists hold superb credentials as participants in current debates, but do not know the historical background. As I hope I demonstrated by practical utility in The Mismeasure of Man (Gould, 1981a), a small and particular — but I think quite impor­tant — intellectual space exists, almost entirely unoccupied, for people who can use historical knowledge to enlighten (not merely to footnote or to pret­tify) current scientific debates, and who can then apply a professional's “feel” for the doing of science to grasp the technical complexities of past debates in a useful manner inaccessible to historians (who have therefore misinter­preted, in significant ways, some important incidents and trends in their sub­ject). I only hope that I have not been wrong in believing that my devotion of a lifetime's enthusiasm to both pursuits might make my efforts useful, in a distinctive way, to my colleagues.


I admire my friend Oliver Sacks extravagantly as a writer, and I could never hope to match him in general quality or human compassion. He once said something that touched me deeply, despite my continuing firm disagreement with his claim (while acknowledging the validity of the single statement rele­vant to the present context). Oliver said that he envied me because, although we had both staked out a large and generous subject for our writing (he on the human mind, me on evolution), I had enjoyed the privilege of devising and developing a general theory that allowed me to coordinate all my work into a coherent and distinctive body, whereas he had only written descrip­tively and aimlessly, albeit with some insight, because no similar central focus underlay his work. I replied that he had surely sold himself short, because he had been beguiled by conventional views about the nature and limits of what may legitimately be called a central scientific theory — and that he certainly held such an organizing concept in his attempt to reintroduce the venerable “case study method” of attention to irreducible peculiarities of individual pa­tients in the practice of cure and healing in medicine. Thus, I argued, he held a central theory about the importance of individuality and contingency in gen­eral medical theory, just as I and others had stressed the centrality of histori­cal contingency in any theoretical analysis and understanding of evolution and its actual results.

Oliver saw the theory of punctuated equilibrium itself, which I developed with Niles Eldredge and discuss at inordinate length in Chapter 9, as my coordinating centerpiece, and I would not deny this statement. But punctuated equilibrium stands for a larger and coherent set of mostly iconoclastic con­cerns, and I must present some intellectual autobiography to explain the rea­sons and the comings together, as best I understand them myself — hence my rip-off of Cardinal Newman's famous title for the best similar effort ever made, albeit in a maximally different domain. In his Apologia Pro Vita Sua (an apology for one's own life), Newman intends the operative word as I do,  {38}  in its original and positive meaning, not in the currently more popular nega­tive sense — “something said or written in defense or justification of what ap­pears to others to be wrong or of what may be liable to disapprobation” (per Webster's).

As my first two scientific commitments, I fell in love with paleontology when I met Tyrannosaurus in the Museum of Natural History at age five, and with evolution at age 11, when I read G. G. Simpson's The Meaning of Evolu­tion, with great excitement but minimal comprehension, after my parents, as members of a book club for folks with intellectual interests but little eco­nomic opportunity or formal credentials, forgot to send back the “we don't want anything this month” card, and received the book they would never have ordered (but that I begged them to keep because I saw the little stick fig­ures of dinosaurs on the dust jacket). Thus, from day one, my developing pro­fessional interests united paleontology and evolution. For some reason still unclear to me, I always found the theory of how evolution works more fasci­nating than the realized pageant of its paleontological results, and my major interest therefore always focused upon principles of macroevolution.* I did come to understand the vague feelings of dissatisfaction (despite Simpson's attempt to resolve them in an orthodox way by incorporating paleontology within the Modern Synthesis) that some paleontologists have always felt with the Darwinian premise that microevolutionary mechanics could construct their entire show just by accumulating incremental results through geological immensity.

As I began my professional preparation for a career in paleontology, this vague dissatisfaction coagulated into two operational foci of discontent. First (and with Niles Eldredge, for we worried this subject virtually to death as graduate students), I became deeply troubled by the Darwinian convention that attributed all non-gradualistic literal appearances to imperfections of the geological record. This traditional argument contained no logical holes, but the practical consequences struck me as unacceptable (especially at the out­set of a career, full of enthusiasm for empirical work, and trained in statis­tical techniques that would permit the discernment of small evolutionary  {39}  changes). For, by the conventional rationale, the study of microevolution be­came virtually nonoperational in paleontology — as one almost never found this anticipated form of gradual change up geological sections, and one there­fore had to interpret the vastly predominant signal of stasis and geologically abrupt appearance as a sign of the record's imperfection, and therefore as no empirical guide to the nature of evolution. Second, I became increasingly dis­turbed that, at the higher level of evolutionary trends within clades, the ma­jority of well documented examples (reduction of stipe number in graptolites, increasing symmetry of crinoidal cups, growing complexity of ammonoid su­tures, for example) had never been adequately explained in the terms de­manded by Darwinian convention — that is, as adaptive improvements of constituent organisms in anagenetic sequences. Most so-called explana­tions amounted to little more than what Lewontin and I, following Kipling, would later call “just-so stories,” or plausible claims without tested evidence, whereas other prominent trends couldn't even generate a plausible story in adaptationist terms at all.

As Eldredge and I devised punctuated equilibrium, I did use the theory to resolve these two puzzles to my satisfaction, and each resolution, when finally generalized and further developed, led to my two major critiques of the first two branches of the essential triad of Darwinian central logic — so Oliver Sacks's identification of punctuated equilibrium as central to my theoretical world holds, although more as a starting point than as a coordinating focus. By accepting the geologically abrupt appearance and subsequent extended stasis of species as a fair description of an evolutionary reality, and not only as a sign of the poverty of paleontological data, we soon recognized that spe­cies met all criteria for definition and operation as genuine Darwinian indi­viduals in the higher-level domain of macroevolution — and this insight (by complex routes discussed in Chapter 9) led us to concepts of species selection in particular and, eventually, to the full hierarchical model of selection as an interesting theoretical challenge and contrast to Darwinian convictions about the exclusivity of organismal selection. In this way, punctuated equilibrium led to the reformulation proposed herein for the first branch of essential Dar­winian logic.

Meanwhile, in trying to understand the nature of stasis, we initially fo­cused (largely in error, I now believe) upon internal constraints, as vaguely represented by various concepts of “homeostasis,” and as exemplified in the model of Galton's polyhedron (see Chapter 4). These thoughts led me to extend my doubts about adaptation and the sufficiency of functionalist mechanisms in general — especially in conjunction with my old worries about paleontological failures to explain cladal trends along traditional adapta­tionist lines. Thus, these aspects of punctuated equilibrium strongly contrib­uted to my developing critiques of adaptationism and purely functional me­chanics on the second branch of essential Darwinian logic (although other arguments struck me as even more important, as discussed below).

Nonetheless, and despite the centrality of punctuated equilibrium in developing a broader critique of conventional Darwinism, my sources extended  {40}  outward into a diverse and quirky network of concerns that seemed, to me and at first, isolated and uncoordinated, and that only later congealed into a coherent critique. For this curious, almost paradoxical, reason, I have be­come even more convinced that the elements of my overall critique hang to­gether, for I never sensed the connections when I initially identified the com­ponents as, individually, the most challenging and intriguing items I had encountered in my study of evolution. When one accumulates a set of things only for their independent appeals, with no inkling that any common intellec­tual ground underlies the apparent miscellany, then one can only gain con­fidence in the “reality” of a conceptual basis discerned only later for the cohe­sion. I would never argue that this critique of strict Darwinism gains any higher probability of truth value for initially infecting me in such an uncoor­dinated and mindless way. But I would assert that a genuinely coherent and general alternative formulation must exist “out there” in the philosophical universe of intellectual possibilities — whatever its empirical validity — if its isolated components could coagulate, and be discerned and selected, so un­consciously.

If I may make a somewhat far-fetched analogy to my favorite Victorian novel, Daniel Deronda (the last effort of Darwin's friend George Eliot), the hero of this story, a Jew raised in a Christian family with no knowledge of his ethnic origins, becomes, as an adult, drawn to a set of apparently inde­pendent activities with no coordinating theme beyond their relationship, en­tirely unknown to Deronda at the time of his initial fascination, to Jewish history and customs. Eventually, he recognizes the unifying theme behind such apparent diversity, and learns the truth of his own genetic background. (I forgive Eliot for this basically silly fable of genealogical determinism be­cause her philosemitic motives, however naive and a bit condescending, shine forth so clearly in the surrounding antisemitic darkness of her times.) But I do feel, to complete the analogy, rather like a modern, if only culturally or psy­chologically predisposed, Deronda who gathered the elements of a coherent critique solely because he loved each item individually — and only later sensed an underlying unity, which therefore cannot be chimaerical, but may claim some logical existence prior to any conscious formulation on my part.

In fact, the case for an external and objective coherence of this alternative view of evolution seems even stronger to me because I gathered the indepen­dent items not only in ignorance of their coordination, but also at a time when I held a conscious and conventional view of Darwinian evolution that would have actively denied their critical unity and meaning. I fledged in sci­ence as a firm adaptationist, utterly beguiled by the absolutist beauty (no doubt, my own simplistic reading of a more subtle, albeit truly hardened, Modern Synthesis) of asserting, a la Cain and other ecological geneticists of the British school, that all aspects of organismal phenotypes, even the most trivial nuances, could be fully explained as adaptations built by natural selec­tion.

I remember two incidents of juvenilia with profound embarrassment to­day: First, an undergraduate evening bull session with the smartest physics  {41}  major at Antioch College, as his skepticism evoked my stronger insistence that our science matched his in reductionistic rigor because “we” now knew for certain that natural selection built everything for optimal advantage, thus making evolution as quantifiable and predictive as classical physics. Second, as a somewhat more sophisticated, but still beguiled, assistant professor, I re­member my profound feeling of sadness and disappointment, nearly amount­ing to an emotional sense of betrayal, upon learning that an anthropological colleague favored drift as the probable reason for apparently trivial genetic differences among isolated groups of Papua-New Guinea peoples. I remem­ber remonstrating with him as follows: Of course your argument conforms to logic and empirical possibility, and I admit that we have no proof either way. But your results are also consistent with selection — and our panselectionist paradigm has forged a theory of such beauty and elegant simplicity that one should never favor exceptions for their mere plausibility, but only for docu­mented necessity. (I recall this discussion with special force because my emo­tional feelings were so strong, and my disappointment in his “unnecessary apostasy” so keen, even though I knew that neither of us had the empirical “goods.”) Finally, if I could, in a species of Devil's bargain, wipe any of my publications off the face of the earth and out of all memory, I would gladly nominate my unfortunately rather popular review article on “Evolutionary paleontology and the science of form” (Gould, 1970a) — a ringing paean to selectionist absolutism, buttressed by the literary barbarism that a “quantifunctional” paleontology, combining the best of biometric and mechanical analyses, could prove panadaptationism even for fossils that could not be run through the hoops of actual experiments.

Against this orthodox background — or, rather, within it and quite unconsciously for many years — I worked piecemeal, producing a set of separate and continually accreting revisionary items along each of the branches of Darwinian central logic, until I realized that a “Platonic” something “up there” in ideological space could coordinate all these critiques and fascinations into a revised general theory with a retained Darwinian base.

The first branch of levels in selection proceeded rather directly and linearly because the generality flowed so clearly from punctuated equilibrium itself, once Eldredge and I finally worked through the implications and extensions of our own formulations (Eldredge and Gould, 1972). Steve Stanley (1975) and Elisabeth Vrba (1980) helped to show us what we had missed in rami­fications leading from the phenomenology of stasis and geologically abrupt appearance, to recognizing species as genuine Darwinian individuals, to des­ignating species as, therefore and potentially, the basic individuals of macro-evolution (comparable with the role of the organism in microevolution), to the validity of species selection, and eventually to the full hierarchical model and its profound departure from the exclusively organismal accounts of con­ventional Darwinism (or the even more reduced and equally monistic genic versions of Williams and Dawkins) — see Vrba and Gould, 1986. Finally, by adopting the interactor rather than the replicator approach to defining selec­tion, and by recognizing emergent fitness, rather than emergent characters, as  {42}  the proper criterion for identifying higher-level selection (Lloyd and Gould, 1993; Gould and Lloyd, 1999), I think that we finally reached, by a circuitous route around many stumbling blocks of my previous stupidity, a consistent and truly operational theory of hierarchical selection (see Chapter 8).

I must also confess to some preconditioning beyond punctuated equilib­rium. I had admired Wynne-Edwards's pluck (1962) from the start, even though I agreed with Williams's (1966) trenchant criticisms of his particular defenses for group selection, rooted in the ability of populations to regulate their own numbers in the interests of group advantage. Still, I felt, for no rea­son beyond vague intuition, that group selection made logical sense and might well find other domains and formulations of greater validity — a feeling that has now been cashed out by modern reformulations of evolutionary the­ory (see especially Wilson and Sober, 1998, and Chapters 8 and 9 herein).

My odyssey on the second branch of balancing internal constraint with external adaptation in understanding the patterning and creative population of novel places in evolutionary morphospace followed a much more com­plex, meandering and diverse set of pathways. As an undergraduate, I loved D'Arcy Thompson's Growth and Form (1917; see Gould, 1971b, for my first “literary” paper), and wrote a senior thesis on his theory of morphology. But I thought that I admired the book only for its incomparable prose, and I at­tacked the anti-Darwinian (and structuralist) components of his theory un­mercifully. I then took up allometry for my first empirical studies, somehow fascinated by structural constraint and correlation of growth, but thinking all the while that my task must center on a restoration of adaptationist themes to this “holdout” bastion of formalist thought — particularly the achievement of biomechanical optima consistent with the Galilean principle of decreasing surface/volume ratios with increasing size in isometric forms. I remain proud of my first review article, dedicated to this subject (Gould, 1966), written when I was still a graduate student, but I am now embarrassed by the fervor of my adaptationist convictions.

I emphasized allometric analysis, now in a directly multivariate reformulation, in my first set of empirical studies on the Bermudian pulmonate snail Poecilozonites (see especially Gould, 1969 — the published version of my Ph.D. dissertation). And yet, of all the long and largely adaptationist treatises in this series, and for some reason that I could not identify at the time, the conclusion that I reached with most satisfaction, and that I somehow re­garded as most theoretically innovative (without knowing why), resided in a short, and otherwise insignificant, article that I wrote for a specialized pale-ontological journal on a case of convergence produced by structural necessity, given modes of coiling and allometry in this genus, rather than by selectionist honing (for some cases rested upon ecophenotypic expression, others on paedomorphosis, and still others on gradual change that could be read as conventionally adaptive): “Precise but fortuitous convergence in Pleistocene land snails” (Gould, 1971c).

Five disparate reasons underlie my more explicit recognition, during the 1970's and early 1980's, of the importance and theoretical interest (and  {43}  iconoclasm versus Darwinian traditions) of nonadaptationist themes rooted in structural and historical constraint. First, I stood under the dome of San Marco during a meeting in Venice and then wrote a notorious paper with Dick Lewontin on the subject of spandrels, or nonadaptive sequelae of prior structural decisions (Gould and Lewontin, 1979 — see Chapter 11, pp. 12461258). Second, I recognized, with Elisabeth Vrba, that the lexicon of evolu­tionary biology possessed no term for the evidently important phenomenon of structures coopted for utility from different sources of origin (including nonadaptive spandrels), and not directly built as adaptations for their cur­rent function. We therefore devised the term “exaptation” (Gould and Vrba, 1982) and explored its implications for structuralist revisions to pure Dar­winian functionalism. Third, I worked with a group of paleontological colleagues (Raup et al, 1973; Raup and Gould, 1974; Gould et al., 1977) to develop more rigorous criteria for identifying the signals that required selec­tionist, rather than stochastic, explanation of apparent order in phyletic pat­terns. This work left me humbled by the insight that our brains seek pattern, while our cultures favor particular kinds of stories for explaining these pat­terns — thus imposing a powerful bias for ascribing conventional determinis­tic causes, particularly adaptationist scenarios in our Darwinian traditions, to patterns well within the range of expected outcomes in purely stochastic sys­tems. This work sobered me against such a priori preferences for adaptation­ist solutions, so often based upon plausible stories about results, rather than rigorous documentation of mechanisms.

Fourth, and most importantly, I read the great European structuralist literatures in writing my book on Ontogeny and Phylogeny (Gould, 1977b). I don't see how anyone could read, from Goethe and Geoffroy down through Severtzov, Remane and Riedl, without developing some appreciation for the plausibility, or at least for the sheer intellectual power, of morphological ex­planations outside the domain of Darwinian functionalism — although my resulting book, for the last time in my career, stuck closely to selectionist or­thodoxy, while describing these alternatives in an accurate and sympathetic manner. Fifth, my growing unhappiness with the speculative character of many adaptationist scenarios increased when, starting in the mid 1970's, the growing vernacular (and some of the technical) literature on sociobiology touted conclusions that struck me as implausible, and that also (in some cases) ran counter to my political and social beliefs as well.

Personal distaste, needless to say, bears no necessary relationship to scien­tific validity. After all, what could be more unpleasant, but also more fac­tually undeniable, than personal mortality? But when distasteful conclusions gain popularity by appealing to supposedly scientific support, and when this “support” rests upon little more than favored speculation in an orthodox mode of increasingly dubious status, then popular misuse can legitimately sharpen a scientist's sense of unhappiness with the flawed theoretical basis be­hind a particular misuse. In any case, I trust that this compendium of reasons will dispel Cain's (1979) hurtful assertion that Lewontin, I, and other evolu­tionists who questioned early forms of sociobiology by developing a general  {44}  critique of adaptationism, had acted cynically, and even anti-scientifically, in opposing biological theories that we knew to be true because we disliked their political implications for explaining human behavior. My own growing doubts about adaptationism arose from several roots, mostly paleontological, with any displeasure about sociobiology serving as a late and minor spur to further examination and synthesis.

I then tried to apply my general critique of pure Darwinian functionalism, and my conviction that important and positive constraints could be actively identified by quantitative morphometric study (and not merely passively in­ferred from failures of adaptationist scenarios) in my work on “covariance sets” in the growth, variation, and evolution of the West Indian pulmonate Cerion (Gould, 1984b and c), a snail that encompasses its maximal diver­sity in overt form among populations within a constraining set of pervasive allometries in growth. I discuss some of this work in my text on the empirical validation of positive constraint (see Chapter 10, pages 10451051).

My doubts on the third branch of extrapolationism and uniformity began even earlier, and in a more inchoate way, but then gained expression in my ef­forts in the history of science, and not so much in my direct empirical work — hence, in part, the reduced attention devoted to this theme (Chapters 6 and 12) compared with the first two branches of selection's agency and efficacy. On a fieldtrip in my freshman geology course, my professor took us to a trav­ertine mound and argued that the deposit must be about 11,000 years old because he had measured the current rate of accumulation and then extrapo­lated back to a beginning. When I asked how he could assume such con­stancy of rate, he replied that the fundamental rule of geological inference, something called “the principle of uniformitarianism” permitted such infer­ences because we must regard the laws of nature as constant if we wish to reach any scientific conclusions about the past. This argument struck me as logically incorrect, and I pledged myself to making a rigorous analysis of the reasons.

As a joint major in geology and philosophy, I studied this issue throughout my undergraduate years, producing a paper entitled “Hume and uniform­itarianism” that eventually transmogrified into my first publication (Gould, 1965), “Is uniformitarianism necessary?” (Norman Newell, my graduate ad­visor, urged me to send the paper to Science where, as I learned to my amuse­ment much later, my future “boss” at Harvard, the senior paleontology pro­fessor Bernie Kummel, rejected it roundly as a reviewer. Properly humbled — although I still regard his reasons as ill founded — I then sent the paper to a specialty journal in geology.)

May I share one shameful memory of this otherwise iconoclastic first paper, from which I still draw some pride? In my undergraduate work on this theme, I made a personal discovery (as others did independently) that became impor­tant in late 20th-century studies of the history of geology. I had been schooled in the conventional view that the catastrophists (aka “bad guys”) had in­voked supernatural sources of paroxysmal dynamics in order to compress the earth's history into the strictures of biblical chronology. I read and reread all  {45}  the classical texts of late 18th and early 19th century catastrophism in their original languages — and I could find no claim for supernatural influences upon the history of the earth. In fact, the catastrophists seemed to be advanc­ing the opposite claim that we should base our causal conclusions upon a lit­eral reading of the empirical record, whereas the uniformitarians (aka “good guys”) seemed to be arguing, in an opposing claim less congenial with the ste­reotypical empiricism of science, that we must make hypothetical inferences about the gradualistic mechanics that a woefully imperfect record does not permit us to observe directly.

But, although I had developed and presented an iconoclastic exegesis of Lyell, I simply lacked the courage to state so general a claim for inverting the standard view about uniformitarians and catastrophists. I assumed that I must be wrong, and that I must have misunderstood catastrophism because I had not read enough, or could not comprehend the subtleties at this fledgling state of a career. So I scoured the catastrophist literature again until I found a quote from William Buckland (both a leading divine and the first reader in ge­ology at Oxford) that could be interpreted as a defense of supernaturalism. I cited the quotation (Gould, 1965, p. 223) and stuck to convention on this broader issue, while presenting an original analysis of multiple meanings — some valid (like the invariance of law) and some invalid (like my professor's claim for constancy in range of rates) — subsumed by Lyell under the singular description of “uniformity” in nature.

This work led me, partly from shame at my initial cowardice, and as others reassessed the scientific character of catastrophism, to a more general analysis of the potential validity of catastrophic claims, and particularly to an under­standing of how assumptions of gradualism had so stymied and constrained our comprehension of the earth's much richer history. These ideas forced me to question the necessary basis for Darwin's key assumption that observable, small-scale processes of microevolution could, by extension through the im­mensity of geological time, explain all patterns in the history of life — namely, the Lyellian belief in uniformity of rate (one of the invalid meanings of the hy­brid concept of uniformitarianism). This exegesis led to a technical book about concepts of time and direction in geology (Gould, 1987b), to an en­larged view that encouraged the development of punctuated equilibrium, and to a position of cautious favor towards such truly catastrophic proposals as the Alvarez theory of mass extinction by extraterrestrial impact — a concept ridiculed by nearly all other paleontologists when first proposed (Alvarez et a\., 1980), but now affirmed for the K-T event, and accepted as an empirical basis for expanding our range of scientifically legitimate hypotheses beyond the smooth extrapolationism demanded by this third branch of Darwinian central logic.

In addition to these disparate accretions of revisionism on the three branches of Darwinian central logic, one further domain — my studies in the history of evolutionary thought — served as a sine qua поп for wresting a co­herent critique from such an inchoate jumble of disparate items. Above all, if I had not studied Darwin's persona and social context so intensely, I doubt  {46}  that I would ever have understood the motivations and consistencies — also the idiosyncrasies of time, place and manner — behind the abstract grandeur of his view of life. History, as I argued before (see p. 35), must not be dis­missed as a humanistic frill upon the adamantine solidity of “real” science, but must be embraced as the coordinating context for any broad view of the logic and reasoning behind a subject so close to the bone of human concern as the science of life's nature and structure. (Of the two greatest revolutions in scientific thought, Darwin surely trumps Copernicus in raw emotional im­pact, if only because the older transition spoke mainly of real estate, and the later of essence.)

Some of my historical writing appeared in the standard professional litera­ture, particularly my thesis about the “hardening” of the Modern Synthesis (Gould, 1980e, 1982a, 1983b), a trend (but also, in part, a drift) towards a stricter and less pluralistic Darwinism. Several full-time historians of science then affirmed this hypothesis (Provine, 1986; Beatty, 1988; Smocovitis, 1996). But much of the historical analysis behind the basic argument of this book had its roots (in my consciousness at least) in the 300 consecutive monthly essays that I wrote from 1974 to 2001 in the popular forum of Nat­ural History magazine, where I tried to develop a distinctive style of “mini in­tellectual biography” in essay form — attempts to epitomize the key ideas of a professional career in a biographic context, and within the strictures of a few thousand words. By thus forcing myself to emphasize essentials and to discard peripherals (while always searching out the truly lovely details that best exemplify any abstraction), I think that I came to understand the ma­jor ideological contrasts between the defining features of Darwinian theory and the centerpieces of alternative views. In this format, I first studied such structuralist alternatives as Goethe's theory of the archetypal leaf, Geoffroy's hypothesis on the vertebral underpinning of all animals, and on dorsoventral inversion of arthropods and vertebrates, and Owen's uncharacteristic English support for this continental view of life. I also developed immense sympathy for the beauty and raw intellectual power of various alternatives, even if I eventually found them wanting in empirical terms. And I came to understand the partial validity, and even the moral suasion, in certain proposals unfairly ridiculed by history's later victors — as in reconsidering the great hippocam­pus debate between Huxley and Owen, and recognizing how Owen used his (ultimately false) view in the service of racial egalitarianism, while Huxley misused his (ultimately correct) interpretation in a fallacious defense of tradi­tional racial ranking.

Finally, my general love of history in the broadest sense spilled over into my empirical work as I began to explore the role of history's great theoretical theme in my empirical work as well — contingency, or the tendency of com­plex systems with substantial stochastic components, and intricate nonlinear interactions among components, to be unpredictable in principle from full knowledge of antecedent conditions, but fully explainable after time's actual unfoldings. This work led to two books on the pageant of life's history (Gould, 1989c; Gould, 1996a). Although this book, by contrast, treats  {47}  general theory and its broad results (pattern vs. pageant in the terms of this text), rather than contingency and the explanation of life's particulars, the science of contingency must ultimately be integrated with the more conventional sci­ence of general theory as explored in this book — for we shall thus attain our best possible understanding of both pattern and pageant, and their different attributes and predictabilities. The closing section of the book (pp. 13321343 of Chapter 12) offers some suggestions for these future efforts.

When I ask myself how all these disparate thoughts and items fell together into the one long argument of this book, I can only cite — and I don't know how else to put this — my love of Darwin and the power of his genius. Only he could have presented such a fecund framework of a fully consistent theory, so radical in form, so complete in logic, and so expansive in implication. No other early evolutionary thinker ever developed such a rich and comprehen­sive starting point. From this inception, I only had to explicate the full origi­nal version, tease out the central elements and commitments, and discuss the subsequent history of debate and revision for these essential features, culmi­nating in a consistent reformulation of the full corpus in a helpful way that leaves Darwin's foundation intact while constructing a larger edifice of in­terestingly different form thereupon. Clearly I do not honor Darwin by hagiography, if only because such obsequious efforts would make any honest character cringe (and would surely cause Darwin to spin in his grave, thus up­setting both the tourists in Westminster Abbey and the adjacent bones of Isaac Newton). I honor Darwin's struggles as much as his successes, and I fo­cus on his few weaknesses as entry points for needed revision — his acknowledged failure to solve the “problem of diversity,” or his special pleading for progress in the absence of any explicit rationale from the operation of his central mechanism of natural selection.

As a final comment, if this section has violated the norms of scientific discourse (at least in our contemporary world, although not in Darwin's age) by the liberty that I have taken in explicating personal motives, errors and cor­rections, at least I have shown how we all grope upward from initial stupid­ity, and how we would never be able to climb without the help and collabora­tion of innumerable colleagues, all engaged in the intensely social enterprise called modern science. I experienced no eureka moment in developing the long argument of this book. I forged the chain link by link, from initial pos­session of a few separate items that I didn't even appreciate as pieces of a sin­gle chain, or of any chain at all. I made my linkages one by one, and then of­ten cut the segments apart, in order to refashion the totality in a different order. So many people helped me along the way — from long dead antecedents by their wise words to younger colleagues by their wisecracks — that I must view this outcome as a social project, even though I, the most arrogant of lite­rati, insisted on writing every word. Perhaps I can best express my profound thanks to the members of such an intellectual collectivity by stating, in the most literal sense, that this book would not exist without their aid and sufferance. My formal dedication to my two dearest and closest paleontological collaborators in this effort to formulate macroevolutionary theory records  {48}  the worthy apex of an extensive pyramid. Scientists fight and squabble as all folks do (and I have scarcely avoided a substantial documentation thereof in this book). But we are, in general, a reasonably honorable lot, and we do em­brace a tendency to help each other because we really do revel in the under­standing of nature's facts and ways — and most of us will even trade some per­sonal acclaim for the goal of faster and firmer learning. For all the tensions and unhappinesses in any life, I can at least say, with all my heart, that I chose to work in the best of all enterprises at the best of all possible times. May our contingent future only improve this matrix for my successors.

Epitomes for a Long Development


Most of this book can be described as extensive narration of work already done, and ideas already expounded elsewhere. But no one should write at such length merely to organize the conventional material of a field, and without an original structure, or a set of unconventional ideas, to propose. I wrote The Structure of Evolutionary Theory because I felt that I had fol­lowed a sufficiently idiosyncratic procedure to devise a sufficiently novel the­oretical structure that then yielded a sufficient number of original insights on specific matters to qualify as a justification for spending so many years of a career, and daring to ask readers for such a non-trivial chunk of their at­tention.

As implied by the foregoing sentence, I think that whatever originality this work possesses might best be conceptualized at three levels of basic structure, primary justifications for the major components of theory, and specific in­sights or discoveries then developed under the aegis of this structure and the­ory. At the first level of basic structure, I believe that three features of organi­zation set the novelty of presentation:

1. Developing an exegesis of essential components in the logic of Darwin­ian theory, as expressed in the agency, efficacy, and scope of selection as an evolutionary mechanism (Chapter 2).

2. Explicating the history of evolutionary thought as a complex and ex­tended debate about these essential components, developed negatively at first by early evolutionists who sought alternative formulations to Darwinism (Chapters 36), and then positively in our times by scientists who recognized the need for extensive revisions and expansions that would build an enlarged structure upon a Darwinian foundation, rather than uproot the theoretical core of selectionism (Chapters 712).

3. Formulating an expanded theory that introduces substantial revisions on each branch of Darwinian central logic, but builds, in its ensemble, a coher­ently enlarged structure with a retained Darwinian base — moving from Dar­win's single level of agency to a hierarchical theory of selection on the first branch; balancing positive sources of internal constraint (for both structural and historical reasons) with the conventional externalism of natural selection on the second branch; and recognizing the disparate inputs of various tiers  {49}  of time, rather than trying to explain all phylogenetic mechanics by uniformitarian extrapolation from microevolutionary processes, on the third branch.

At the second level of validation for proposed revisions in the structure of evolutionary theory, I have tried to develop broad arguments and empirical justifications for major changes and expansions on each of the three branches of Darwinian central logic. On the first branch of agency, the theory of punc­tuated equilibrium itself, initially formulated by Niles Eldedge and me, estab­lishes the species as a true and potent Darwinian individual, and grants a minimal guarantee of descriptive independence to macroevolution by requir­ing a treatment of trends as the differential success of stable species rather than the adaptive anagenesis of lineages by accumulated and extrapolated organismal selection alone. Beyond punctuated equilibrium, the general ra­tionale for a hierarchical theory of selection, as presented here through the interactor approach based on emergent fitnesses at higher levels, may estab­lish a complete (and tolerably novel) framework not only for grasping the consistent logic of hierarchical selection, but also for viewing each level as po­tent in its own distinctive way, and for recognizing the totality of evolution­ary outcomes as a realized balance among these potencies, and not as the achieved optimality of a single causal locus — a substantial difference from Darwinian traditions for conceiving the dynamics of evolutionary change. In working through the differences among levels — see Chapter 8, pp. 714744 — I was particularly struck by the surprising, but accurate and challeng­ing, analogies (Lamarckian inheritance at the organismal level with adaptive anagenesis at the species level, for example); and by the different modes of equally effective change implied by disparate structural reasons for the estab­lishment of individuality at various levels (particularly, the domination of se­lection over drift and drive at the organismal level vs. the potent balance among all three mechanisms at the species level).

On the second branch of efficacy, I have tried to make the most comprehensive case yet advanced for internal constraint as a positive director and channeler of evolutionary change, and not only as a negative brake upon pure Darwinian functionalism. I proceed by explicating two conceptually different forms of constraint — structural constraints as consequences of physical prin­ciples, and historical constraints as channels from particular pasts. I argue that each category challenges a different central tenet of Darwinism — struc­tural constraint by establishing a substantial space for non-selectionist origin of important evolutionary features, and historical constraint for explaining the markedly inhomogeneous filling of morphospace as flow down ancient internal channels of deep homology, and not primarily as a mapping of adap­tive design upon current ecological landscapes. Beyond any novelty in this general formulation, I have attempted to develop a conceptual space, and to establish practical criteria, for the identification of non-adaptive sequelae (spandrels), the evolutionary importance of their later cooptation for utility (exaptation), and the importance of such reservoirs of potential (exaptive pools) in explicating the important concept of “evolvability” in structural rather than purely adaptational terms.  {50} 

On the third branch of scope, my contribution cannot claim much novelty, if only because I have not worked professionally in this area of paleontological research. But I do explicate, perhaps more fully than before, both the historical and conceptual reasons for regarding catastrophic mass extinction, and catastrophic mechanics in general (within their limited scope of validity), not as anti-selectionist per se, but rather as fracturing the extrapolationist premise of Darwinian central logic, and requiring that substantial aspects of phyletic pattern be explained as interaction between temporal extensions of microevolution and different processes that only become visible and effective at higher tiers of time. I try to resolve “the paradox of the first tier” (the em­pirical failure of Darwin's logically airtight argument for a vector of progress) by arguing that punctuated equilibrium at the second tier of phyletic trends, and mass extinction at the third tier of faunal overturn, impose enough of their own, distinctive and different, patterning to forestall the domination or pure imprint of extrapolated microevolutionary results upon the general pag­eant of life's history.

Finally, at the third level of those lovely details (where both God and the devil dwell, and where, ultimately, both the joy and power of science reside), I trust that any originality I have introduced at “higher” levels of theoretical structure gains primary expression and utility in the resolution of previously puzzling details, and in the identification of “little things” that had escaped previous notice or explicit examination.

For example, most original analyses and discoveries in the historical first half of this book flow directly from my organizing theme of identifying essen­tial components in Darwinian logic, and then tracing both the early attempts to defeat, and our later efforts to modify and expand them through time. I was thus able to discover and identify Darwin's major encounter with higher level selection not in his recognized discussion of group selection for human altruism, but in his previously unexplicated admission of species selection to resolve the problem of diversity (see Chapter 3, pp. 246250). In this case, I “lucked out” through an odd reason for previous ignorance of such an im­portant textual revision — for Darwin omitted this material in his compressed and hasty discussion of diversity in Chapter 4 of the Origin (on this subject, the only Darwinian source generally known to professional biologists, who would immediately highlight the importance of any acknowledgment of spe­cies selection). But Darwin agonized over levels of selection at explicit length in the unpublished “long version” that only saw the light of printed day in 1975 (Stauffer, 1975), and that virtually no practicing biologist has ever read (whereas historians of science who do study this longer text usually lack suf­ficient knowledge of the technical debate about levels of selection to under­stand the meaning of Darwin's passages or to appreciate their import).

The same context led me to appreciate the previously unanalyzed development of a full hierarchical model by Weismann in his later works (Chapter 3, pp. 223224), a formulation that Weismann himself identified as the most important theoretical achievement of his later career. Previous historians had written about his much longer and earlier explications of lower level selection  {51}  (germinal selection in his terms), if only in the context of modern reductionistic breakdowns of Darwinism to selection among “selfish genes.” But they had missed his later reversal and expansion to a full hierarchical model, despite Weismann's own emphasis. Similarly, de Vries's clear understanding of Darwinian logic had also been ignored because de Vries, as an opponent of the efficacy of Darwinian organismal selection (a painful decision for him, given his psychological fealty to Darwin, also explored herein), applied the logic to higher levels, and even devised the term “species selection” (Chapter 5, pp. 446451) — a concept and coining previously entirely unremarked by historians (much to the embarrassment of scientists, including yours truly, who coined and explicated the same term much later in full expectation of pristine originality!).

Similarly, my sense of the logic in conflicts between constraint and adapta­tion (or internal vs. external, or formal vs. functional approaches) on the second branch helped me to pinpoint, or to make sense of, several important historical events and arguments that have not been properly treated or under­stood. Historians of science had not previously discussed orthogenetic theo­ries in this fairest light, and had not distinguished the very different formula­tions of Hyatt, Eimer, and Whitman in terms of their increasingly greater willingness to accommodate Darwinian themes as well (see Chapter 5). The same framework allowed me to identify the crucial importance, and brilliant epitomization, of this issue in the final paragraphs of Chapter 6 (“Difficulties on Theory”) in Darwin's Origin, a significance that had not been highlighted before.

I also traced the dichotomy of anglophonic preferences for functionalist accounts vs. continental leanings towards formalism back through the evo­lutionary reconstruction of the argument in the mid 19th century into the creationist formulations of Paley vs. Agassiz (Chapter 4), thus illustrating a pedigree for this fundamental issue in morphology that evolution may have recast in causal terms, but did not budge in basic commitments to the mean­ing of morphology. Among the little tidbits that emerge from such analy­ses, I even discovered that Darwin borrowed his clearest admission of co-opted utility from non-adaptive origins (unfused skull sutures in mammalian neonates, essential for passage through the birth canal, but also existing in birds and reptiles born from more capacious eggs) from the longer and more nuanced descriptions of Richard Owen, Britain's anomalous defender of for­malism.

I also included some historical analyses in the book's second half on mod­ern advances because I thought they could make an original contribution to arguments usually developed only in contemporary terms and findings. I have already mentioned my analysis of how the initial pluralism of the Modern Synthesis (embracing any mode of change consistent with known genetic mechanisms) hardened through subsequent editions of the founding volumes into pronounced preferences for adaptationist accounts framed only in terms of natural selection (Chapter 7). In addition, I think that my reexhumation of the debate between Falconer and Darwin on fossil elephants provides a  {52}  good introduction to punctuated equilibrium (Chapter 9, pp. 745749). The largely unknown paradox of Lankester's original definition of homoplasy as a category of homology, rather than in the opposite status held by the term to­day, provides the best entry I could devise for understanding the vital, but lit­tle appreciated and rarely acknowledged, theoretical differences between par­allelism and convergence. In the absence of this context and distinction, the key importance of evo-devo and the discovery of deep homology among dis­tant phyla cannot properly be grasped as a challenge and expansion of Dar­winian expectations (Chapter 10).

I hope that my sympathetic portrayal of D'Arcy Thompson's theory of form (Chapter 11), despite my general disagreement with his argument, will help colleagues to understand the thrust and potential power of this unusual formulation of structuralist constraint on external grounds of universal phys­ics. Although I am chagrined that I discovered Nietzsche's account of the dis­tinction between current utility and historical origin so late in my work, I know no better introduction — from one of history's greatest philosophers to boot, and in his analysis of morality, not of any scientific subject — to the the­oretical importance of spandrels and exaptation in the rebalancing of con­straint and adaptation within evolutionary theory (Chapter 11, pp. 12141218). In a final historical analysis of the second part, I think that Dar­win's own rationale for progress (Chapter 12, pp. 12961303), rooted not in the mechanics of natural selection itself, but in an ecological argument for extrapolation of biotic competition through time in a perpetually crowded world — an aspect of Darwin's thought that has very rarely been appreciated, formulated or discussed by historians — provided the best context I could de­vise for understanding why catastrophic mass extinction in particular, and non-extrapolation through tiers of time in general, play such havoc with Dar­win's need for uniformity on the third branch of his essential logic.

The original claims in the book's second half on modern reformulations of evolutionary theory rest, necessarily and primarily, on theoretical insights and unusual conceptual parsings, rather than on novel data — if only be­cause custom dictates that my extensive empirical documentation be pre­sented in “review” format by collating published studies in support or refuta­tion of general themes under discussion. But I have sometimes presented existing data in novel contexts — as in my analysis of the proper category for understanding the exaptive value of genes lost by founder drift in establish­ing the social cohesion (albeit transient) that has made the Argentine ant Linepithema humile such a successful invader of non-native Californian habi­tats (Chapter 11, pp. 12821284). I have also cited my own empirical studies, previously published but original in the more conventional sense, to support important pieces of more general arguments, including validation of punctu­ated equilibrium by dissection of a single bedding plane to reveal transition by absolute age dating of individual shells (Goodfriend and Gould, 1996), the “employment” of constraint by selection to yield several adaptive features by one heterochronic change in a case of neoteny in Gryphaea (Jones and Gould, 1999), and the explanation of most ordered geographic variation within  {53}  a major subregion of Cerion as consequences of allometric correlations in growth (Gould, 1984b).

I tried (and utterly failed) to compose a selective listing, as provided above for the book's historical half, for original ideas about theoretical details devel­oped in revising the three branches of Darwinian central logic in the book's second half on modern reformulations of evolutionary theory. I ripped up several attempts that read like the hodge-podge of a random laundry list rather than the ordered “sweet places” on a logical continuum. These high­lights, I finally recognized, have little meaning outside the broader context of a linearly developing argument for each branch, and I will therefore make a second attempt, within the more detailed epitome of the next and final sec­tion of this chapter, to designate the points that struck me with the force of “aha,” or that conveyed a hint of deeper, surprising, or more radical implica­tions for reasons that I couldn't quite fathom directly, but that tickled my in­tuition at the edge of that wonderful, if elongate, German word: Finger — spitzengefuhl, or feeling at the tip of one's finger. Most inchoate excitements of this sort lead to nowhere but foolishness and waste of time, but every once in a while, the following of one's nose catches a whiff of novelty. At least we must trust ourselves enough to try — and not take ourselves so seriously that we forget to laugh at our more frequent and inevitable stumbles.


I have insisted, borrowing Darwin's famous line in my arrogance, that this “whole volume is one long argument,” flowing logically and sequentially from a clear beginning in Darwin's Origin to our current reformulations of evolutionary theory. But this structural thread of Ariadne can easily become lost in the labyrinth of my tendencies to expatiate on little factual gems, or to follow the thoughts of leading scientists into small, if lovely, byways of their mental complexities. Hence, I need to present summaries and epitomes as guidelines.

Long books, like large bureaucracies, can easily get bogged down in a ba­roque layering of summary within summary. The United States House of Representatives has a Committee on Committees (I kid you not), undoubt­edly embellished with subcommittees thereof. And we must not forget Jona­than Swift's famous verse on the fractality of growing triviality in scholarly commentary:

So, naturalists observe, a flea

Hath smaller fleas that on him prey;

And these have smaller still to bite 'em

And so proceed ad infinitum.

Thus every poet, in his kind,

Is bit by him that comes behind.

I wrote, on page 13, that this book includes three levels of embedding for this long argument — the summary in this chapter, the epitome of Darwin in  {54}  Chapter 2, and the development of the totality. Now, and most sheepishly, I add two more, for a fractal total of five — the listed abstract, in pure “book order,” of this section, and (God help us) the epitome of this epitome, pre­sented now to introduce and guide the list.

I develop my argument throughout this book by asserting, first, that the central logic of Darwinism can be depicted as a branching tree with three ma­jor limbs devoted to selection's agency, efficacy and scope. Second, that Dar­win, despite his heroic and explicit efforts, could not fully “cash out” his the­ory in terms of the stated commitments on each branch — and that he had to allow crucial exceptions, or at least express substantial fears, in each domain (admitting species selection to resolve the problem of diversity; permitting an uncomfortably large role for formalist correlations of growth as compromis­ers of strict adaptationism; expressing worry that mass extinction, if more than an artifact of an imperfect fossil record, would derail the extrapolationist premise of his system). Third, that the subsequent history of evolution­ary debate has focused so strongly upon the key claims of these three essential branches that we may use engagement with them as a primary criterion for distinguishing the central from the secondary when we need to gauge the im­portance of challenges to the Darwinian consensus. Fourth, that we should not be surprised by the prominence of these three themes, for they embody (in their biological specificity) the broadest underlying issues in scientific expla­nation, and in the nature of change and history: levels of structure and causal­ity, rates of alteration, directions of causal flow, the possibility of causal uni­fication by reduction to the lowest level vs. autonomy and interaction of irreducible levels, punctuational vs. gradual change, causal and temporal tier­ing vs. smooth extrapolation. Fifth, that the most interesting and impor­tant debates in our contemporary science continue to engage the same three themes, thus requiring the vista of history to appreciate the continuity and logical ordering that extends right back to Darwinian beginnings. Sixth, that our best modern understanding of the structure of evolutionary theory has re­versed the harmful dichotomization of earlier debates (Darwinian fealty vs. destructive attempts to trivialize or overturn the mechanism of selection) by confronting the same inadequacies of strict Darwinism, but this time intro­ducing important additions and revised formulations that preserve the Dar­winian foundation, but build a theory of substantial expansion and novelty upon a retained selectionist core.

This logic and development may be defended as tolerably impersonal and universal, but any book of this length and complexity, and of so idiosyncratic a style and structure, must also own its authorial singularities. The Structure of Evolutionary Theory emerges, first of all, from my professional focus as a paleontologist and a student of macroevolution, defined, as explained on page 38, as descriptive phenomenology prior to any decision about the need for distinctive theory (my view) or the possibility of full subsumption un­der microevolutionary principles (the view of Darwin and the Modern Syn­thesis). The contingency of history guarantees that any body of theory will underdetermine important details, and even general flows, in the realized  {55}  pageant of life's phylogeny on Earth — and such a claim for nontheoretical independence of macroevolution generates no dispute, even between rigid reductionists who grant no separate theoretical space to macroevolution, and biologists, like myself, who envisage an important role for distinctive macroevolutionary theory within an expanded and reformulated Darwinian view of life.

In his description of the reductionist view of classical Darwinism — his own opinion in positive support, not a simplistic caricature in opposition — Hoffman (1989, p. 39) writes: “The neodarwinian paradigm therefore asserts that this history of life at all levels — including and even beyond the level of speciation and species extinction events, embracing all macroevolutionary phenomena — is fully accounted for by the processes that operate within pop­ulations and species.” I dedicate my book to refuting this traditional claim, and to advocating a helpful role for an independent set of macroevolutionary principles that expand, reformulate, operate in harmony with, or (at most) work orthogonally as additions to, the extrapolated, and persistently relevant (but not exclusive, or even dominant) forces of Darwinian microevolution.

This perspective of synergy confutes the contrary, and ultimately destruc­tive, attempts by late 19th and early 20th century macroevolutionists to de­velop substitute mechanisms that would disprove or trivialize Darwinism, and that spread such a pall of suspicion over the important search for non-reductionistic and expansive evolutionary theories — a most unfortunate (if historically understandable) trend that stifled, for several generations, the unification and fruitful expansion of evolutionary theory to all levels and temporal tiers of biology. Thus, for example, my attempt to develop a specia-tional theory of macroevolution (Chapters 8 and 9), with species treated as irreducible Darwinian individuals playing causal roles analogous to those oc­cupied by organisms in Darwinian microevolution, represents an extension of Darwinian styles of explanation to another hierarchical level of analysis (with interestingly different causal twists and resulting patterns), not a refuta­tion of natural selection from an alien realm. (Such a speciational theory, however, does counter Hoffman's reductionistic claim of full theoretical suf­ficiency for “processes that operate within populations and species” — for, given the stasis of species under punctuated equilibrium, such macroevolu­tionary patterns originate by higher-order sorting among stable species, and not primarily by processes occurring anagenetically within the lifetime of these higher-level Darwinian individuals.) Similarly, the different rules of cat­astrophic mass extinctions require additions to the extrapolated Darwinian and microevolutionary causes of phyletic patterns, but do not refute or deny the relevance of conventional uniformitarian accretions through geological time. (In fact, a more comprehensive theory that seeks to integrate the rela­tive strengths, and interestingly disparate effects, of such different levels and forms of continuationist vs. catastrophic causality offers greater richness to Darwinian perspectives as both underpinnings and important contributors to a larger totality.)

A second authorial input must arise from the distinctive ontogeny of past  {56}  work. The Structure of Evolutionary Theory occupies a much broader terri­tory than my first lengthy technical book of an earlier career, Ontogeny and Phylogeny (1977b). The motivating conceit of the first book rested upon my choice of a much smaller compass defined by a much clearer tradition of definition and research. I thought — thus my designation of this strategy as a conceit — that I could quote, in extenso and from original sources, every im­portant statement, from von Baer and before to de Beer and after, on the rela­tionship between development and evolution. This potential for comprehen­siveness brought me much pleasure and operational motivation.

In fact, I soon realized that I could not succeed, even within this limited sphere — and I therefore punted shamelessly in the final result. I did manage to quote every important passage on the theoretical relationship between these central subjects of biology, but I passed, nearly completely, on the actual use of these putative relationships in specific proposals for phylogenetic recon­structions. And, as all historians of science and practitioners of evolutionary biology know, this genre of “phylogenizing” represented by far (at least by weight) the dominant expression of this theoretical rubric in the technical lit­erature. I would, by the way, defend my decision as entirely reasonable and proper, and not merely as practically necessary, because these specific phylo­genetic invocations made effectively no contribution to the development of evolutionary theory — my central concern in the book — and remained both speculative and transient to boot. But I do remember the humbling experi­ence of realizing that a truly full coverage could only represent a pipe dream, if applied to any important subject in a vigorous domain of research!

My personal love of such thoroughness (with the necessary trade-off of limitation in domain) posed a substantial problem when I decided to expand my range from ontogeny and phylogeny to the structure of evolutionary the­ory. Of all genres in scholarship, I stand most strongly out of personal sympa­thy with broad-brush views that attempt to encompass entire fields (the his­tory of philosophy from Plato to Pogo, or of transportation from Noah to NASA) in a breathless summary paragraph for each of many thousand inci­dents. Even the most honorable efforts by great scholars — former Librarian of Congress Daniel Boorstin's The Explorers, for instance — make me cringe for simplistic legends repeated and interesting complexities omitted. At some level, truly important and subtle themes can only be misrepresented by such a strategy.

But how then to treat the structure of evolutionary theory in a reputable, even an enlightening, way? Surely we cannot abandon all hope for writing honorably about such broad subjects simply because the genre of comprehen­sive listing by executive summary must propagate more mythology and mis­information than intrigue or understanding. As a personal solution to this crucial scholarly dilemma, and in developing the distinctive strategy of this book, I employed a device that I learned by doing, through many years of composing essays — a genre that I pursued by writing comprehensive personal treatments of small details, fully documentable in the space available, but  {57}  also conveying important and general principles in their cascading implica­tions. I vowed that I would try to encompass the structure of evolutionary theory in its proper intellectual richness, but that I would do so by exhaustive treatment of well chosen exemplifying details, not by rapid summaries of in­adequate bits and pieces catalogued for all relevant participants.

Under this premise, the central task then evolves (if I may use such a metaphor) into an extended exercise in discrimination. The solution may be la­beled as elitist, but how else can selection in intellectual history be under­taken? One must choose the best and the brightest, the movers and shakers by the sieve of history's harsh judgment (and not by the transiency of immedi­ate popularity) — and let their subtle and detailed formulations stand as a se­ries of episodes, each conveyed by an essay of adequate coverage. Luckily, the history of evolutionary thought — as one of the truly thrilling and expansive subjects of our mental lives — has attracted some of the most brilliant and fas­cinating doers and thinkers of intellectual history. Thus, we are blessed with more than adequate material to light the pathway of this particular odyssey in science. Luckily too, the founding figure of Darwin himself established such a clear basis of brave commitment that I could characterize, and then trace down to our own times, an essential logic that has defined and directed one of the most important and wide-ranging debates in the history of science into a coherent structure, ripe for treatment by my favored method of full coverage for the few truly central items (by knowing them through their fruits and logics, and by leaving less important, if gaudy, swatches gently aside in order to devote adequate attention to essential threads).

A third, and final, authorial distinction — my treatment of history and my integration of the history of science with contemporary research on evolu­tionary theory — emerges directly from this strategy of coverage in depth for a small subset of essential items and episodes. My historical treatments tend to resolve themselves into a set of mini intellectual biographies (as exemplified and defended on page 46) for almost all the central players in the history of Darwinian traditions in evolutionary thought. I can only hope that this pecu­liar kind of intellectual comprehensiveness will strike some readers as enlight­ening for the “quick entree” thus provided into the essential work of the peo­ple who led, and the concepts that defined, the history of the greatest and most consequential revolution in the history of biological science. (In most cases — a Goethe, Cuvier, Weismann, de Vries, Fisher or Simpson, for exam­ple — I chose people for their intrinsic and transcendent excellence. In fewer instances — an Eimer or Hyatt as proponents of orthogenesis, for example — I selected eminently worthy scientists not as great general thinkers, but as best exponents of a distinctive approach to an important subject in the history of debate on essentials of evolutionary theory.)

A few figures in history have been so prescient in their principal contribu­tion, and so acute and broad-ranging in their general perceptions, that they define (or at least intrude upon) almost any major piece of a comprehensive discussion (A. N. Whitehead famously remarked, for example, that all philosophy  {58}  might be regarded as a footnote to Plato). Evolutionary biology pos­sesses the great good fortune to embrace such a figure — Charles Darwin, of course — at the center of its origin and subsequent history. Thus, Darwin emerges again and again, often controlling the logic of discussion, through­out this book — in his own full foundational exegesis (Chapter 2); but then, in later chapters, as the principal subject, and best possible exemplification, of other important subbranches on all three boughs of his essential logic (his re­luctant acceptance of higher levels of selection in Chapter 3; his formalist contrast to his own functionalism in stressing “correlations of growth” in Chapter 4; his views on direction and progress in the history of life in Chapter 6, and, even in the book's second half on modern developments, for his dis­cussion of discordance between historical origin and current utility as a point of departure for my treatment of exaptation in Chapter 11, and his attempt to underplay and undermine mass extinction as an introduction to my cri­tique of uniformitarianism and extrapolationism in the final Chapter 12). Who could ask for a more attractive and effective coordinating “device” to tie the disparate strands of such an otherwise disorderly enterprise together than the genial and brilliant persona of the man who first gave real substance to the grandeur in this view of life?

Whatever my dubiety about the role and efficacy of abstracts (too often, as we would all admit in honest moments, our only contact with a work that we nonetheless then feel free to criticize in full assurance of our rectitude), I can­not deny that a work of this length, imbued moreover with a tendency to pen­etrate byways along a basic route that seems (at least to this author) ade­quately linear and logical, demands some attempt to list its principal claims in textual order. Hence, I now impose upon you the following abstract:

Chapter 2: An exegesis of the origin

1. All major pre-Darwinian evolutionary theories, Lamarck's in particu­lar, contrasted a primary force of linear progress with a distinctly secondary and disturbing force of adaptation that drew lineages off a main line into par­ticular and specialized relationships with immediate environments. In his most radical intellectual move, expressing both the transforming depth and the conceptual originality of the theory of natural selection, Darwin denied the existence of a primary progressive force, while promoting the lateral force of adaptation to near exclusivity. In so privileging uniformitarian extrapola­tion as an explanatory device, Darwin imbued natural selection, the lateral force, with sufficient power to generate evolutionary change at all scales by accumulating tiny adaptive increments through the immensity of geological time.

2. The Origin of Species exceeds all other scientific “classics” of past centu­ries in immediate and continued relevance to the basic theoretical formula­tions and debates of current practitioners. Careful exegesis of Darwin's logic and intentions, through textual analysis of the Origin, therefore assumes un­usual importance for the contemporary practice of science (not to mention its undeniable historical value in se).  {59} 

3. Darwin famously characterized the Origin as “one long argument” without explicitly stating “for what?” Assumptions about the focus of this long argument have ranged from the restrictively narrow (for natural selec­tion, or even for evolution) to the overly broad (for application of the most general hypotheticodeductive model in scientific argument, as Ghiselin has claimed). I take a middle position and characterize the “long argument” as an attempt to establish a methodological approach and intellectual foundation for rigorous analysis in historical science — a foundation that could then be used to validate evolution.

4. The “long argument” for historical science operates at two poles — methodological and theoretical. The methodological pole includes a set of procedures for making strong inferences about phyletic history from data of an imperfect record that cannot, in any case, “see” past causes directly, but can only draw conclusions from preserved results of these causes. Darwin devel­ops four general procedures, all based on one of the three essential premises of his theory's central logic: the explanation of large-scale results by extrapo­lation from short-term processes. In order of decreasing information avail­able for making the required inference, these four procedures include: (1) ex­trapolation to longer times and effects of evolutionary changes actually observed in historic times (usually by analogy to domestication and horticul­ture); (2) exemplification and ordering of several phenomena as sequential stages of a single historical process (fringing reefs, barrier reefs and atolls as stages in the formation of coral reefs by subsidence of central islands, for ex­ample); (3) inference of history as the only conceivable coordinating explana­tion for a large set of otherwise disparate observations (consilience); and (4) inference of history from single objects based on quirks, oddities and imper­fections that must denote pathways of prior change.

5.  The theoretical pole rests upon the three essential components of Darwinian logic: (1) agency, or organismal struggle as the appropriate (and nearly exclusive) level of operation for natural selection; (2) efficacy, or natu­ral selection as the creative force of evolutionary change (with complexly co­ordinated sequelae of inferred principles about the nature of variation, and of commitments to gradualism and adaptationism as foci of evolutionary analy­sis); and (3) scope, or extrapolationism (as described in point 4 just above). The logical coordination of these commitments, and their establishment as a brilliantly coherent and intellectually radical theory of evolution, can best be understood by recognizing that Darwin transferred the paradoxical argument of Adam Smith's economics into biology (best organization for the general polity arising as a side consequence of permitting individuals to struggle for Aemselves alone) in order to devise a mechanism — natural selection — that would acknowledge Paley's phenomenology (the good design of organisms said harmony of ecosystems), while inverting its causal basis in the most radical of all conceivable ways (explaining the central phenomenon of adaptation by historical evolution rather than by immediate creation, and recognizing nature's sensible order as a side consequence of unfettered struggle among individuals, rather than a sign of divine intent and benevolence).  {60} 

6. The first theme of agency: Darwin's commitment to the organismal level as the effectively exclusive locus of natural selection occupies a more central, and truly defining, role than most historians and evolutionists have recog­nized. Invocation of this most reductionistic locus then available (in igno­rance of the mechanism of inheritance) embodies the intellectual radicalism of Darwin's theory — using Adam Smith to overturn Paley, and holding that all higher-order harmony, previously attributed to divine intention, arises only as a side-consequence of selfish “struggle” for personal advantage at the lowest organismal level. Darwin devoted far more of the Origin to defending this organismal locus than most exegetes have acknowledged, particularly in centering his only two chapters on specific difficulties in natural selection (7 on Instinct and 8 on Hybridism) to resolutions provided by insistence upon organismal agency — explaining the establishment of adaptive sterile castes in social insects by selection upon queens as individuals, and resolving sterility in interspecific crosses as an unselected sequel of differences accumulated by organismal selection in each of two isolated populations, rather than as a di­rect result of higher-level species selection, as Wallace affirmed and as Darwin strove mightily and consciously to avoid. We can also trace his struggle to af­firm organismal exclusivity in his reluctances, underplayings and walling off (as unique and unrepeated elsewhere in nature) of the one exception (for hu­man altruism) that the logic of his system forced upon his preferences.

7. For his defense of the second theme of efficacy — his assertion of natural selection as the only potent source of creative evolutionary change — Darwin recognized that his weak and negative force, although surely a vera causa (true cause), could only play this creative role if variation met three crucial re­quirements: copious in extent, small in range of departure from the mean, and isotropic (or undirected towards adaptive needs of the organism). I would argue that Darwin's most brilliant intellectual move lay in his accurate identification, through the logical needs of his theory and not from any actual knowledge of heredity's mechanism, of these three major attributes of varia­tion — because he recognized that natural selection could not otherwise oper­ate as a creative force in the evolution of novelties.

8.  Gradualism enters Darwin's system as another deductive intellectual consequence of asserting that natural selection acts as the creative mechanism of evolutionary change. Gradualism has three distinct meanings in Darwinian traditions, with only the second (or intermediate) statement relevant to the central assertion of selection's creativity. First, gradualism as simple historical continuity of stuff or information underlies the basic factuality of evolution vs. creation, and does not validate any particular mechanism of evolutionary change. Second, gradualism as insensible intermediacy of transitional forms specifies the Goldilockean “middle position” required by the mechanism of natural selection to refute the possibility that saltational variation might en­gender creative change all at once, thus relegating selection to a negative role of removing the unfit. Third, gradualism as a geological claim for slowness and smoothness (but not constancy) of rate plays a crucial role in the third theme (see point 10 of this list) of selection's  {61}  scope, or the extrapolatability of microevolution to explain all patterns in geological time — and is therefore the aspect of gradualism that punctuated equilibrium refutes (for punctuated equilibrium questions Darwin's uniformitarian and continuationist beliefs, but not his mechanism of natural selection). This parsing of three distinctly different forms of gradualism, all embraced by Darwin for different rea­sons, alleviates the misunderstanding behind some unfortunate terminologi­cal wrangles without substance that have generated much heat (but little light) in recent debates.

9. The adaptationist program as a primary strategy of research emerges as the third major implication of advocating natural selection as the primary creative force in evolutionary change — for this Darwinian style of evolution must proceed step by step, with each tiny increment of change rendering or­ganisms better adapted to alterations in local environments. To summarize all the key implications of this second theme of efficacy, the creativity of natural selection makes adaptation central, isotropy of variation necessary, and grad­ualism pervasive.

10.  Restriction of agency to the organismal level, and assertions of selec­tion's creativity, set a biological basis for the third essential claim of Dar­winian logic — selection's scope, or the argument that this incremental and gradualistic style of microevolution can, by smooth extrapolation through the immensity of geological time, build the full extent of life's anatomical change and taxonomic diversity by simple accumulation. I focus my shorter discussion of this third essential theme not upon biological needs (already covered in the first two themes), but upon the requirement for similar grad­ualistic styles of change in the geological stage that must present the evolu­tionary play — particularly in Darwin's embrace of Lyellian uniformity, and his denial of catastrophism (through arguments about the imperfection of the fossil record to allay the literal appearance of such rapidity in geological data), for even a fully consistent, intellectually sound, and operationally po­tent theory will not regulate actual events if surrounding conditions debar its operation.

11. I use Kellogg's brilliant approach to the evaluation of Darwinian theory (published in 1907 in anticipation of centennial celebrations for Darwin's birth and the sesquicentenary of the Origin) to distinguish alternatives that deny the fundamental postulate of selection's creativity from auxiliaries that enlarge, adumbrate, or reformulate the theory of natural selection in basically helpful and consistent ways. I show that Darwinism may be epitomized by its three essential claims of agency, efficacy, and scope — and that the history of debate has always centered upon these themes, with critiques focusing upon destructive alternatives or constructive auxiliaries. I argue, as the major thesis of this book, that modern debates have developed important and coherent auxiliary critiques on all three branches of essential Darwinian logic, and that these debates may lead to a fundamentally revised evolutionary theory with a retained Darwinian core.

Chapter 3: Seeds of hierarchy

1. Nearly all scientific revolutions originate as replacements and refuta­tions of  {62}  previous explanatory schemes, not as pure additions to a former state of acknowledged ignorance. Lamarck's evolutionary theory, known to anglophonic readers as a first full account through the fair but critical descriptions of Lyell (in Volume 2, 1832, of the Principles of Geology), and from Chambers's promotion in the Vestiges of 1844, provided a context for Darwin's ref­utation. Darwin's single-level theory, based on the full efficacy of locally adaptive changes at the smallest scale, countered the only available alter­native of Lamarckism by relocating the major phenomenon that generated change and required explanation (local adaptation for Darwin, general prog­ress for Lamarck), and (far more radically) by reversing the conventional Paleyan explanation for the good design of organisms and the harmony of ecosystems (direct divine construction at the highest level vs. sequelae of nat­ural selection working at the lowest level of organismal advantage).

2. Lamarck, a dedicated materialist with a two-factor theory of evolution as a contrast between linear progress up life's ladder and tangential deflec­tions of diversity through local adaptation, has been widely misunderstood (and reviled), both in Darwin's time and today, as a vitalist and pure expo­nent of “soft” or Lamarckian inheritance (which he accepted as the “folk wisdom” of his day, and invoked primarily to explain the secondary process of lateral adaptation).

3. Darwin's theory of natural selection shared a functionalist basis with Lamarck in joint emphasis upon adaptation to external environment as the instigator of evolutionary change. But the two theories differ most radically in Darwin's citation of a single locus and mechanism of change — with the full range of evolutionary results proceeding by natural selection for local adapta­tion of populations to changing immediate environments, and all higher-level phenomenology emerging by sequential accumulation of such tiny incre­ments through the immensity of geological time. By contrast, Lamarck advo­cated a two-factor theory, with local adaptation as a merely secondary and diverging process (and, as we all know of course, arising by soft inheritance of acquired features generated by adaptive effort during an organism's life, rather than by natural selection of fortuitous variation), set against a primary process of progressive complexification up the ladder of life. Thus, Darwin embraced Lamarck's secondary force (instantiated by a different mechanism), denied the existence of Lamarck's primary force, and argued that the second­ary force of local adaptation also produced the large-scale results attributed by Lamarck to the primary force. Thus, this first major debate between evolu­tionary alternatives contrasted Lamarck's hierarchical theory with Darwin's single-level account. Hierarchy has been an important issue from the start (al­though, obviously, modern versions of hierarchical selection theory, advo­cated as the centerpiece of this book, bear no relationship, either genealogical or ideological, to this false, but fascinating, Lamarckian original).

4. Darwin explicitly rejected Lamarck's two-factor theory, correctly identifying the disabling paradox that rendered the theory nonoperational: “what is important cannot be observed or manipulated (the higher-level force of progress), and what can be observed and manipulated (the tangential force of local adaptation) cannot explain the most important phenomenon (progress  {63}  in complexification).” Darwin developed the first testable and operational theory of evolution by locating all causality in the palpable mechanism of natural selection.

5.  In the first generation of Darwinian debate, August Weismann, clearly the most brilliant theorist of his time, and the only biologist (besides Darwin) who fully grasped the logic and implications of selection, wrestled with levels of selection throughout his career, and along an interesting path, finally devel­oping a full hierarchical theory that he explicitly identified as the most impor­tant conclusion of his later work. He began by trying to refute Lamarckian inheritance (and Herbert Spencer's vigorous defense thereof) by advocating the Allmacht (omnipotence, or literally “all might” or complete sufficiency) of natural selection. He first attributed the degeneration of previously useful structures (a bigger problem for Darwinism than the explanation of adaptive features) to what he called “panmixia” (not the modern meaning of the term, but the effect of recombination, in sexual reproduction, between adaptive ele­ments and inadaptive elements no longer subject to negative selection); then realized that this process could not explain complete elimination, thus lead­ing him to propose a lower level of subcellular selection, potentially acting in opposition to organismal selection, and called “germinal selection”; and finally recognized that if levels of selection existed below the organismal, then the same logic implies the existence and potency of supraorganismal levels as well.

6. Darwin himself provides the best 19th century example — previously unrecognized because Darwin omitted this material, originally written for the unpublished “long version,” from the Origin — of the need for a hierarchical theory of selection in any full account of the phenomenology of evolution. Entirely consistent single-level theories cannot be carried through to comple­tion. Darwin admitted important components of species selection in capping his (still unsatisfactory) explanation for an issue that he ranked second in im­portance only to explaining the anagenesis of populations by natural selec­tion: the resolution of organic variety and plenitude by a “principle of diver­gence” (his terminology). I document the largely unrecognized emphasis that he placed upon this principle of divergence (for example, the Origin's famous single figure does not illustrate natural selection, as generally misinterpreted, but rather the principle of divergence). Darwin struggled to explain this de­scriptively higher-level phenomenon of taxonomic diversification as a fully predictable consequence of ordinary organismal selection, but he could not proceed beyond an argument that he himself finally recognized as forced, and even a bit hokey: the claim that natural selection will always favor extreme variants at the tails of a distribution for a local population in a particular ecology (the Origin's diagram represents an exemplification of this claim). Eventually, Darwin realized that he needed to invoke species selection for a fell explanation of the success of speciose clades — and this unknown argument, rather than his well-documented defense of group selection for human altruism, represents Darwin's most generalized invocation of selection at supraorganismal levels.  {64} 

7. Hierarchical models of evolutionary processes (at least descriptively so, but causally as well) have been featured and defended by evolutionary theo­rists from the beginning of our science, although not always by good or valid arguments. This inadequately recognized theme explains the major contrast between Lamarck and Darwin, and coordinates the various disputes between Wallace and Darwin. Wallace simply didn't grasp the concept of levels at all, and remained so committed to adaptationism that he ranged up and down the hierarchy, oblivious of the conceptual problems thus entailed, until he found a level to justify his adaptationist bent. Darwin, by contrast, com­pletely understood the problem of levels, and the reasons behind his strong preference for a reductionist and single-level theory of organismal agency — although he reluctantly admitted a need for species selection to resolve the problem of divergence. We can also understand why Wallace's 1858 Ternate paper, sent to Darwin and precipitating the “delicate arrangement,” did not proceed as far to a resolution as later tradition holds, when we recognize Wallace's conceptual confusion about levels of selection.

Chapter 4: Intemalism and laws of form: pre-darwinian alternatives

1. In a brilliant closing section to his general chapter 6, entitled “difficulties on theory,” Darwin summarized the logical structure of the most important challenge to his system, and organized his most cogent defense for his func­tionalist theory of selection, by explicating the classical dichotomy between “unity of type” and “conditions of existence” — or the formalism of Geoffroy vs. the functionalism of Cuvier — entirely in selectionist terms, and to his ad­vantage. He attributed “conditions of existence” to immediate adaptation by natural selection, and then explicated “unity of type” as constraints of inheri­tance of homologous structures, originally evolved as adaptations in a distant ancestor. Thus, he identified natural selection as the underlying “higher law” for explaining all morphology as present adaptation or as constraint based on past adaptation. He also admitted, while cleverly restricting their range and frequency, a few other factors and forces in evolutionary explanation.

2.  A fascinating, and previously unexplored, contrast may be drawn be­tween the strikingly similar dichotomy, although rooted in creationist ex­planations, of Paley's functionalist and adaptationist theory of divine con­struction for individualized biomechanical optimality vs. Agassiz's formalist theory of divine ordination of taxonomic structure as an incarnation of God's thoughts according to “laws of form” reflecting modes and categories of eter­nal thought. Clearly, this ancient (and still continuing) contrast between structural and functional conceptions of morphology transcends and pre­dates any particular mechanism, even the supposedly primary contrast of cre­ation vs. evolution, proposed to explain the actual construction of organic di­versity.

3. In the late 18th century, the great poet (and naturalist) Goethe developed a fascinating (and, in the light of modern discoveries in evo-devo, more than partly correct) archetypal theory in the structuralist or formalist mode — and explicitly critical of functionalist, teleological and adaptationist alternatives  {65}  — for the diversity of organs growing off the stems and roots of plants. He viewed cotyledons, and all the standard parts of flowers (sepals, petals, stamens and carpels), as modifications of a leaf archetype.

4.  The famous early 19th century argument, culminating in the public debate of 1830 between Georges Cuvier and Etienne Geoffroy St. Hilaire (and analyzed by Goethe in his final paper before his death), did not, as com­monly misinterpreted, pit evolutionary theories against creationist accounts (although Geoffroy favored a limited theory of evolution, while Cuvier re­mained strongly opposed), but rather represented the most striking and en­during incident in this older and persistent struggle between formalist (Geoffroy) and functionalist (Cuvier) explanations of morphology and taxonomic order. Geoffroy advocated the abstract vertebra as an archetype for all animals, beginning (largely successfully) with a common basis for anatomical differences between teleosts and tetrapods, moving to the putatively common design of insects and vertebrates (still with some success, partly confirmed by the Hoxology of modern evo-devo, but also including some “howlers” like the homology of arthropod limbs with vertebrate ribs), and crashing with the proposed homology of vertebrates and a cephalopod doubled back upon it­self (the comparison that sufficiently aroused Cuvier's growing ire into a call for public debate). Geoffroy's theory of dorsoventral inversion between insects and vertebrates was not a silly evolutionary conjecture about “the worm that turned” (as later caricatures often portray), and did not represent an evolutionary explanation at all, but rather expressed a formalist comparison based upon a common underlying structure, ecologically oriented one way in vertebrates (central nervous system up), and the other way in arthropods. The common impression of Cuvier's victory must be reassessed as a complex “draw,” with Geoffroy's position abetted by the fortuity of his longer life and his courting of prominent literary friends as supporters (including Balzac and Georges Sand).

5.  Adaptationist preferences have enjoyed a long anglophonic tradition, beginning with the treatises of Ray and Boyle, in Newton's founding genera­tion, on final causes; then extending, in creationist terms, through Paley and the Bridgewater Treatises; and finally culminating in the radically reversed evolutionary explanations (but still retaining the same functionalist and adaptationist commitments) of Darwin, extending forward to Fisher and the Modern Synthesis. By contrast, continental traditions have favored formalist and structuralist explanations of morphology, from the creationist accounts of Agassiz, through the transitional systems of Goethe and Geoffroy, to the fully evolutionary accounts of Goldschmidt and Schindewolf in the mid 20th century. Interestingly, the complex views of Richard Owen, so widely misun­derstood as an opponent of evolution (when he only rejected the predomi­nant functionalism of traditional British approaches to morphology), may best be grasped when we understand him as a rare anglophonic exponent of a predominantly formalist theory. Owen, following Geoffroy, tried to explain the entire vertebrate skeleton, including the skull and limbs, as a set of modi­fications upon a vertebral archetype.

6. Darwin maintained a genuine interest in formalist constraints upon  {66}  adaptationist optimality for individualized features of anatomy — a theme that he epitomized as “correlations of growth.” But he developed an explicit framework and rationale, most thoroughly discussed not in the Origin but in his longest 1868 book on The Variation of Animals and Plants Under Do­mestication, that relegated such formalist effects to a clearly subservient and secondary status, compared with natural selection and adaptation, in evolu­tionary causality.

Chapter 5: Channels and saltations in post-Darwinian formalism

1. Galton's Polyhedron, the metaphor and model devised by Darwin's bril­liant and eccentric cousin Francis Galton, and then fruitfully used by many evolutionary critics of Darwinism, including St. George Mivart, W. K. Brooks, Hugo de Vries, and Richard Goldschmidt, clearly expresses the two great, and both logically and historically conjoined, themes of formalist (or structuralist, or internalist, in other terminologies) challenges to functionalist (or adaptationist, or externalist) theories in the Darwinian tradition. This model of evolution by facet-flipping to limited possibilities of adjacent planes in inherited structure stresses the two themes — channels set by internal con­straint, and evolutionary transition by discontinuous saltation — that struc­turalist alternatives tend to embrace and that pure Darwinism must combat as challenges to basic components of its essential logic (for channels direct the pathways of evolutionary change from the inside, albeit in potentially posi­tive and adaptive ways, even though some external force, like natural selection, may be required as an initiating impulse; whereas saltational change violates the Darwinian requirement for selection's creativity by vesting the scope and direction of change in the nature and magnitude of internal jumps, and not in sequences of adaptive accumulations mediated by natural selection at each step).

2.  Orthogenesis, as a general term for evolutionary directionality along channels of internal constraint, rather than external pathways of natural se­lection, existed in several versions, ranging from helpful auxiliaries to Dar­winism, to outright alternatives that denied any creative potency to selection. Theodor Eimer, who coined the term orthogenesis, presented a middle ver­sion that tried to integrate internal channels of orthogenesis with external pathways of functionalist determination. But Eimer defended Lamarckian mechanics for his functionalism, thus leading him to oppose natural selection (he spoke of the Ohnmacht, or “without power,” of selection, contrasted with Weismann's Allmacht, or “all power”) despite his pluralistic linkage of formalist and functionalist explanations.

3.  The orthogenetic theory of the late 19th century American paleontolo­gist Alpheus Hyatt embodied maximal opposition to natural selection, and must be viewed as alternative, rather than auxiliary, to Darwinism. Hyatt conceived the pathway of ontogeny, modified only by heterochronic changes permitted under the biogenetic law, as the internal directing channel that nat­ural selection could tweak, but not derail. Illustrating the influence of theory over perception, Hyatt found several parallel lineages of snails, running along  {67}  different segments of a common pathway, but all supposedly living in an identical environment — where others had reconstructed typical Darwinian monophyletic trees of phylogeny from the same stratigraphic section of fresh­water planorbids. Hyatt, who engaged in a long and ultimately frustrating correspondence with Darwin on this subject, believed that lineages followed a preordained “ontogeny” of phyletic youth, maturity and old age, thus at­tributing the different internal responses of lineages living in the same envi­ronment to their residence in different stages of an ontogenetically fixed and shared phyletic pathway (a preset internal channel with a vengeance).

4. Charles Otis Whitman, a great early 20th century American naturalist, developed the most congenial auxiliary theory (to Darwinism) of ortho­genesis in his extensive work on the evolution of color patterns in Darwin's own favorite organism, the domestic pigeon. Whitman argued that domestic pigeons in particular, and dove-like birds in general, followed a strong chan­nel of internal predisposition leading in one direction from checkers to bars, and eventually to the obliteration of all color. (Darwin, by interesting con­trast, argued for a reverse tendency from bars to checkers, but also held, as his basic theory obviously implies, that selection largely determines any par­ticular event and that no internal predisposition can trump the dictates of im­mediate function.)

5. In his 1894 book on Materials for the Study of Variation (where he coined the term homeosis), William Bateson presented an extensive catalog of cases in discontinuous variation among individuals in a population and be­tween populations of closely related organisms. He used these examples to develop a formalist theory of saltational evolution, strongly opposed to the adaptationist assumptions of Darwinian accounts. (Bateson's acerbic criti­cisms of adaptationist scenario-building and story-telling in the speculative mode emphasize a common linkage between structuralist preferences for me­chanical explanation, and distaste for the adaptationist assumption that func­tional necessity leads and the evolution of form follows.) Although Bateson coined the term genetics, his personal commitment to a “vibratory” theory of heredity, based on physical laws of classical mechanics — an intuition that he could never “cash out” as a testable theory — prevented his allegiance to the growing influence of Mendelian principles.

6. Hugo de Vries, the brilliant Dutch botanist who understood the logic of selectionism so thoroughly and acutely (but largely in contrast with the only other biologists, Weismann and Darwin himself, who also grasped all the richness and range of implications, but with favor), developed a saltational theory of evolution, but explicitly denied any predisposition of lineages to follow internal channels of constraint. (He thus showed the potential inde­pendence of the frequently linked formalist themes of channeling and salta­tion, a conjunction espoused by Bateson and Goldschmidt for example, but denied in the other direction by Whitman, who favored channeling but de­nied saltation by supporting a gradualist theory of orthogenetic change.) This fascinating scholar regarded Darwin as his intellectual hero and never forgot the kindness and encouragement conveyed by his mentor and guru during  {68}  their one personal meeting early in de Vries's career. But de Vries, who devel­oped the theory of intracellular pangenesis (the ultimate source for the term “gene”) in the late 19th century, and then (quite fortuitously and long after he had reached saltational conclusions for other reasons) became one of Men­del's rediscoverers, based his truly saltational theory of immediate macromutational origin of species on his work with the evening primrose, Oenothera lamarckiana, where he mistook an odd chromosomal organization that gen­erates occasional saltations for a biological generality. De Vries, who under­stood the logic of selectionism so well, who knew that his macromutational theory refuted several essential components of Darwinian logic, but who could not bear (for largely psychological reasons) to forsake his intellectual and personal hero, insisted upon his larger fealty to Darwin, even though he had banned Darwinian mechanisms from the master's own realm of the ori­gin of species. So de Vries developed a hierarchical theory that, while denying selection for the origin of species, restored selectionist logic at the higher level of phyletic trends by explicitly proposing “species selection” (his term) as a mechanism for generating broader phylogenetic patterns.

7. By proposing a comprehensive formalist theory in the heyday of developing Darwinian orthodoxy, Richard Goldschmidt became the whipping boy of the Modern Synthesis — and for entirely understandable reasons. Gold­schmidt showed his grasp, and his keen ability to utilize, microevolutionary theory by supporting this approach and philosophy in his work on variation and intraspecific evolution within the gypsy moth, Lymantria dispar. But he then expressed his apostasy by advocating discontinuity of causality, and pro­posing a largely nonselectionist and formalist account for macroevolution from the origin of species to higher levels of phyletic pattern. Goldschmidt in­tegrated both themes of saltation (in his concept of “systemic mutation” based on his increasingly lonely, and ultimately indefensible, battle to deny the corpuscular gene) and channeling (in his more famous, if ridiculed, idea of “hopeful monsters,” or macromutants channeled along viable lines set by internal pathways of ontogeny, sexual differences, etc.). The developmental theme of the “hopeful monster” (despite its inappropriate name, virtually guaranteed to inspire ridicule and opposition), based on the important con­cept of “rate genes,” came first in Goldschmidt's thought, and always occu­pied more of his attention and research. Unfortunately, he bound this inter­esting challenge from development, a partially valid concept that could have been incorporated into a Darwinian framework as an auxiliary hypothesis (and now has been accepted, to a large extent, if under different names), to his truly oppositional and ultimately incorrect theory of systemic mutation, therefore winning anathema for his entire system. Goldschmidt may have acted as the architect of his own undoing, but much of his work should evoke sympathetic attention today.

Chapter 6: Pattern and progress on the geological stage

1. Darwin based his argument for a broad and general vector of progress in life's history not on the “bare bones” operation of natural selection (where he  {69}  had explicitly denied such an outcome as the most radical implication of his theory), but on subsidiary ecological claims for the predominance of biotic over abiotic competition, and for a geological history of plenitude in a persis­tently crowded ecological world, where one species must displace another to gain entry into ecosystems (the metaphor of the wedge). Darwin used these ecological sequelae, along with the gradualist and incrementalist logic of natural selection itself, as primary justifications for his third essential claim of selection's scope, or the uniformitarian extension of small-scale microevolution, in a smoothly continuationist manner, to explain all patterns of macro-evolution by accumulation of increments through the immensity of geological time.

2.  Such a claim requires that the geological stage operate in an appropri­ate, and “Goldilockean,” manner — not too much change to debar the opera­tion and domination of this slowly and smoothly accumulative biological mode, and not too little to provide insufficient impetus (within Darwin's externalist and functionalist theory) for attributing the amount of change ac­tually observed to natural selection.

3.  The primary claim of “too much” derived from the school of “catastrophism” in geology — a movement that has been unfairly stigmatized by later history, following Lyell's successful and largely rhetorical mischaracterization (he was a lawyer by profession), as an unscientific defense of super-naturalism to cram the observed results of geology into the strictures of bibli­cal chronology, but that actually took the opposite position of strict empirical literalism (whereas uniformitarians argued that the numerous literal appear­ances of rapidity in the geological record must be “interpreted” as misleading consequences of how gradual change must be expressed in a woefully imper­fect set of strata). The great catastrophist Cuvier, in particular, was an En­lightenment rationalist, not a theological apologist — and he based his defense of catastrophism upon his literalist reading of the paleontological and geolog­ical record.

4.  The primary claim of “too little” geology followed Lord Kelvin's increasingly diminished estimates for the age of the earth (incorrectly made — although Kelvin accurately described the necessary, but (as it turned out) empirically false, logic required to validate his views — by assuming that heat now flowing from the earth represented a continuing loss from an originally molten state). Darwin worried intensely over Kelvin's claims, even referring to him as an “odious spectre” in a letter to Wallace. Darwin feared that Kel­vin's low estimates would not permit enough time to generate the history of fife under his slowly acting theory of gradualistic and accumulative change. Although this story has been told often, and has become familiar to scientists, an important (and decisive) aspect of the tale has rarely been exposed: Darwin fought this battle alone, and his strong distress illustrates the maximal, and unique, extent of his gradualistic and continuationist commitments. His closest colleagues, Wallace and Huxley, did not find Kelvin's low estimates unacceptable, but argued that we had only been led to expect such slow change from our previous conception of the earth's age, and that faster rates  {70}  of phyletic change, implied by Kelvin's dates, were entirely acceptable under their reading of evolution.

Chapter 7: The modern synthesis as a limited consensus

1.  From the anarchic situation that prevailed at the Darwinian centennial celebrations of 1909 (confidence in the factuality of evolution, linked with agnosticism about theories and mechanics, as the first fruits of Mendelism seemed, initially, to refute the gradualism and incrementalism of natural se­lection), the Modern Synthesis eventually emerged in two stages (following the union of Darwinian and Mendelian perspectives in the work of Fisher and others): first, by a welcome restriction that eliminated Kellogg's three al­ternatives in oppositional modes that would have destroyed Darwinism (Lamarckism as a substitute functionalism, and saltationism and orthogene­sis as formalist alternatives), and reasserted, now in a context of Mendelian particulate inheritance, the adequacy of natural selection as a creative force; and second, by an increasingly dubious hardening, culminating in centennial celebrations for the Origin in 1959, that substituted an increasingly rigid adaptationism for an earlier pluralism that embraced all mechanisms (includ­ing genetic drift) consistent with known genetic principles, while favoring se­lection as a primary force.

2.  In his founding book of 1930, The Genetical Theory of Natural Selec­tion, R. A. Fisher showed how slow, gradualist evolution in large, panmictic populations (treated almost as an ahistorical system, analogous to effectively infinite populations of identical gas molecules free to move and diffuse by physical principles) could validate strict Darwinism under Mendelian partic­ulate inheritance (with Darwin's own acceptance of blending inheritance ex­posed as a more serious impediment than Darwin himself had realized), and disprove saltational alternatives by the inverse correlation of frequency and magnitude in variation. To these mathematical and general chapters, Fisher appended a long closing section devoted to his eugenical theory that Western society had begun to degenerate seriously as a consequence of the social pro­motion of infertility (the rise in class level of “good” genetic stock, largely by their correlated tendency to have fewer children, thereby husbanding their economic resources to potentiate their social elevation). Fisher conceived this eugenical “blight” as entirely Darwinian in character — invisible in its gradual expression generation by generation, but ultimately more deadly than the ex­plicit saltational degenerations stressed by most eugenicists.

3.  In contrast with the initial pluralism of Haldane and Huxley (in the book that coined the Modern Synthesis), and of the first editions of founding documents for the second phase of the Synthesis (Dobzhansky's 1937 Genet­ics and the Origin of Species, Mayr's 1942 Systematics and the Origin of Spe­cies, and Simpson's 1944 Tempo and Mode in Evolution), later editions of these three documents encapsulated the hardening of this second phase, as initial pluralism yielded to an increasingly firm and exclusive commitment to adaptationist scenarios, and to natural selection as a virtually exclusive mech­anism of change. Even Sewall Wright's views on genetic drift and shifting  {71}  balance altered from initial stress upon stochastic alternatives to selection to an auxiliary role for drift (as an impetus for the exploration of new, and poten­tially higher, adaptive peaks) as one aspect of a more inclusive and basically adaptationist process. The complex reasons for this hardening include some empirical documentations of selection, but also involve a set of basically so­cial and institutional factors not based on increasing factual adequacy.

4. If this hardening on the second Darwinian branch of selection's efficacy reflects a general trend within evolutionary theory, then we should find a sim­ilar Darwinian strengthening (and narrowing) on the other two branches of selection's agency (organismal vs. higher levels) and scope (adequacy to ex­plain the entire geological record by extrapolated microevolution). The tri­umph (for good reasons at the time) of Williams over Wynne-Edwards af­firms this trend for agency, although Williams's important clarification then unfortunately hardened (among epigones) into a dogmatic and a priori re­jection of any hint of group selection. Similarly, the Synthesis's increasing confidence in the exclusivity of gradualistic microevolution deprived paleon­tology of any independent theoretical space, and relegated the field to docu­mentation of an admittedly underdetermined pageant, built by the exclusive agency of microevolutionary principles. Several synthesists even denied the efficacy of differential speciation as an input to macroevolutionary pattern (branding the speciosity of some clades as a “luxury” rather than a crucial in­put to survival and flourishing), and attributed all higher-level change to extensions of gradualistic and adaptive anagenesis within unbranched lineages.

5.  The trends to development, initial pluralism and later hardening of the Modern Synthesis win clearest expression in two sources of data: comparison of statements by leading scientists at the two contrasting centennial cele­brations of 1909 and 1959 (for Darwin's birth and for the publication of the Origin); and by documentation of hardening in the summary statements (and increasingly dogmatic dismissal of alternatives) in leading textbooks for sec­ondary and undergraduate courses in biology.

Chapter 8: Species as individuals in the hierarchical theory of selection

1. Selectionist mechanics, in the most abstract and general formulation, work by interaction of individuals and environments (broadly construed to include all biotic and abiotic elements), such that some individuals secure dif­ferential reproductive success as a consequence of higher fitness conferred by some of their distinctive features, leading to differential plurifaction of indi­viduals with these features (relative to other individuals with contrasting fea­tures), thus gradually transforming the population in adaptive ways. But the logic of this statement implies that organisms cannot be the only bio­logical entities that manifest the requisite properties of Darwinian individual­ity — properties that include both vernacular criteria (definite birth and death points, sufficient stability during a lifetime, to distinguish true entities from unboundable segments of continua), and more specifically Darwinian criteria (production of daughters, and inheritance of parental traits by daughters). In  {72}  particular, and by these criteria, species must be construed not only as classes (as traditionally conceived), but also as distinct historical entities acting as good Darwinian individuals — and therefore potentially subject to selection. In fact, a full genealogical hierarchy of inclusion — with rising levels of genes, cell lineages, organisms, demes, species and clades — features clearly definable Darwinian individuals, subject to processes of selection, at each level, thus validating (in logic and theory, but not necessarily in the potency of actual practice in nature) an extension and reformulation of Darwin's exclusively organismal theory into a fully hierarchical theory of selection.

2.  The validity of the “interactor approach” to defining the mechanics of selection, and the fallacy of the “replicator approach” expose, as logically in­valid, all modern attempts to preserve Darwinian exclusivity of level, but to offer an even more reductionistic account in terms of genes, rather than or­ganisms, as agents — with organisms construed as passive containers for the genes that operate as exclusive agents of natural selection. This false argu­ment, based upon the true but irrelevant identification of genes as faithful replicators, must be replaced by the conceptually opposite formulation of a hierarchical theory of selection, with genes identified as only one valid, and lowest, level in a hierarchy of equally potent, and interestingly different, lev­els of Darwinian individuality: genes, cell lineages, organisms, demes, species and clades. Replication identifies a valid and important criterion for the cru­cial task of bookkeeping or tracing evolutionary change; but replicators can­not specify the causality of selectionist processes, which must be based upon the recognition and definition of interactors with environments. Even Wil­liams and Dawkins, the two leading exponents of exclusive gene selectionism, have acknowledged and properly described the hierarchical causality of inter­action (while proferring increasingly elaborate and implausible verbal de­fenses of gene selection in arguments about parallel hierarchies and Necker cubing of legitimate alternatives rooted in criteria of replication vs. interac­tion). Thus, Williams and Dawkins seem to grasp the validity of hierarchical selection through a glass darkly, while still trying explicitly to defend their in­creasingly indefensible preferences for exclusive gene selectionism.

3.  The logic of hierarchical selection cannot be gainsaid, and even Fisher admitted the consistency, even the theoretical necessity, while denying the empirical potency, of species selection. Fisher based his interesting and powerful argument on his assumption that low N for species in clades (relative to or­ganisms in populations) must debar any efficacy for species selection in a world of continuous and gradualistic anagenesis rooted in organismal selec­tion. However, Fisher's argument, although logically tight, fails empirically because species tend to be stable and directionally unchanging (however fluctuating) during their geological lifetimes, and the theoretically “weaker” force of species selection may therefore operate as the “only game in town” for macroevolution. The arguments for potency of species selection are stron­ger than corresponding assertions for interdemic selection (largely because species actively maintain their boundaries as Darwinian individuals, whereas demes remain subject to breakup and invasion). But, despite these intrinsic  {73}  weaknesses and problems, interdemic selection has now been empirically val­idated as an important force in evolution — thus strengthening a prima facie case for the even greater importance of species selection in macroevolution.

4. Two theoretical resolutions and clarifications have established both a sound theoretical basis, and a strong argument for the empirical potency, of species selection as an important component of macroevolution: first, the rec­ognition of differential proliferation rather than downward effect as the most operational criterion for defining and recognizing species selection; second, the acknowledgment that emergent fitnesses under the interactor approach, rather than emergent features treated as active adaptations of the species, constitute the proper criterion for identifying species selection. The former in­sistence upon emergent features (by me and other researchers, and in error), while logically sound and properly identifying a small subset of best and most interesting cases, relegated the subject to infrequent operational utility, and thus to relative impotence. The proper criterion (under the interactor ap­proach) of emergent fitness universalizes the subject by permitting general identification in the immediacy of the current mechanics of selection, and not requiring knowledge — often unavailable given the limits of historical ar­chives — of adaptive construction and utility in ancestral states.

5. The six levels recognized for convenience, and not accompanied by any claim of completion or exclusivity — gene, cell lineage, organism, deme, spe­cies and clade — feature two important principles that make the theory of hi­erarchical selection so different from, while still in the lineage and tradition of, exclusivistic Darwinian organismal selection. First, adjacent levels may in­teract in the full range of conceivable ways — in synergy, orthogonally, or in opposition. Opposition has been stressed in the existing literature, but only because this mode is easier to recognize, and not for any argument of greater importance in principle. Second, the levels operate non-fractally, with fas­cinating and distinguishing differences in mode of functioning, and relative importance of components, for each level. For example, the different mecha­nisms by which organisms and species maintain their equally strong individu­ality dictate that selection should dominate at the organismal level, while selection, drift, and drive should all play important and balanced roles at the species level.

6. To cite just one difference (from conventions of the organismal level) for each nonstandard level, and to make the key point about distinctiveness of levels in an almost anecdotal manner: random change may be most promi­nent in relative frequency at the level of the gene-individual; true gene selec­tion also plays an important, if limited, role (largely in the mode that has been given the unfortunate name — for its implication of opposition, almost in ethical terms, to the supposed standard of proper organismal selection — of “selfish DNA”); however, the Dawkinsian argument for exclusivity of genic selection only records the confusion of a preferred level of bookkeeping with an erroneous claim for a privileged locus of selection. Selection among cell-lineages, although ancestrally important in the evolution of multicellular organisms, has largely been suppressed by the organismal level in the interests  {74}  of its own integrity; failure of this suppression leads to the pyrrhic victory of cell-lineages that we call cancer. Interdemic selection, although once so widely rejected, probably plays an essential role in the evolution of social coopera­tion in general, and not only for such specific phenomena as human altruism. Species-level selection, combined with other species-level properties of drive and drift, establishes the independent basis for a distinctive speciational the­ory and reformulation of macroevolution. The highest level of clade selec­tion, although sometimes operative, may be relatively weak by an extension of Fisher's argument about low N.

7. I explore the distinctive differences between levels of selection by trying to exemplify and “play out” the detailed disparities in a “grand analogy” be­tween the conventional operation of organismic selection and the relative conceptual novelty of species selection. As an idiosyncratic sample of poten­tial reforms and surprises, consider the following claims: First, the formula­tion of a general taxonomy for sources of change in hierarchically ordered systems, based on a primary distinction of “drive” for directed changes aris­ing within an individual, based on change among lower-level individuals as constituent parts; and “sorting,” with two causally distinct subcategories of “selection” and “drift” for change based on alterations of relative frequencies among individuals at the focal level itself. Second, the recognition, by follow­ing the logic of the analogy, of some strikingly counterintuitive comparisons that become both interesting and revealing upon subsequent reflection — in­cluding the likeness of Lamarckian change, construed as ontogenetic drive at the organismal level, with standard anagenetic transformation as organismal drive at the species level (transformation by directional change of constituent parts of a higher-level individual, in this case the organisms of a species); this similarity may also highlight the rather different reasons for general unimpor­tance of both levels of drive — Lamarckism for the well-known reason of the­oretical non-occurrence in a Mendelian world, and anagenesis based on the controversial claim for its evident plausibility in theory (as a basic Darwinian process), but rarity in fact, given the dominant relative frequency of punctu­ated equilibrium. Third, the establishment of a framework for distinguishing directional speciation as a form of reproductive drive (inherently biased dif­ferences in autapomorphies of descendant species vs. ancestral states) from true species selection as a higher order sorting among daughter species that arise with phenotypic differences randomly distributed about parental means. I believe that we have missed this crucial distinction because the ana­log of directional speciation at the organismal level — drives induced by muta­tion pressure — occur so rarely (for conventional reasons of organismal selec­tion's power to suppress them) that we haven't considered the greater potency of analogous processes at other levels. Fourth, the importance of testing “Wright's Rule” — the claim that speciation is random with respect to the di­rection of evolutionary trends within clades — because the major alternative of directional speciation as the cause of trends holds such potential power at the species level, whereas its analog (drives of mutation pressure) assumes so little importance at the organismal level. Fifth, the potentially far greater  {75}  importance of drift (both species drift and founder drift) vs. selection as a mech­anism of sorting at the species level, but not at the organismal level, where se­lection predominates in standard formulations. Sixth, the identification of an intrinsically, and probably unbreakable (in most cases), negative correlation between speciation and extinction propensities as the primary constraint op­erating to prevent the takeover of life by a few megaclades (which might dom­inate by enhancing speciation while retarding extinction among constituent species — or perhaps the Coleoptera have prevailed by this means). Seventh, the recognition that the organismal level operates uniquely in securing the in­tegrity of its individuals by devices (physiological homeostasis among organs, and spatial bounding by an external surface) that “clear out” both drive from below and drift at its own level as mechanisms operating at high relative fre­quency — thus leaving selection in its most dominant position at this level. Perhaps our Darwinian prejudice for regarding selection as by far the most effective, or virtually the only important, process of evolutionary change arises more from the parochialism of our organismal focus (given our own personal residence in this category) than from any universal characterization of all levels in evolution.

Chapter 9: Punctuated equilibrium and the validation of macroevolutionary theory

1. The clear predominance of an empirical pattern of stasis and abrupt geological appearance as the history of most fossil species has always been acknowledged by paleontologists, and remains the standard testimony (as documented herein) of the best specialists in nearly every taxonomic group. In Darwinian traditions, this pattern has been attributed to imperfections of the geological record that impose this false signal upon the norm of a truly gradualistic history. Darwin's argument may work in principle for punctuational origin, but stasis is data and cannot be so encompassed.

2. This traditional argument from imperfection has stymied the study of evolution by paleontologists because the record's primary (and operational) signal has been dismissed as misleading, or as “no data.” Punctuated equilib­rium, while not denying imperfection, regards this signal as a basically accu­rate record of evolution's standard mode at the level of the origin of species. In particular, before the formulation of punctuated equilibrium, stasis had been read as an embarrassing indication of absence of evidence for the de­sired subject of study — that is, of data for evolution itself, falsely defined as gradual change — and this eminently testable, fully operational, and intellec­tually fascinating (and positive) subject of stasis had never been subjected to quantitative empirical study, a situation that has changed dramatically during the last 25 years.

3. The key empirical ingredients of punctuated equilibrium — punctuation, stasis, and their relative frequencies — can be made testable and defined operationally. The theory only refers to the origin and development of species in geological time, and must not be misconstrued (as so often done) as a claim for true saltation at a lower organismal level, or for catastrophic mass extinction  {76}  at a higher faunal level. Punctuation must be scaled relative to the later duration of species in stasis, and we suggest 1-2 percent (analogous to human gestation vs. the length of human life) as an upper bound. Punctuated equi­librium can be distinguished from other causes of rapid change (including anagenetic passage through bottlenecks and the traditional claim of imperfect preservation for a truly gradualistic event) by the criterion of ancestral sur­vival following the branching of a descendant. Punctuations can be revealed by positive evidence (rather than inferred from compression on a single bed­ding plane) in admittedly rare situations, but not so infrequent in absolute number, of unusual fineness of stratigraphic resolution or ability to date the individual specimens of a single bedding plane. Stasis is not defined as abso­lute phenotypic immobility, but as fluctuation of means through time at a magnitude not statistically broader than the range of geographic variation among modern populations of similar species, and not directional in any pre­ferred way, especially not towards the phenotype of descendants. Punctuated equilibrium will be validated, as all such theories in natural history must be (including natural selection itself), by predominant relative frequency, not by exclusivity. Gradualism certainly can and does occur, but at very low relative frequencies when all species of a fauna are tabulated, and when we overcome our conventional bias for studying only the small percentage of species quali­tatively recognized beforehand as having changed through time.

4. Punctuated equilibrium emerges as the expected scaling of ordinary allopatric speciation into geological time, and does not suggest or imply radi­cally different evolutionary mechanisms at the level of the origin of species. (Other proposed mechanisms of speciation, including most sympatric modes, envision rates of speciation even faster than conventional allopatry, and are therefore even more consistent with punctuated equilibrium.) The theoreti­cally radical features of punctuated equilibrium flow from its proposals for macroevolution, with species treated as higher-level Darwinian individuals analogous to organisms in microevolution.

5. The difficulty of defining species in the fossil record does not threaten the validity of punctuated equilibrium for several reasons. First, in the few studies with adequate data for genetic and experimental resolution, paleo­species (even for such difficult and morphologically labile species as colonial cheilostome bryozoans) have been documented as excellent surrogates, com­parable as units to conventional biospecies. Second, the potential underesti­mation of biospecies by paleospecies only imposes a bias that makes punctu­ated equilibrium harder to recognize. The fossil record's strongly positive signal for punctuated equilibrium, in the light of this bias, only increases the probability of the pattern's importance and high relative frequency. Third, the potential overestimation of biospecies by paleospecies is probably false in any case, and also of little practical concern because no paleontologist would as­sert punctuated equilibrium from the evidence of oversplit taxa in faunal lists, but only from direct biometric study of stasis and punctuation in actual data.

6. We originally, and probably wrongly, tried to validate punctuated equilibrium by asserting that, in principle, most evolutionary change should be  {77}  concentrated at events of speciation themselves. Subsequent work in evolu­tionary biology has not confirmed any a priori preference for concentration in such episodes. Futuyama's incisive macroevolutionary argument — that re­alized change will not become geologically stabilized and conserved unless such change can be “tied up” in the unalienable individuality of a new spe­cies — offers a far richer, far more interesting, and theoretically justified ra­tionale for correlating episodes of evolutionary change with speciation.

7. Section III presents a wide-ranging discussion of why proposed empirical refutations of punctuated equilibrium either do not hold in fact, or do not bear the logical weight claimed in their presentation. Refutations for single cases are often valid, but do not challenge the general hypothesis because we anticipate a low relative frequency for gradualism, and these cases may reside in this minor category. Claims for predominant gradualism in the entire clade of planktonic forams may hold as exceptional (although, even here, the ma­jority of lineages remain unstudied, in large part because they seem, at least subjectively, to remain in stasis, and have therefore not attracted the attention of traditional researchers, who wish to study evolution, but then equate evo­lution with gradualism). However, in these asexual forms with vast popula­tions, gradualism at this level may just represent the expected higher-level expression of punctuational clone selection, as Lenski has affirmed in the most thorough study of evolution in a modern bacterial species — and just as gradual cladal trends in multicellular lineages emerge as the expected conse­quences of sequential punctuated equilibrium at the species level (trends as stairsteps rather than inclined planes, so to speak). Claims for genetic gradu­alism do not challenge punctuated equilibrium, and may well be anticipated as the proper expression at the genic level (especially given the high relative frequency of random nucleotide substitutions) of morphological stasis in the phenotypic history of species. Punctuated equilibrium has done well in tests of conformity with general models, particularly in the conclusion that exten­sive polytomy in cladistic models may arise not only (as usually interpreted) from insufficient data to resolve a sequence of close dichotomies, but also as the expectation of punctuated equilibrium for successive branching of daugh­ter species from an unchanged parental form in stasis. In fact, the frequency of polytomy vs. dichotomy may be used as a test for the relative frequency of punctuated equilibrium in well resolved cladograms — a test well passed in data presented by Wagner and Erwin.

8. Section IV then summarizes the data on empirical affirmations of punctuated equilibrium, first on documented patterns of stasis in unbranched lineages; second on punctuational cladogenesis affirmed by the criterion of ancestral survival; third on predominant relative frequencies for punctuated equilibrium in entire biotas (with particularly impressive affirmations by Hallam, Kelley, and Stanley and Yang for mollusks; and by Prothero and Heaton for Oligocene Big Badlands mammals, where a study of all taxa yielded 177 species that followed the expectations of punctuated equilibrium and three cases of potential gradualism, only one significant); fourth on predominant relative frequencies for punctuated equilibrium in entire clades, with emphasis  {78}  on Vrba's antelopes and, especially, Cheetham's rigorously quantitative and multivariate data of evolution in the bryozoan genus Metrarabdotos, perhaps the best documented and most impressive case of exclusive punctu­ated equilibrium ever developed. Finally, we can learn much from variation in relative frequencies among taxa, times, and environments — and interesting inferences have been drawn from recorded differences, particularly in Shel­don's counterintuitive linkage of stasis to rapidly changing, and gradualism to stable, environments.

9. Among many reasons proposed to explain the predominance of stasis, a phenomenon not even acknowledged as a “real” and positive aspect of evo­lution before punctuated equilibrium gave it some appropriate theoretical space, habitat tracking (favored by Eldredge), constraints imposed by the nature of subdivided populations (favored by Lieberman), and normalizing clade selection (proposed by Williams) represent the most novel and interest­ing proposals.

10. Among the implications of a predominantly punctuational origin of stable species-individuals for macroevolutionary theory, we must rethink trends (the primary phenomenon of macroevolution, at least in terms of dedi­cated discussion in existing literature) as products of the differential success of certain kinds of species, rather than as the adaptive anagenesis of lin­eages — a radical reformulation with consequences extending to a new set of explanations no longer rooted (as in all traditional resolutions) in the adap­tive advantages conferred upon organisms, but potentially vested in such structural principles as sequelae (by hitchhiking or as spandrels) of fortuitous phenotypic linkage to higher speciation rates of certain taxa. In further exten­sions, macroevolution itself must be reconfigured in speciational terms, with attendant implications for a wide range of phenomena, including Cope's rule (structurally ordained biases of speciation away from a lower size limit occu­pied by founding members of the clade, rather than adaptive anagenesis to­wards organismal benefits of large size), living fossils (members of clades with persistently minimal rates of speciation, and therefore no capacity for ever generating much change in a speciational scheme, rather than forms that are either depauperate of variation, or have occupied morphological optima for untold ages), and reinterpretation of cladal trends long misinterpreted as tri­umphs of progressive evolution (and now reevaluated in terms of variational range in species numbers, rather than vectors of mean morphology across all species at any time — leading, for example, to a recognition that modern horses represent the single surviving twig of a once luxurious, and now de­pleted, clade, and not the apex of a continually progressing trend). By the same argument, generalized to all of life, we understand the stability and con­tinued domination of bacteria as the outstanding feature of life's history, with the much vaunted progress of complexity towards mammalian elegance rein­terpreted as a limited drift of a minor component of diversity into the only open space of complexity's theoretical distribution. But, to encompass this re­formulation, we need to focus upon the diversity and variation among life's species, not upon the supposed vectors of its central tendencies, or even its  {79}  peripheral superiorities. Hominid evolution must also be rethought as reduction of diversity to a single species of admittedly spectacular (but perhaps quite transient) current success. In addition, the last 50,000 years or more of hu­man phenotypic stability becomes a theoretical expectation under punctuated equilibrium, and not the anomaly so often envisaged (and attributed to the suppression of natural selection by cultural evolution) both by the lay public and by many professionals as well.

11. Further extensions of punctuated equilibrium include the controversial phenomenon of “coordinated stasis,” or the proposition that entire faunas, and not merely their component species, tend to remain surprisingly stable in composition over durations far longer than any model based on independent behavior of species (even under punctuated equilibrium) would allow, al­though other researchers attribute the same results to extended consequences of sudden external pulses and resulting faunal turnovers, while still others deny the empirics of coordination and continue to view species as more inde­pendent, one from the other, even in the classical faunas (like the Devonian Hamilton Group) that serve as “types” for coordinated stasis.

12. Punctuated equilibrium has inspired several attempts, of varying suc­cess in my limited judgment, to construct mathematical models (or to simu­late its central phenomena in simple computer systems of evolving “artificial life”) that may help us to identify the degree of generality in modes of change that this particular biological system, at this particular level of speciation, exemplifies and records. Punctuated equilibrium has also proved its utility in extension by meaningful analogy (based on common underlying principles of change) to the generation of punctuational hypotheses at other levels, and for other kinds of phenomena, where similar gradualistic biases had prevailed and had stymied new approaches to research. These extensions range from phyletic and ecological examples below the species level to interesting ana­logs of both stasis and punctuation above the species level. Non-trending, the analog of stasis in large clades, for example, had been previously disre­garded — following the same fate as stasis in species — as a boring manifesta­tion of non-evolution, but has now been recognized and documented as a real and fascinating phenomenon in itself. Punctuational analogs have proven their utility for understanding the differential pace of morphological innova­tion within large clades, and for resolving a variety of punctuational phenom­ena in ecological systems, including such issues of the immediate moment as rates of change in benthic faunas (previously the province of hypotheses about glacially slow and steady change in constantly depauperate environ­ments), and such questions of broadest geological scale as the newly recognized stepped and punctuational “morphology” (correcting the hypothetical growth through substantial time of all previous gradualistic accounts) of mutual biomechanical improvement in competing clades involved in “arms шее,” and generating a pattern known as “escalation.”

13. Punctuational models have also been useful, even innovative in break­ing conceptual logjams, in nonbiological fields ranging from closely cognate studies of the history of human tools (including extended stasis in the Homo  {80}  erectus toolkit), and nontrending, despite classical (and false) claims to the contrary by both experts, the Abbe Breuil and Andre Leroi-Gourhan, for the 25,000 year history of elegance in parietal cave art of France and Spain — and extending into more distant fields like learning theory (plateaus and innova­tive punctuations), studies of the dynamics of human organizations, patterns of human history, and the evolution of technologies, including a fascinating account of the history of books, through punctuations of the clay tablet, the scroll, the codex, and our current electronic reformation (wherever it may lead), and long periods of morphological stasis (graced with such vital inno­vations as printing, imposed upon the unaltered phenotype of the codex, or standard “book”).

14. In a long and final section, I indulge myself, and perhaps provide some useful primary source material for future historians of scientific conflicts, by recording the plethora of non-scientific citations, ranging from the absurd to the insightful, for punctuated equilibrium (including creationist misuses and their politically effective exposure by scientists in courtroom trials that de­feated creationist legislative initiatives; and the treatment of punctuated equi­librium, often very good but sometimes very bad, by journalists and by au­thors of textbooks — the primary arenas of vernacular passage). I also trace and repudiate the “dark side” of non-scientific reactions by professional col­leagues who emoted at challenges to their comfort, rather than reacting criti­cally and sharply (as most others did, and as discussed extensively in the main body of the chapter) to the interesting novelty, accompanied by some promi­nent errors of inevitable and initial groping on our part, spawned by the basic hypothesis and cascading implications of punctuated equilibrium.

Chapter 10: The integration of constraint and adaptation: historical constraint and the evolution of development

1. Although the directing of evolutionary change by forces other than natu­ral selection has loosely been described as “constraint,” the term, even while acknowledged as a domain for exceptions to standard Darwinian mecha­nisms, has almost always been conceived as a “negative” force or phenome­non, a mode of preventing (through lack of variation, for example) a popula­tion's attainment of greater adaptation. But constraint, both in our science (and in vernacular English as well), also has strongly positive meanings in two quite different senses: first, or empirically, as channeled directionality for reasons of past history (conserved as homology) or physical principles; and second, or conceptually, as an nonstandard force (therefore interesting ipso facto) acting differently from what orthodoxy would predict.

2. The classical and most familiar category of internal channeling (the first, or empirical, citation of constraint as a positive theme) resides in preferred di­rections for evolutionary change supplied by inherited allometries and their phylogenetic potentiation by heterochrony. As “place holders” for an exten­sive literature, I present two examples from my own work: first, the illustra­tion of synergy with natural selection (to exemplify the positive, rather than oppositional, meaning), where an inherited internal channel builds two  {81}  important adaptations by means of one heterochronic alteration, as neoteny in descendant Gryphaea species of the English Jurassic produces shells of both markedly increased size (by retention of juvenile growth rates over an un­changed lifetime) and stabilized shape to prevent foundering in muddy envi­ronments (achieved by “bringing forward” the proportions of attached juve­niles into the unattached stage of adult ontogeny); second, an illustration of pervasiveness and equal (or greater) power than selective forces (to exemplify the strength and high relative frequency of such positive influences), as geo­graphic variation of the type species, Cerion uva, on Aruba, Bonaire, and Curasao, a subject of intense quantitative study and disagreement in the past, becomes resolved in multivariate terms, with clear distinction between local adaptive differences and the pervasive general pattern of an extensive suite of automatic sequelae, generated by nonadaptive variation in the geometry of coiling a continuous tube, under definite allometric regularities for the genus, around an axis.

3. For the second, or conceptually positive, meaning of constraint as a term for nonstandard causes of evolutionary change, I present a model that com­pares the conventional outcomes of direct natural selection, leading to local adaptation, with two sources that can also yield adaptive results, but for rea­sons of channeling by internal constraints rather than by direct construc­tion under external forces of natural selection. In this triangular model for aptive structures, the functional vertex represents features conventionally built by natural selection for current utilities. At the historical vertex, cur­rently aptive features probably originated for conventionally adaptive rea­sons in distant ancestors; but these features are now developmentally chan­neled as homologies that constrain and positively direct both patterns of immediate change and the inhomogeneous occupation of morphospace (espe­cially as indicated by “deep homologies” of retained developmental patterns among phyla that diverged from common ancestry more than 500 million years ago). At the structural vertex, two very different reasons underlie the origin of potentially aptive features for initially nonadaptive reasons: physical principles that build “good” form by the direct action of physical laws upon plastic material (as in D'Arcy Thompson's theory of form), and architectural sequelae (spandrels) that arise as nonadaptive consequences of other features, and then become available for later cooptation (as exaptations) to aptive ends in descendant taxa. These two structural reasons differ strongly in the ahis-toricist implications of direct physical production independent of phyletic context vs. the explicit historical analysis needed to identify the particular foundation for the origin of spandrels in any individual lineage.

4.  As a conceptual basis for understanding the importance of recent ad­vances in evo-devo (the study of the evolution of development), the largely unknown history of debate about categories of homology, particularly the distinction between convergence and parallelism, provides our best ordering device — for we then learn to recognize the key contrast between parallelism as a positive deep constraint of homology in underlying generators (and therefore as a structuralist theme in evolution) and convergence as the opposite  {82}  sign of domination for external natural selection upon a yielding internal substrate that imposes no constraint (and therefore as a functionalist theme in evolution). As a beginning paradox, we must grasp why E. Ray Lankester coined the term homoplasy as a category of homology, whereas today's termi­nology ranks the concepts as polar opposites. Lankester wanted to contrast homology of overt structure (homogeny in his terms, or homology sensu stricto) with homology of underlying generators (later called parallelism) building the same structure in two separate lineages (homoplasy, or homol­ogy sensu lato, in Lankester's terms). Because parallelism could not be cashed out in operational terms (as science had no way, until our current revolution in evo-devo, to characterize, or even to recognize, these underlying genera­tors), proper conceptual distinctions between parallelism and convergence have generally not been made, and the two terms have even (and often) been united as subtypes of homoplasy (now defined in the current, and utterly non-Lankesterian sense, as opposite to homology). I trace the complex and con­fused history of this discussion, and show that structuralist thinkers, with doubts about panadaptationism, have always been most sensitive to this is­sue, and most insistent upon separating and distinguishing parallelism as the chief category of positive developmental constraint — a category that has now, for the first time, become scientifically operational.

5. I summarize the revolutionary empirics and conceptualizations of evo-devo in four themes, united by a common goal: to rebalance constraint and adaptation as causes and forces of evolution, and to acknowledge the perva­siveness and importance — also the synergy with natural selection, rather than opposition to Darwinian themes — of developmental constraint as a positive, structuralist, and internal force. The first theme explores the implications — for internally directed evolutionary pathways and consequent clumping of taxa in morphospace — of the remarkable and utterly unanticipated discovery of extensive “deep homology” among phyla separated at least since the Cam­brian explosion, as expressed by shared and highly conserved genes regulat­ing fundamental processes of development. I first discuss the role and ac­tion of some of these developmental systems — the ABC genes of Arabidopsis in regulating circlets of structures in floral morphology, the Hox genes of Drosophila in regulating differentiation of organs along the AP axis, and the role of the Pax-6 system in the development of eyes — in validating (only par­tially, of course) the archetypal theories of 19th century transcendental mor­phology, long regarded as contrary to strictly selectionist views of life's his­tory — particularly Goethe's theory of the leaf archetype, and Geoffroy's idea of the vertebral groundplan of AP differentiation. I then discuss the even more exciting subject of homologically conserved systems across distant phyla, as expressed in high sequence similarity of important regulators, com­mon rules of development (particularly the “Hoxology” followed in both ar­thropod and vertebrate ontogeny), and similar action of homeotic mutations that impact Hoxological rules by loss or gain of function. Geoffroy was partially right in asserting segmental homology between arthropods and ver­tebrates, particularly for the comparison of insect metameres with  {83}  rhombomeric segments in the developing vertebrate brain (a small part, perhaps, of the AP axis of most modern vertebrates, but the major component of the earliest fossil vertebrates), where the segments themselves may form differ­ently, but where rules of Hoxology then work in the same manner during later differentiation. I also defend the substantial validity of Geoffroy's other “crazy” comparison — the dorsoventral inversion of the same basic body plan between arthropods and vertebrates.

6. The second theme stresses the even more positive role of parallelism, based on common action of regulators shared by deep homology, in directing the evolutionary pathways of distantly related phyla into similar channels of adaptations thus more easily generated (thereby defining this phenomenon as synergistic and consistent with an expanded Darwinian theory, and not confrontational or dismissive of selection). I discuss such broad scale examples as the stunning discovery of substantial parallelism in the supposedly classical, “poster boy” expression of the opposite phenomenon of convergence — the development of eyes in arthropods, vertebrates, and cephalopods. The overt adult phenotype, of course, remains largely convergent, but homology of the underlying regulators demonstrates the strong internal channeling of parallel­ism. The vertebrate and squid version of Pax-6 can, in fact, both rescue the development of eyes in Drosophila and produce ectopic expression of eyes in such odd places as limbs. I also discuss smaller-scale examples of “conver­gence,” reinterpreted as parallelism, for even more precise similarities among separate lineages within coherent clades — particularly the independent con­version of thoracic limbs to maxillipeds, by identical homeotic changes in the same Hox genes, in several groups of crustaceans. Finally, I caution against Overextension and overenthusiasm by pointing out that genuine developmen­tal homologies may be far too broad in design, and far too unspecific in mor­phology, to merit a designation as parallelism, as in the role of distal-less in regulating “outpouchings” so generalized in basic structure, yet so different in form, as annelid parapodia, tunicate ampullae and echinoderm tube feet. I designate these overly broad similarities (that should not be designated as parallelism, or used as evidence for constraint by internal channeling) as “Pharaonic bricks” — that is, building blocks of such generality and multi­purpose utility that they cannot be labeled as constraints (with the obvious reductio ad absurdum of DNA as the homological basis of all life). By con­trast, the “Corinthian columns” of more specific conservations define the proper category of important positive constraint by internal channelings of parallelism based on homology of underlying regulators (just as the specific form of a Corinthian column, with its acanthus-leafed capital, represents a tightly constrained historical lineage that strongly influences the particular shape and utility of the entire resulting building).

7. My third and shorter theme — for this subject, though “classical” throughout the history of evolutionary thought, holds, I believe, less validity and scope than the others — treats the role of homologous regulators in pro­ducing rapid, even truly saltational, changes channeled into limited possibili­ties of developmental pathways (as in Goldschmidt's defense of discontinuous  {84}  evolution based upon mutations in rate genes that control ontogenetic trajec­tories). I discuss the false arguments often invoked to infer such saltational changes, but then document some limited, but occasionally important, cases of such discontinuous, but strongly channeled, change in macroevolution.

8. The fourth theme of top-down channeling from full ancestral comple­ments, rather than bottom-up accretion along effectively unconstrained path­ways of local adaptation, explores the role of positive constraint in establish­ing the markedly non-random and inhomogeneous population of potential morphospace by actual organisms throughout the history of life. Ed Lewis, in brilliantly elucidating the action of Hox genes in the development of Drosophila, quite understandably assumed (albeit falsely, as we later discov­ered to our surprise) that evolution from initial homonomy to increasing complexity of AP differentiation had been achieved by addition of Hox genes, particularly to suppress abdominal legs and convert the second pair of wings to halteres. In fact, the opposite process of tinkering with established rules, primarily by increased localization of action and differentiation in tim­ing (and also by duplication of sets, at least for vertebrate Hox genes), has largely established the increasing diversity and complexity of differentiation in bilaterian phyla. The (presumably quite homonomous) common ancestor of arthropods and vertebrates already possessed a full complement of Hox genes, and even the bilaterian common ancestor already possessed at least seven elements of the set. Moreover, the genomes of the most homonomous modern groups of onycophorans and myriapods also include a full set of Hox genes — so differentiation of phenotypic complexity must originate as a de­rived feature of Hox action, exapted from a different initial role. The Cam­brian explosion remains a crucial and genuine phenomenon of phenotypic diversification, a conclusion unthreatened by a putatively earlier common an­cestry of animal phyla in a strictly genealogical (not phenotypic) sense. The further evolution of admittedly luxuriant, even awesome, variety in major phyla of complex animals has followed definite pathways of internal channel­ing, positively abetted (as much as negatively constrained) by homologous developmental rules acting as potentiators for more rapid and effective selec­tion (as in the loss of snake limbs and iteration of prepelvic segments), and not as brakes or limitations upon Darwinian efficacy.

Chapter 11: The integration of constraint and adaptation: structural constraints, spandrels, and exaptation

1. D'Arcy Thompson's idiosyncratic, but brilliantly crafted and expressed, theory of form (1917,1942) presents a 20th century prototype for the generalist, or ahistorical, form of structural constraint: adaptation produced not by a functionalist mechanism like natural selection (or Lamarckism), but di­rectly and automatically impressed by physical forces operating under invari­ant laws of nature. This theory enjoyed some success in explaining the corre­lation of form and function in very simple and labile forms (particularly as influenced by scale-bound changes in surface/volume ratios). But similarly nongenetic (and nonphyletic) explanations do not apply to complex  {85}  creatures, and even D'Arcy Thompson admitted that his mechanism could not encompass, say, “hipponess,” but, at most, only the smooth transformations of these basic designs among closely related forms of similar Bauplan (the true theoretical significance of his much misunderstood theory of transformed coordinates). In summary, D'Arcy Thompson, the great student of Aristotle, erred in mixing the master's modes of causality — by assuming that the adap­tive value (or final cause) of well designed morphology could specify the physical forces (or efficient causes) that actually built the structures.

2. Stuart Kauffman and Brian Goodwin have presented the most cogent modern arguments in this tradition of direct physical causation. These argu­ments hold substantial power for explaining some features of relatively sim­ple biological systems, say from life's beginnings to the origin of prokaryotic cells, where basic organic chemistry and the physics of self-organizing sys­tems can play out their timeless and general rules. Such models also have sub­stantial utility in describing very broad features of the ecology and energy dynamics of living systems in general terms that transcend any particular taxonomic composition. But this approach founders, as did D'Arcy Thompson's as well, when the contingent and phyletically bound histories of particular complex lineages fall under scrutiny — and such systems do constitute the “bread and butter” of macroevolution. Nonetheless, Kauffman's powerful notion of “order for free,” or adaptive configurations that emerge from the ahistoric (even abiological) nature of systems, and need not be explained by particular invocations of some functional force like natural selection, should give us pause before we speculate about Darwinian causes only from evidence of functionality. This “order for free” aids, and does not confute, such func­tional forces as selection by providing easier (even automatic) pathways to­wards a common desideratum of adaptive biological systems.

3. I then turn to the second, and (in my judgment) far more important, theme of structural constraint in the fully historicist and phyletic context of aptive evolution by cooptation of structures already present for other reasons (often nonadaptive in their origin), rather than by direct adaptation for cur­rent function via natural selection. The central principle of a fundamental logical difference between reasons for historical origin and current functional utility — a vital component in all historical analysis, as clearly recognized but insufficiently emphasized by Darwin, and then unfortunately underplayed or forgotten by later acolytes — was brilliantly identified and dissected by Friedrich Nietzsche in his Genealogy of Morals, where he contrasted the ori­gin of punishment in a primal will to power, with the (often very different) utility of punishment in our current social and political systems.

4. Darwin himself invoked this principle of disconnection between histori­cal origin and current utility both in the Origin's first edition, and particularly in later responses to St. George Mivart's critique (the basis for the only chap­ter that Darwin added to later editions of the Origin) on the supposed inabil­ity of natural selection to explain the incipient (and apparently useless) stages of adaptive structures. Darwin asserted the principle of functional shift to ar­gue that, although incipient stages could not have functioned in the manner  {86}  of their final form, they might still have arisen by natural selection for a dif­ferent initial utility (feathers first evolved for thermoregulation and later co-opted for flight, for example). Darwin used this principle of cooptation, or functional shift, in two important ways that enriched and expanded his the­ory away from a caricatured panselectionist version — as the primary ground of historical contingency in phyletic sequences (for one cannot predict the di­rection of subsequent cooptation from different primary utilities), and as a source of structural constraint upon evolutionary pathways. But these Dar­winian invocations stopped short of a radical claim for frequent and impor­tant nonadaptive origins of structures co-opted to later utility. That is, Darwin rarely proceeded beyond the principle of originally adaptive origin for differ­ent function, with later cooptation to altered utility.

5. This important principle of cooptation of preexisting structures origi­nally built for different reasons has been so underemphasized in Darwinian traditions that the language of evolutionary theory does not even include a term for this central process — which Elisabeth Vrba and I called “exaptation” (Gould and Vrba, 1982). (The available, but generally disfavored, term “pre-adaptation” only speaks of potential before the fact, and has been widely re­jected in any case for its unfortunate, but inevitable, linguistic implication of foreordination in evolution, the very opposite of the intended meaning!)

6. I present a list of criteria for recognizing exaptations and separating them from true adaptations. I also discuss some outstanding examples of exaptation from the recent literature, with particular emphasis on the multi­ple exaptation of lens crystallins (in part for their fortuitous transparency, but for many other cooptable characteristics as well) in so many vertebrates and from so many independent and different original functions.

7. The exaptation of structures that arose for different adaptive reasons remains within selectionist orthodoxy (while granting structural constraint a large influence over historical pathways, in contrast with crude panadaptationism) by confirming a Darwinian basis for the adaptive origin of struc­tures, whatever their later history of exaptive shift. On the other hand, the theoretically radical version of this second, or historicist, style of structural constraint in evolution posits an important role for an additional phenome­non in macroevolution: the truly nonadaptive origin of structures that may later be exapted for subsequent utility. Many sources of such nonadaptive origin may be specified (see point 10 below), but inevitable architectural con­sequences of other features — the spandrels of Gould and Lewontin's termi­nology (1979) — probably rank as most frequent and most important in the history of lineages.

8. Spandrels (although unnamed and ungeneralized) have been acknowl­edged in Darwinian traditions, but relegated to insignificant relative frequen­cies by invalid arguments for their rarity, their structural inconsequentiality (the mold marks on an old bottle, for example), or their temporally subse­quent status as sequelae — with the first two claims empirically false, and the last claim logically false as a further confusion between historical origin and current utility.  {87} 

9. I affirm the importance and high relative frequency of spandrels, and therefore of nonadaptive origin, in evolutionary theory by two major argu­ments for ubiquity. First, for intrinsic structural reasons, the number of po­tential spandrels greatly increases as organisms and their traits become more complex. (The spandrels of the human brain must greatly outnumber the im­mediately adaptive reasons for increase in size; the spandrels of the cylindri­cal umbilical space of a gastropod shell, by contrast, may be far more limited, although exaptive use as a brooding chamber has been important in several lineages.) Second, under hierarchical models of selection, features evolved for any reason at one level generate automatic consequences at other levels — and these consequences can only be classified as cross-level spandrels (since they are “injected into” the new level, rather than actively evolved there).

10. The full classification of spandrels and modes of exaptation offers a resolving taxonomy and solution — primarily through the key concept of the “exaptive pool” — for the compelling and heretofore confusing (yet much discussed) problem of “evolvability.” Former confusion has centered upon the apparent paradox that ordinary organismal selection, the supposed ca­nonical mechanism of evolutionary change, would seem (at least as its pri­mary overt effect) to restrict and limit future possibilities by specializing forms to complexities of immediate environments, and therefore to act against an “evolvability” that largely defines the future macroevolutionary prospects of any lineage. The solution lies in recognizing that spandrels, al­though architecturally consequential, are not doomed to a secondary or un­important status thereby. Spandrels, and all other forms of exaptive potential, define the ground of evolvability, and play as important a role in macro-evolutionary potential as conventional adaptation does for the immediacy of microevolutionary success. I emphasize the centrality of the exaptive pool for solving the problem of evolvability by presenting a full taxonomy of catego­ries for the pool's richness, focusing on a primary distinction between “frank­lins” (or inherent potentials of structures evolved for other adaptive roles — that is, the classical Darwinian functional shifts that do not depart from adaptationism), and “miltons” (or true nonadaptations, arising from several sources, with spandrels as a primary category, and then available for later cooptation from the exaptive pool — that is, the class of nonadaptive origins that does challenge the dominant role of panadaptationism in evolutionary theory).

11. I argue that the concept of cross-level spandrels vastly increases the range, power and importance of nonadaptation in evolution, and also unites the two central themes of this book by showing how the hierarchically ex­panded theory of selection also implies a greatly increased scope for non-adaptive structural constraint as an important factor in the potentiation of macroevolution.

Chapter 12: Tiers of time and trials of extrapolationism

1. Darwin clearly recognized the threat of catastrophic mass extinction to the extrapolationist and uniformitarian premises underlying his claim for full  {88}  explanation of macroevolutionary results by microevolutionary causes (and not as a challenge to the efficacy of natural selection itself). Darwin therefore employed his usual argument about the imperfection of geological records to “spread out” apparent mass extinction over sufficient time for resolution by ordinary processes working at maximal rates (and therefore only increasing the intensity of selection).

2. The transition of the impact scenario (as a catastrophic trigger for the K-T extinction) from apostasy at its proposal in 1980 to effective factuality (based on the consilience of disparate evidence from iridium layers, shocked quartz and, especially, the discovery of a crater of appropriate size and age at Chicxulub) has reinstated the global paroxysms of classical catastrophism (in its genuinely scientific form, not its dismissive Lyellian caricature) as a legiti­mate scientific mechanism outside the Darwinian paradigm, but operating in conjunction with Darwinian forces to generate the full pattern of life's his­tory, and not, as previously (and unhelpfully) formulated, as an exclusive al­ternative to disprove or to trivialize Darwinian mechanisms.

3. If catastrophic causes and triggers for mass extinction prove to be gen­eral, or at least predominant in relative frequency (and not just peculiar to the K-T event), then this macroevolutionary phenomenon will challenge the cru­cial extrapolationist premise of Darwinism by being more frequent, more rapid, more intense and more different in effect than Darwinian biology (and Lyellian geology) can allow. Under truly catastrophic models, two sets of reasons, inconsistent with Darwinian extrapolationism by microevolu­tionary accumulation, become potentially important agents of macroevolu­tionary patterning: effectively random extinction (for clades of low N), and, more importantly, extinction under “different rules” from reasons regulating the adaptive origin and success of autapomorphic cladal features in normal times.

4. Catastrophic mass extinction, while breaking the extrapolationist credo, may suggest an overly simplified and dichotomous macroevolutionary model based on alternating regimes of “background” vs. “mass” extinction. Rather, we should expand this insight about distinctive mechanisms at different scales into a more general model of several rising tiers of time — with conven­tional Darwinian microevolution dominating at the ecological tier of short times and intraspecific dynamics; punctuated equilibrium dominating at the geological tier of phyletic trends based on interspecific dynamics (with species arising in geological moments, and then treated as stable “atoms,” or basic units of macroevolution, analogous to organisms in microevolution); and mass extinction (perhaps often catastrophic) acting as a major force of over­all macroevolutionary pattern in the global history of relative waxing and waning of clades. (I also contrast this preferred model of time's tiering with the other possible style of explanation, which I reject but find interesting nonetheless, for denying full generality to smooth Darwinian upward extrap­olation from the lowest level — namely, an equally smooth and monistic downward extrapolation from catastrophic mortality in mass extinction to  {89}  diminishing, but equally random and sudden, effects at all scales, as proposed in Raup's “field of bullets” model.)

5. In a paradoxical epilogue, I argue (despite my role as a longtime cham­pion of the importance and scientific respectability of unpredictable con­tingency in the explanation of historical patterns) that the enlargement and reformulation of Darwinism, as proposed in this book, will recapture for gen­eral theory (by adding a distinctive and irreducible set of macroevolutionary causes to our armamentarium of evolutionary principles) a large part of macroevolutionary pattern that Darwin himself, as an equally firm supporter of contingency, willingly granted to the realm of historical unpredictability because he could not encompass these results within his own limited causal structure of strict reliance upon smooth extrapolation from microevolutionary processes by accumulation through the immensity of geological time.

A FINAL THOUGHT. May I simply end by quoting the line that I wrote at the completion of a similar abstract (but vastly shorter, in a much less weighty book) for my first technical tome, Ontogeny and Phylogeny (1977b, p. 9): “This epitome is a pitiful abbreviation of a much longer and, I hope, more subtle development. Please read the book!”



The Essence of Darwinism and the
Basis of Modern Orthodoxy:
An Exegesis of the Origin of Species

A Revolution in the Small

Our theatrical and literary standards recognize only a few basic types of heroes. Most are preeminently strong and brave; some, in an occasional bone thrown to the marginal world of intellectuals, may even be allowed to tri­umph by brilliance. But one small section of the pantheon has long been re­served for a sideshow of improbables: the meek, the mild, the foolish, the insignificant, the ornamental — in short, for characters so disdained that they pass beneath notice and become demons of effectiveness by their invisibility. Consider the secretaries or chauffeurs who learn essential secrets because pa­trician bosses scarcely acknowledge their personhood and say almost any­thing in their presence; or the pageboys and schoolgirls who walk unnoticed through enemy lines with essential messages to partisans in conquered terri­tories.

Though few scholars have considered the issue in this light, I would argue that the intellectual agent of Darwin's victory falls into this anomalous cate­gory. To be sure, Darwin succeeded because he devised a mechanism, natural selection that possessed an unbeatable combination of testability and truth. But, at a more general level, Darwin triumphed by allowing the formerly meek to inherit the entire world of evolutionary theory.

Darwin's theory explicitly rejected and overturned the two evolutionary systems well known in Britain during his time (see next chapter for details) — Lamarck's (via Lyell`s exegesis in the Principles of Geology) and Chambers's (in the anonymously printed Vestiges of the Natural History of Creation). Both these theories sunk a deep root in the most powerful of cultural biases by describing evolution as an interaction of two opposing forces. The first — considered dominant, intrinsic and fundamental — yielded progress on the old euphonious (and sexist) theme of “the march from monad to man.” The sec­ond — designated as secondary, diversionary and superimposed — interrupted the upward flow and produced lateral dead-ends of specialized adaptations, from eyeless moles to long necked giraffes. Darwin, in his greatest stroke of  {94}  genius, took this secondary force, proposed a new mechanism for its opera­tion (natural selection), and then redefined this former source for superficial tinkering as fully sufficient to render all of evolution — thus branding the sep­arate and more exalted force of progress as illusory.

Such an argument poses an obvious logical dilemma: how can such power be granted to a force formerly viewed as so inconsequential? After all, evolu­tion must still construct the full pageant of life's history and the entire taxonomic panorama, even if we abandon the concept of linear order. Darwin's answer records the depth of his debt to Lyell, the man more responsible than any other for shaping Darwin's basic view of nature. Time, just time! (pro­vided that the “inconsequential” force of adaptation can work without limit, accumulating its tiny effects through geological immensity). The theory's full richness cannot be exhausted by the common statement that Darwinism pre­sents a biological version of the “uniformitarianism” championed by Lyell for geology, but I cannot think of a more accurate or more encompassing one-liner. (In a revealing letter to Leonard Horner, written in 1844, Darwin ex­claimed: “I always feel as if my books came half out of Lyell's brains ... for I have always thought that the great merit of the Principles [of Geology], was that it altered the whole tone of one's mind and therefore that when seeing a thing never seen by Lyell, one yet saw it partially through his eyes” (cited in Darwin, 1987, p. 55).)

Darwin, in his struggle to formulate an evolutionary mechanism during his annus mirabilis (actually a bit more than two years) between the docking of the Beagle and the Malthusian insight of late 1838, had embraced, but ulti­mately rejected, a variety of contrary theories — including saltation, inher­ently adaptive variation, and intrinsic senescence of species (see Gruber and Barrett, 1974; Kohn, 1980). A common thread unites all these abandoned ap­proaches: for they all postulate an internal drive based either on large pushes from variation (saltationism) or on inherent directionality of change. Most use ontogenetic metaphors, and make evolution as inevitable and as pur­poseful as development. Natural selection, by contrast, relies entirely upon small, isotropic, nondirectional variation as raw material, and views exten­sive transformation as the accumulation of tiny changes wrought by struggle between organisms and their (largely biotic) environment. Trial and error, one step at a time, becomes the central metaphor of Darwinism.

This theme of relentless accumulation of tiny changes through immense time, the uniformitarian doctrine of Charles Lyell, served as Darwin's touch­stone throughout his intellectual life. Uniformitarianism provides the key to his first scientific book (Darwin, 1842) on the formation of coral atolls by gradual subsidence of oceanic islands, long continued. And the same theme defines the central subject of his parting shot (1881), a book on the formation of vegetable mould by earthworms. Darwin, for lifelong reasons of personal style, did not choose to write a summary or confessional in lofty philosophi­cal terms, but he did want to make an exit with guns blazing on his favorite topic. Ironically, Darwin's overt subject of worms has led to a common inter­pretation quite opposite to his own intent — his misrepresentation as a dod­dering old naturalist who couldn't judge  {95}  the difference in importance between fishbait and fomenting revolution, and who, in recognizing evolution, just happened to be in the right place at the right time. In fact, Darwin's worm book presents an artfully chosen example of the deeper principle that under­lay all his work, including the discovery of evolution — the uniformitarian power of small changes cumulated over great durations. What better example than the humble worm, working literally beneath our notice, but making, grain by grain, both our best soils and the topography of England. In the pref­ace (1881, p. 6), Darwin explicitly draws the analogy to evolution by refuting the opinions of a certain Mr. Fish (wonderful name, given the context), who denied that worms could account for much “considering their weakness and their size”: “Here we have an instance of that inability to sum up the effects of a continually recurrent cause, which has often retarded the progress of sci­ence, as formerly in the case of geology, and more recently in that of the prin­ciple of evolution.”

Darwin waxed almost messianic in advancing this theme in the Origin of Species, for he understood that readers could not grasp his argument for evo­lution until they embraced this uniformitarian vision with their hearts. He confessed the a priori improbability of his assertion, given the norms and tra­ditions of western thought: “Nothing at first can appear more difficult to be­lieve than that the more complex organs and instincts should have been per­fected, not by means superior to, though analogous with, human reason, but by the accumulation of innumerable slight variations, each good for the indi­vidual possessor” (1859, p. 459). In his short concluding section on our gen­eral reluctance to accept evolution, he did not — probably for diplomatic rea­sons — identify specific cultural or religious barriers; instead, he spoke of our unfamiliarity with the crucial uniformitarian postulate: “But the chief cause of our natural unwillingness to admit that one species has given birth to other and distinct species, is that we are always slow in admitting any great change of which we do not see the intermediate steps . . . the mind cannot possibly grasp the full meaning of the term of a hundred million years; it cannot add up and perceive the full effects of many slight variations, accumulated during an almost infinite number of generations” (1859, p. 481).

To impress readers with the power of natural selection, Darwin continually stressed the cumulative effect of small changes. He reserved his best literary lines, his finest metaphors, for this linchpin of his argument — as in this famil­iar passage: “It may be said that natural selection is daily and hourly scruti­nizing, throughout the world, every variation, even the slightest; rejecting that which is bad, preserving and adding up all that is good; silently and in­sensibly working, whenever and wherever opportunity offers, at the improve­ment of each organic being in relation to its organic and inorganic conditions of life. We see nothing of these slow changes in progress, until the hand of time has marked the long lapse of ages” (1859, p. 84). Examine the smallest changes and variations, Darwin almost begs us. Let nothing pass beneath your notice. Cumulate, cumulate, and cumulate:

Certainly no clear line of demarcation has as yet been drawn between species and sub-species . . .; or, again, between sub-species and  {96}  well-marked varieties, or between lesser varieties and individual differences. These differences blend into each other in an insensible series; and a se­ries impresses the mind with the idea of an actual passage. Hence I look at individual differences, though of small interest to the systematist, as of high importance for us, as being the first step towards such slight varie­ties as are barely thought worth recording in works on natural history (1859, p. 51).

I need hardly stress Darwin's impact as one of the half dozen or so most revolutionary thinkers in western history. I want, instead, to emphasize a more curious aspect of his status — his continuing relevance, indeed his benev­olent hovering over almost all our current proceedings. We may revere New­ton and Lavoisier as men of equal impact, but do modern physicists and chemists actively engage the ideas of these founders, as they pursue their daily work? Darwin, on the other hand, continues to bestride our world like a colossus — so much so that I can only begin this book on the structure of evo­lutionary theory by laying out Darwin's detailed vision as a modern starting point, a current orthodoxy only lightly modified by more than a century of work. I do, in this book, advocate some major restructuring, in the light of new concepts and findings, and with the approbation of more and more col­leagues as our understanding of evolution broadens. But Darwin remains our context — and my proposed restructuring represents an extension, not a re­placement, of his vision. The hierarchical theory of selection builds a world different from Darwin's in many important respects, but we do so by ex­tending his mechanism of selection to a larger realm than he acknowledged — that is, to levels both below and above his focus on the struggle among or­ganisms.

When Cassius spoke his words about Caesar (paraphrased above), he added his puzzlement at Caesar's extraordinary success: “Upon what meat doth this our Caesar feed, that he is grown so great.” I shall argue in this chapter that Darwin's continued, pervasive relevance arises from his capacity for revolutionary innovation at two opposite poles of scientific practice — the immediate strategy of formulating a methodology for everyday research, and the most general discussion of causes and phenomena in the natural world (the questions that will not go away, and that air continually from college bull sessions, to TV talk shows, to learned treatises on the nature of things). Dar­win's residence at both poles of immediate methodology and broadest theo­retical generality begins with his distinctive attitude towards the central im­portance of daily, palpable events in nature, and their power to account for all evolution by cumulation — hence my choice of an opening topic for this chapter (see Fig. 2-1).

Caesar voiced his suspicions of Cassius, fearing men who think too much (may all despots thus beware). But his grudging words of praise might well be invoked to epitomize the reasons for Darwin's unparalleled success: “He reads much; he is a great observer, and he looks quite through the deeds of men.”


Darwin as a Historical Methodologist


An old quip, highlighting the intractability of philosophical dualism, pro­claims: “what's matter? never mind; what's mind? doesn't matter.” Predarwinian evolutionary systems embodied the same kind of Catch-22, this time in painful and practical terms, destined to ensnare any budding naturalist who hoped to study organisms by direct confrontation with testable hypothe­ses. Lamarck's system, for example, contrasted an intrinsic force of progress with a diversionary, and clearly secondary, force of adaptation to changing local environments. The secondary process worked in the immediate here and  {98}  now, and might be engaged empirically by studies of adaptation and heredity. But the more important primary force, the source of natural order and the ul­timate cause of human mentality, lurked in the background of time's immen­sity, and at the inaccessible interior of the very nature of matter. This charac­terization creates an intolerable dilemma for anyone who holds (as Darwin did) that science must be defined as testable doing, not just noble thinking. Recalling my opening quip, Lamarck's system virtually mocked the empirical approach to science, and forestalled any growing confidence in evolution: what is important cannot be seen; what can be seen is not important.

Darwin used a brilliant argument to cut through this dilemma, thus mak­ing the study of evolution a practical science. He acknowledged Lamarck's implied claim that small scale adaptation to local environment defines the tractable subject matter of evolution. But he refuted the disabling contention that adaptation in this mode only diverted the “real” force of evolution into side channels and dead ends. And he revised previous evolutionary thinking in the most radical way — by denying that Lamarck's “real” force existed at all, and by encompassing its supposed results as consequences of the “subsid­iary” force accumulated to grandeur by the simple expedient of relentless action over sufficient time. Darwin established our profession not only by discovering a force — natural selection — that seems both powerful and true; he also, perhaps more importantly, made evolution accessible to science by granting to empiricists their most precious gifts of tractability and testability. The essence of Darwin's theory (specified in the next section) owes as much to his practical triumph at this immediate scale of daily work, as to his broadest perception that western views of nature had been seriously awry, and largely backwards.

Darwin, as we all know, began the last chapter of the Origin with a claim that “this whole volume is one long argument” (1859, p. 459). Fine, but an argument for what? For evolution itself? In part, of course, but such a general theme cannot mark the full intent of Darwin's statement, for the bulk of the Origin moves well beyond the basic arguments for evolution's factuality, as Darwin proceeds to craft a defense for natural selection and for the philoso­phy of nature so entailed. “One long argument” for natural selection, then? Again, in part; but we now confront the obverse of my last statement: too much of the Origin details basic evidence for evolution, independent of any particular mechanism of change. Instead, we must ask what deeper subject underlies both the defense of evolution as a fact and the proposal of a mecha­nism to explain its operation? How should we characterize the “one long ar­gument” that pervades the entire book?

Ghiselin (1969) correctly identified the underlying theme as the construc­tion, and defense by example, of a methodology — a mode of practice — for testing both the fact and mechanism of evolutionary change. But I cannot agree with Ghiselin that Darwin's consistent use of “hypothetico-deductive” reasoning constitutes his long argument (see Kitcher, 1985), for this style of scientific procedure, whatever its merits or problems, has been advocated as a general methodology for all scientific activity (see Hempel, 1965). Darwin, I  {99}  believe, sought to construct and defend a working method for the special sub­ject matter of evolutionary inquiry — that is, for the data of history.

Inferences about history, so crucial to any evolutionary work, had been plagued by problems of confidence that seemed to bar any truly scientific in­quiry into the past. Darwin knew that evolution would not win respect until methods of historical inference could be established and illustrated with all the confidence of Galileo viewing the moons of Jupiter. He therefore set out to formulate rules for inference in history. I view the Origin as one long illus­tration of these rules. Historical inference sets the more general theme under­lying both the establishment of evolution as a fact, and the defense of natural selection as its mechanism. The “one long argument” of the Origin presents a comprehensive strategy and compendium of modes for historical inference (see fuller exposition of this view in Gould, 1986). We must grasp Darwin's practical campaign on this battlefield in order to understand his radical phi­losophy, and to identify the features of his theory that count as essential to any definition of “Darwinism.”


Reading Darwin has been a persisting and central joy in my intellectual life. Lyell and Huxley may have been greater prose stylists, with more consistency in the ring and power of their words. Yet I give the nod to Darwin, and not only for the greater depth and power of his ideas. Darwin often wrote quite ordinary prose, page after page. But then, frequently enough to rivet the at­tention of any careful reader, his passion bursts through, and he makes a point with such insight and force (almost always by metaphor) that under­standing breaks like sunrise. Every evolutionist can cite a list of favorite Dar­winian passages, written on well-worn index cards for lectures (or, now, eter­nally embedded in PowerPoint files), posted on the office door or prominently displayed above the typewriter (now the computer terminal), or simply (and lovingly) committed to memory.

Several of my favorite passages celebrate the broadened understanding of nature that derives from recognizing organisms as products of history, rather than objects created in their present state. Darwin writes (1859, 485486):*  {100} 

When we no longer look at an organic being as a savage looks at a ship, as at something wholly beyond his comprehension; when we regard every production of nature as one which has had a history; when we contemplate every complex structure and instinct as the summing up of many contrivances, each useful to the possessor, nearly in the same way as when we look at any great mechanical invention as the summing of the labor, the experience, the reason, and even the blunders of numerous workmen; when we thus view each organic being, how far more interesting, I speak from experience, will the study of natural history become!

By contrast, Darwin's chief quarrel with creationism resides not so much in its provable falseness, but in its bankrupt status as an intellectual argument — for a claim of creation teaches us nothing at all, but only states (in words that some people may consider exalted) that a particular creature or feature exists, a fact established well enough by a simple glance: “Nothing can be more hopeless than to attempt to explain the similarity of pattern in members of the same class, by utility or by the doctrine of final causes ... On the ordinary view of the independent creation of each being, we can only say that so it is; — that it has so pleased the Creator to construct each animal and plant” (p. 435).

Moreover, and more negatively, creation marks the surrender of any at­tempt to understand connections and patterns. We express no causal insight whatever when we say that taxonomic order reflects the plan of a creator — for unless we can know the will of God, such a statement only stands as a re­dundant description of the order itself. (And God told us long ago, when he spoke to Job from the whirlwind, that we cannot know his will — “canst thou draw out leviathan with a hook?”) Darwin, an ever genial man in the face of endless assaults upon his patience, directed several of his rare caustic com­ments against the ultimate idea stopping claim that God so made it, praise his name. Darwin notes, for example, that horses are sometimes born with faint striping on their hides. A creationist can only assert that God made each equine species of zebras, horses, and asses alike, with such tendencies to vary and thereby to display, if only occasionally, the more comprehensive type.  {101}  Evolution, on the other hand, supplies a true cause for an anomaly by posit­ing community of descent with retention of ancestral states by heredity — something that might be tested in many ways, once we understand the me­chanics of inheritance. (The following passage appears just before Darwin's summary to Chapter 5 on laws of variation.) Darwin lambastes the cre­ationist alternative as causally meaningless: “To admit this view is, as it seems to me, to reject a real for an unreal, or at least for an unknown, cause. It makes the works of God a mere mockery and deception; I would almost as soon believe with the old and ignorant cosmogonists, that fossil shells had never lived, but had been created in stone so as to mock the shells now living on the sea-shore” (p. 167).

If we must locate our confidence about evolution in evidence for history — in part directly from the fossil record, but usually indirectly by inference from modern organisms — by what rules of reason, or canons of evidence, shall his­tory then be established? Darwin's “long argument,” in my view, can best be characterized as a complex solution to this question, illustrated with copious examples. We must first, however, specify the kinds of questions that can­not be answered. Many revealing statements in the Origin circumscribe the proper realm of historical inference by abjuring what cannot be known, or usefully comprehended under current limits. Darwin, for example, and fol­lowing Hutton, Lyell and many other great thinkers, foreswore (as beyond the realm of science) all inquiry into the ultimate origins of things.* In the first paragraph of Chapter 7 on instincts, for example, Darwin writes (1859, p. 207): “I must premise, that I have nothing to do with the origin of the pri­mary mental powers, any more than I have with that of life itself.” Darwin in­voked the same comparison in discussing the evolution of eyes, one of his greatest challenges (and firmest successes). He states that he will confine his attention to transitions in a structural sequence from simple to complex, and not engage the prior issue — answerable in principle, but beyond the range of knowledge in his day — of how sensitivity to light could arise within nervous tissue in the first place (1859, p. 187): “How a nerve comes to be sensitive to light, hardly concerns us more than how life itself first originated.” Most crucially, and in a savvy argument that saved his entire system in the face of con­temporary ignorance on a central issue, Darwin argues over and over again that we may bypass the vital question of how heredity works, and how varia­tions arise — and only illustrate how evolution can occur, given the commonplace  {102}  observation that sufficient variation does exist, and is inherited often enough:

Whatever the cause may be of each slight difference in the offspring from their parents — and a cause for each must exist — it is the steady accumu­lation, through natural selection, of such differences, when beneficial to the individual, that gives rise to all the more important modifications of structure, by which the innumerable beings on the face of this earth are enabled to struggle with each other, and the best adapted to sur­vive (p. 170 — see also p. 131 for Darwin's argument that when we as­cribe variation to “chance,” we only mean to express our ignorance of causes).

Having established a domain of testability by exclusion, Darwin laid out his methodology for history — never explicitly to be sure, but with such accu­mulating force by example that the entire book becomes “one long argu­ment” for the tractability of his new science. Those of us who practice the sci­ences of reconstructing specific events and unravelling temporal sequences have always fought a battle for appropriate status and respect, no less so to­day than in Darwin's time (see Gould, 1986), against those who would view such work as a “lesser” activity, or not part of science at all. History presents two special problems: (1) frequent absence of evidence, given imperfections of preservation; and (2) uniqueness of sequences, unrepeatable in their con­tingent complexity, and thereby distancing the data of history from such stan­dard concepts as prediction, and experimentation.

We may epitomize the dilemma in the following way: many people define science as the study of causal processes. Past processes are, in principle, unobservable. We must therefore work by inference from results preserved in the historical record. We must study modern results produced by processes that can be directly observed and even manipulated by experiment — and we must then infer the causes of past results by their “sufficient similarity” (Steno's principle — see Gould, 1981c) with present results. This procedure re­quires, as Mill (1881) and other philosophers recognized long ago, a method­ological assumption of temporal invariance for laws of nature. Historical study manifests its special character by placing primary emphasis upon com­parison and degrees of similarity, rather than the canonical methods of sim­plification, manipulation, controlled experiment, and prediction.

Darwin had done some paleontological work, particularly in his treatises on barnacles (1851-1858), and his important discoveries of South American fossil vertebrates (formally named and described by Owen, at Darwin's invi­tation). But Darwin was not primarily a paleontologist, and he did not intend to base his argument for evolution on the evidence of fossils — especially since he viewed the stratigraphic record, with its vast preponderance of gaps over evidence, as more a hindrance than an aid to his theory (see chapters 9 and 10 of the Origin). Thus, of the two major sources for historical reconstruction — direct but imperfect information from fossils, and indirect but copious data from modern organisms — Darwin preferred the second as his wellspring of  {103}  documentation. The Origin therefore focuses upon the establishment of a methodology for making inferences about history from features of modern organisms — and then using these multifarious inferences to prove both the fact of evolution and the probability of natural selection as a primary mecha­nism of change.


Darwin, as a subtle and brilliant thinker, must be read on several levels. Con­sider just three, at decreasing domains of overt display, but increasing realms of generality: On the surface — a lovely, and not a pejorative, location for any student of nature — each book treats a particular puzzle: different forms of flowers on the same plant (1877), modes of formation for coral atolls (1842), formation of soil by worms (1881), styles of movement in climbing plants (1880a), the fertilization of orchids by insects (1862). At an intermediary level, as Ghiselin (1969) showed in his innovative study of the entire Darwin­ian corpus, each book forms part of a comprehensive argument for evolution itself. But I believe that we must also recognize a third, even deeper and more comprehensive layer of coordinating generality — Darwin's struggle to con­struct and apply a workable method for historical inference: a series of proce­dures offering sufficient confidence to place the sciences of history on a par with the finest experimental work in physics and chemistry. I have come to re­gard each of Darwin's books as, all at the same time, a treatment of a particu­lar puzzle (level one), an argument for an evolutionary worldview (level two), and a treatise on historical methodology (level three). But the methodological focus of level three has usually been overlooked because Darwin chose to work by practice rather than proclamation.

Darwin recognized that several methods of historical inference must be developed, each tailored to the nature and quality of available evidence. We may order his procedures by decreasing density of available information. I recognize four waystations in the continuum and argue that each finds a pri­mary illustration in one of Darwin's books on a specific puzzle in natural his­tory. The Origin of Species, as his comprehensive view of nature, uses all four methods, and may therefore be read as a summation of his seminal contribu­tion to the methodology of historical science. I shall list, and then illustrate with examples from the Origin, these four principles ordered by decreasing density of information.

Uniformity. Or working up by extrapolation from direct observations on rates and modes of change in modern organisms. Call this, if you will, the worm principle to honor Darwin's last book (1881), which explains the topsoil and topography of England by extrapolating the measured work of worms through all scales of time, from the weight of castings left daily on a patch of sod to the historical and geological realms of millennia to millions of years.

Sequencing. Or the definition and ordering of various configurations,  {104}  previously regarded as unrelated and independent, into stages of a single his­torical process. Here we cannot observe the changes between configurations directly and we must therefore work by recognizing them as temporally or­dered products of a single underlying process of change. Call this, if you will, the coral reef principle to honor Darwin's first book (1842) on a scientific subject. His successful theory proposes a single historical process for the for­mation of coral atolls by recognizing three configurations of reefs — fringing reefs, barrier reefs, and atolls — as sequential stages in the foundering of oce­anic islands.

Consilience (concordance of several). We now reach a break in types of information. Methods 1 and 2 permit the reconstruction of histor­ical sequences, either by extrapolating up from the most palpable and testable of daily changes (method 1), or by ordering a series of configurations as tem­poral stages (method 2). In many cases, however, we cannot reconstruct se­quences, and must infer history from the configuration of a single object or circumstance. Of the two major methods for inferring history from single configurations, consilience calls upon a greater range of evidence. This word, coined by William Whewell in 1840, means “jumping together.” By this term, Whewell referred to proof by coordination of so many otherwise unrelated consequences under a single causal explanation that no other organization of data seems conceivable. In a sense, consilience defines the larger method un­derlying all Darwin's inference from historical records. In a more specific con­text, I use consilience (see Gould, 1986) for Darwin's principal tactic of bring­ing so many different points of evidence to bear on a single subject, that history wins assent as an explanation by overwhelming confirmation and unique coordination. Call this, if you will, the different flowers principle to honor the extraordinary range of evidence that Darwin gathered (1877) to forge a historical explanation for why some taxa bear different forms of flowers on the same plant.

Discordance (dissonance of one). Here we reach a rock bottom of minimalism — unfortunately all too common in a world of limited informa­tion. We observe a single object, but not enough relevant items to forge consilience about its status as the product of history. How can we work from unique objects? How shall we infer history from a giraffe? Darwin tells us to search for a particular form of discordance — some imperfection or fail­ure of coordination between an organism and its current circumstances. If such a quirk, oddity, or imperfection — making no sense as an optimal and immutable design in a current context — wins explanation as a holdover or vestige from a past state in different circumstances, then historical change may be inferred. Call this, if you will, the orchid principle (though I have also designated it as the panda principle for my own favorite example, perforce unknown to Darwin, of the panda's false thumb, Gould, 1980d), to honor Darwin's argument (1862) for orchids as products of history. Their intricate adaptations to attract insects for fertilization cannot be read as wonders of optimal design, specially created for current utilities, for they represent con­traptions, jury-rigged from the available parts of ordinary flowers.  {105} 

The Origin of Species presents an ingenious compendium of all four methods.

UNIFORMITY. People who do not understand science in their bones, and who think that revolutionary treatises must be presented as ideological mani­festos at broadest scale, often express surprise and disappointment in reading the Origin, especially at Darwin's opening chapter. They expect fanfare, and they get fantails — pigeons, that is. But Darwin ordered his book by conscious intent and strategy. He knew that he had to demonstrate evolution with data, not simply proclaim his new view of life by rhetoric. Uniformitarianism em­bodied his best method based on maximal information — so he started from the smallest scale, change in domestication, and worked up to the history of life. As a member of two London pigeon fancying clubs (which he had joined, not from an abiding affection for this scourge of cities, but to gain practical information about evolution in the small), Darwin led from his acquired strength.

What better starting point, under method 1, than indubitable proof of historical change in domesticated plants and animals. The logic of the Origin employs one long analogy between artificial and natural selection, with uni­formity as the joining point. Darwin writes in his introduction (p. 4): “At the commencement of my observations it seemed to me probable that a careful study of domesticated animals and of cultivated plants would offer the best chance of making out this obscure problem. Nor have I been disappointed; in this and in all other perplexing cases I have invariably found that our knowl­edge, imperfect though it be, of variation under domestication, afforded the best and safest clue.”

Darwin continually drives home this analogy and extrapolation: if by arti­ficial selection at small scale (as we know for certain), why not by natural se­lection at larger scale: “If it profit a plant to have its seeds more and more widely disseminated by the wind, I can see no greater difficulty in this being effected through natural selection, than in the cotton-planter increasing and improving by selection the down in the pods on his cotton-trees” (p. 86).

But this argument by uniformitarian extrapolation presents a serious dif­ficulty (exploited by Fleeming Jenkin, 1867, in the famous critique that Dar­win ranked so highly, and took so seriously in revising the Origin): change surely occurs in domestication, but suppose that species function like glass spheres with a modal configuration at the center and unbridgeable limits to variation representing the surface. Artificial selection could then bring mor­phology from the center to the surface, but no further — and the key argument for smooth extrapolation to all change over any time would fail. Darwin therefore staked a verbal claim for no limit. “What limit can be put to this power, acting during long ages and rigidly scrutinizing the whole constitu­tion, structure, and habits of each creature — favoring the good and rejecting the bad? I can see no limit to this power, in slowly and beautifully adapting each form to the most complex relations of life” (p. 469).

Darwin then applied the full sequence of extrapolation to the natural  {106}  world, beginning with individual variants as the source of subspecies, then moving to subspecies as incipient species, and finally to species as potential ancestors for branches of life's tree — a full range of scales from variation within a population to the entire pageant of life: “I look at individual differ­ences, though of small interest to the systematist, as of high importance for us, as being the first step towards such slight varieties as are barely thought worth recording in works of natural history. And as I look at varieties which are in any degree more distinct and permanent, as steps leading to more strongly marked and more permanent varieties; and at these latter, as leading to subspecies, and to species” (p. 51).

Darwin invoked this first method, a strong argument based on maximal information at smallest scale, as his favored choice when available. To cite just three instances as a sampler: (1) the paleontological panorama may be read as a story of gradual evolution because species in adjacent strata show minimal differences, but these differences increase gradually as stratigraphic distance expands (p. 335). (2) When we find hints of the feather patterns of rock pi­geon in highly modified breeds, we do not hesitate to interpret these designs as vestiges of an ancestral stock; therefore, the faint stripes that we sometimes observe in coats of young horses point to a common origin for all species in the clade of horses, asses and zebras (pp. 166167). (3) Marine molluscs of­ten exhibit brighter colors in warmer waters. We note this pattern both among varieties of a single species living in cold and warm waters, and among related species. A creationist explanation requires uncomfortable spe­cial pleading: God sometimes makes a species with bright shells in warm cli­mates, but he allows other species to vary naturally, in the same geographic pattern, within a single created kind. An evolutionist, using method one, will recognize these phenomena as two stages in a single sequence of extrapola­tion from smaller to larger scale (p. 133).

SEQUENCING. We can use a second style of inference about temporal order when we cannot obtain adequate data about the nature of immediate changes at smallest scale. Since historical processes begin at different times and pro­ceed at varying rates, all stages of a sequence may exist simultaneously (for example, stage one in case A, which began very recently; stage two in case B, which began at the same time, but has proceeded at an uncommonly rapid rate; and stage three in case C, which began long ago). Thus, fringing reefs, barrier reefs and atolls all exist now. When we recognize these forms as se­quential stages of a single process, we may infer the pathway of history.

Darwin epitomizes method two in writing (p. 51): “A series impresses the mind with the idea of an actual passage.” Invoking his usual starting point, Darwin presents a first example from breeds of domesticated pigeons. The more adequate data of method one — observed steps of passage, accumulating to greater and greater difference in time — no longer exist, for the transitional populations have died, and only a set of morphological “islands,” represent­ing a set of established breeds, remains. But these islands can be ordered as a plausible sequence of change between ancestral rock pigeons and the most aberrant of artificially produced breeds: “Although an English carrier or  {107}  short-faced tumbler differs immensely in certain characters from the rock pi­geon, yet by comparing the several sub-breeds of these breeds, more espe­cially those brought from distant countries, we can make an almost perfect series between the extremes of structure” (p. 27).

Darwin uses method two in a special and crucial way throughout the Ori­gin. Several of the most telling critiques against Darwin's style of evolution by gradualistic continuity — best represented in Mivart's famous argument (1871) about inviability of “incipient stages of useful structures” (see Chap­ter 11 for full treatment) — held that insensibly graded passages between puta­tive ancestors and descendants could not even be conceptualized, much less documented. Charges of inconceivability took several forms, each reducible to the claim that you can't get from here to there, however well the beginning and end points may function. Consider the two most prominent formula­tions: (1) Early stages (when rudimentary) could provide no adaptive advan­tage, however valuable the final product (2) Major functional changes cannot occur because intermediary stages would fall into a never-never land of invia­bility, with the original (and essential) function lost, and the new operation not yet established.

Darwin offered a twofold response to these arguments, both using this sec­ond historical method of sequencing. He first presented theoretical arguments for the conceivability, even the likelihood, of intermediary stages in supposed cases of impossibility. He argued that early stages, too small to work in their eventual manner, could have performed different functions at the outset, and been coopted later for another style of life. (Incipient wings, originally used in thermoregulation, became organs of flight when they evolved to sufficiently large size to provide “fortuitous” aerodynamic benefits — see Kingsolver and Koehl, 1985, for an experimental validation of this scenario, and Gould, 1991b, for general discussion). As the misleadingly named principle of “pre-adaptation,” this concept of functional shift became an important principle in evolutionary theory (see Chapter 11). Darwin writes, using a verbal intensifier rarely found in his prose: “In considering transitions of organs, it is so important to bear in mind the probability of conversion from one function to another” (p. 191).

As a response to charges of inviability for intermediary stages, Darwin invoked the important principle of redundancy as a norm for organic structures and functions. Most important functions can be performed by more than one organ; and most organs work in more than one way. By coupling these two aspects of redundancy, transitions in single organs can easily be conceived. An organ doesn't mysteriously invent a new function, but usually intensifies and specializes a previously minor use, while shedding an old primary opera­tion. This previously major function can then be lost because other organs continue to do the same necessary job.

Ironically, we now recognize Darwin's favorite example of such redun­dancy as not only incorrect, but truly backwards (Gould, 1989b) — the evolu­tion of lungs from swimbladders. (In fact, swimbladders evolved from lungs, see Liem, 1988). Darwin ran his transition in the wrong way, but his argu­ment for redundancy as the key to viability for intermediary steps remains  {108}  correct and crucially important, for the logic works equally well in either direction. Ancestral fishes maintained two systems for breathing — gills and lungs (as do modern lungfish, taxonomically called Dipnoi, or “two breath­ing”). The original lung probably played a subsidiary role in buoyancy; this function could be enhanced, and the original use in breathing deleted, be­cause gills could adopt the entire respiratory burden. Darwin wrote (pp. 204205): “For instance, a swim bladder has apparently been converted into an air-breathing lung. The same organ having performed simultaneously very different functions, and then having been specialized for one function; and two very distinct organs having performed at the same time the same func­tion, the one having been perfected whilst aided by the other, must often have largely facilitated transitions.”

As a second response, Darwin proceeded beyond conceivability and tried to document actual sequences for supposedly impossible transitions — as in the evolution of a light-sensitive spot into an “organ of extreme perfection” like the vertebrate eye. These sequences cannot represent true phylogenies (since they consist solely of living species), but they do constitute structural series illustrating the conceivability of transitions. After admitting, for exam­ple, that the gradual evolution of such a miracle of workmanship as the eye “seems, I freely confess, absurd in the highest possible degree” (p. 186), Dar­win presents a structural series of disparate animals, including working con­figurations proclaimed impossible by opponents: “Yet reason tells me, that if numerous gradations from a perfect and complex eye to one very imperfect and simple, each grade being useful to its possessor, can be shown to exist... then the difficulty of believing that a perfect and complex eye could be formed by natural selection, though insuperable by our imagination, can hardly be considered real” (p. 186).

Darwin applies this principle to behavior and its products, as well as to form. For the exquisite mathematical regularity of the honeycomb, he writes (p. 225): “Let us look to the great principle of gradation, and see whether Na­ture does not reveal to us her method of work.” (See also page 210 on com­plex instincts and their explanation by the establishment of structural series.)

CONSILIENCE (CONCORDANCE OF SEVERAL). Darwin took great pride in his formulation of natural selection as a theory for the mechanism of phyletic change. But he granted even more importance to his relentless pre­sentation of dense documentation for the factuality of change — for only such a cascade of data would force the scientific world to take evolution seriously. (The contrast between the Origin as a compendium of facts, and Lamarck's Philosophie zoologique as a purely theoretical treatise, strikes me as an even more distinguishing difference than the disparate causal mechanisms pro­posed by the two authors.) Facts literally pour from almost every page of the Origin, a feature that became even more apparent following Darwin's forced change of plans, and his decision to compress his projected longer work into the “abstract” that we call the Origin of Species — a revised strategy that led him to omit almost every reference and footnote, and almost all discursive discussion between bits of information. In some parts, the Origin reaches an  {109}  almost frenetic pace in its cascading of facts, one upon the other. Only Dar­win's meticulous sense of order and logic of argument save the work from dis­abling elision and overload.

Whenever he introduces a major subject, Darwin fires a volley of disparate facts, all related to the argument at hand — usually the claim that a particular phenomenon originated as a product of history. This style of organization vir­tually guarantees that Whewell's “consilience of inductions” must become the standard method of the Origin. Darwin's greatest intellectual strength lay in his ability to forge connections and perceive webs of implication (that more conventional thinking in linear order might miss). When Darwin could not cite direct evidence for actual stages in an evolutionary sequence, he re­lied upon consilience — and sunk enough roots in enough directions to pro­vide adequate support for a single sturdy trunk of explanation.

Again, Darwin starts with pigeons, unleashing a cannonade of disparate arguments, all pointing to the conclusion that modern breeds of pigeons de­rive from a single ancestral stock. None of these facts permits the construc­tion of an actual temporal series (methods one and two); but all identify the features of a current configuration that point to history as the underlying cause. Darwin, as usual, proceeds by particular example, but I doubt that a better general description of consilience could be formulated:

From these several reasons, namely, the improbability of man having formerly got seven or eight supposed species of pigeons to breed freely under domestication; these supposed species being quite unknown in a wild state, and their becoming nowhere feral; these species having very abnormal characters in certain respects, as compared with all other Columbidae, though so like in most other respects to the rock pigeon; the blue color and various marks occasionally appearing in all the breeds, both when kept pure and when crossed; the mongrel offspring being perfectly fertile; — from these several reasons, taken together, I can feel no doubt that all our domestic breeds have descended from the Columba livia with its geographical subspecies (pp. 26-27).

Every scholar could cite a favorite case of Darwinian consilience. For my part, I especially admire Darwin's uncharacteristically long discussion (pp. 388406) on transport from continental sources and subsequent evolu­tion to explain the biotas of oceanic islands. Consider the main items in Dar­win's own order of presentation:

(1) The general paucity of endemic species on islands, contrasted with comparable areas of continents; why should God put fewer species on islands?

(2) The frequent displacement of endemic island biotas by continental spe­cies introduced by human transport. If God created species for islands, why should species designed for continents so often prove superior in competi­tion: “He who admits the doctrine of the creation of each separate species, will have to admit, that a sufficient number of the best adapted plants and an­imals have not been created on oceanic islands; for man has unintentionally stocked them from various sources far more fully and perfectly than has na­ture” (p. 390).  {110} 

(3) Taxonomic disparity of endemic species within groups records ease of access, not created fit to oceanic environments: “Thus in the Galapagos Is­lands nearly every land bird, but only two out of the eleven marine birds, are peculiar; and it is obvious that marine birds could arrive at these islands more easily than land birds” (pp. 390-391).

(4) Biotas of oceanic islands often lack the characteristic groups of similar habitats on continents. On these islands, endemic members of other groups often assume the ecological roles almost always occupied by more appropri­ate or more competitive taxa in the richer faunas of continents — for example, reptiles on the Galapagos, or wingless birds on New Zealand, acting as surro­gates for mammals.

(5) In endemic island species, features operating as adaptations in related species on continents often lose utility when their island residences do not feature the same environment: “For instance, in certain islands not tenanted by mammals, some of the endemic plants have beautifully hooked seeds; yet few relations are more striking than the adaptation of hooked seeds for transportal by the wool and fur of quadrupeds. This case presents no dif­ficulty on my view, for a hooked seed might be transported to an island by some other means; and the plant then becoming slightly modified, but still re­taining its hooked seeds, would form an endemic species, having as useless an appendage as any rudimentary organ” (p. 392).

(6) Peculiar morphological consequences often ensue when creatures seize places usually inhabited by other forms that could not reach an island. Many plants, herbaceous in habit on continents, become arboraceous on islands otherwise devoid of trees.

(7) Suitable organisms frequently fail to gain access to islands. Why do so many oceanic islands lack frogs, toads, and newts that seem so admirably adapted for such an environment? “But why, on the theory of creation, they should not have been created there, it would be very difficult to explain” (p. 393).

(8) Correlation of biota with distance. Darwin could find no report of terrestrial mammals on islands more than 300 miles from a continent. He pre­sents the obvious evolutionary explanation for a disturbing creationist co­nundrum:

It cannot be said, on the ordinary view of creation, that there has not been time for the creation of mammals; many volcanic islands are suf­ficiently ancient, as shown by the stupendous degradation which they have suffered and by their tertiary strata: there has also been time for the production of endemic species belonging to other classes ... why, it may be asked, has the supposed creative force produced bats and no other mammals on remote islands? On my view this question can easily be an­swered; for no terrestrial mammal can be transported across a wide space of sea, but bats can fly across (p. 394).

(9) Correlation with ease of access. Creatures often manage to cross shal­low water barriers between a continent and island, but fail to negotiate deep-water gaps of the same distance.  {111} 

(10) Taxonomic affinity of island endemics — perhaps the most obvious point of all: why are the closest relatives of island endemics nearly always found on the nearest continent or on other adjacent islands?

Any honorable creationist, after suffering such a combination of blows, all implicating a history of evolution as the only sensible coordinating explana­tion, should throw in the towel and, like a beaten prizefighter, acknowledge Darwin as the Muhammad Ali of biology.

DISCORDANCE (DISSONANCE OF ONE). Consilience works as a cumulative argument for inferring history from objects and phenomena, rather than di­rectly from sequences. You develop a line of attack, list numerous points, and then close in for the kill. But the empirical world often fails to provide such a bounty of evidence. Often, scientists must reason from a single object or situ­ation — just the thing itself, not a network of arguments suitable for a broad consilience. Can history be inferred from such minimal information?

Thinkers, like soldiers, often show their true mettle in greatest adversity. I am particularly attracted by Darwin's approach to method 4, and have often cited his arguments in these “worst cases” as my primary illustration of his genius (Gould, 1986) — for Darwin met his greatest difficulty, and then not only devised a resolution, but also developed an argument of power and range. In other words, he turned potential trouble into one of his greatest strengths.

To infer history from a single object, Darwin asserts, one must locate fea­tures (preferably several, so the argument may shade into method three) that make no sense, or at least present striking anomalies, in the current life of the organism. One must then show that these features did fit into a clearly inferable past environment. In such cases, history — as expressed by preserva­tion of signs from the past — provides the only sensible explanation for mod­ern quirks, imperfections, oddities, and anomalies.

Darwin structured the Origin of Species as a trilogy. The first four chapters lay out the basic argument for natural selection. The middle five treat dif­ficulties with the theory, and ancillary subjects that must be incorporated or explained away (rules of variation, nature of geological evidence, instincts, hybridism, and general objections). The final five chapters present the grand consilience by summarizing evidence for evolution itself — not so much for natural selection as a mechanism — from a broad range of disparate fields: ge­ology*, geographic variation, morphology, taxonomy, embryology, and so forth.

The last part of the trilogy features method four. One might almost say that chapters 10-14 constitute one long list of examples for inferring history  {112}  from the oddities and imperfections of modern objects. (This arrangement of the last part struck me with particular force, as I reread the Origin before writing this book, and realized that the introductory paragraph for almost ev­ery new subject — from geographic variation to rudimentary organs — explic­itly restates the general argument for method four.) Of course, the rest of the Origin also abounds with cases of method four, beginning as usual with ex­amples from domestication. (Darwin argues that the chicks of wildfowl hide in grass and bushes to give their mother an opportunity for escape by flight. Domesticated chickens retain this habit, which no longer makes sense “for the mother-hen has almost lost by disuse the power of flight” — p. 216.)

Of subjects treated in this final part of the Origin's trilogy, rudimentary or­gans represent, almost by definition, the “holotype” of method four. Darwin's definition, in the first sentence of his discussion, emphasizes this theme — “organs or parts in this strange condition, bearing the stamp of inutility” (p. 450). Nature tries to give us a history lesson, Darwin argues in some frus­tration, but we resist the message as inconsistent with received wisdom about natural harmony: “On the view of each organic being and each separate or­gan having been specially created, how utterly inexplicable it is that parts, like the teeth in the embryonic calf or like the shrivelled wings under the sol­dered wing-covers of some beetles, should thus so frequently bear the plain stamp of inutility! Nature may be said to have taken pains to reveal, by rudi­mentary organs and by homologous structures, her scheme of modification, which it seems that we wilfully will not understand” (p. 480). What else but imprints of history can explain rudimentary organs? Darwin ridicules the special pleading of creationist accounts as fancy ways of saying nothing at all. “In works on natural history rudimentary organs are generally said to have been created 'for the sake of symmetry,' or in order 'to complete the scheme of nature;' but this seems to me no explanation, merely a restatement of the fact. Would it be thought sufficient to say that because planets revolve in elliptic courses round the sun, satellites follow the same course round the planets, for the sake of symmetry, and to complete the scheme of nature?” (p. 453). Always searching for analogies with a short-term human history that we cannot deny, Darwin compares rudimentary organs with silent let­ters, once sounded, in the orthography of words: “Rudimentary organs may be compared with the letters in a word, still retained in the spelling, but be­come useless in the pronunciation, but which serve as a clue in seeking for its derivation” (p. 455).

Darwin continues the same argument as an underpinning for all discus­sions on other aspects of organic form. He introduces morphology as “the most interesting department of natural history, [which] may be said to be its very soul” (p. 434) and continues immediately with an example of method four: “What can be more curious than that the hand of a man, formed for grasping, that of a mole for digging, the leg of the horse, the paddle of the porpoise, and the wing of the bat, should all be constructed on the same pat­tern, and should include the same bones, in the same relative positions” (p. 434).

Similarly, the section on embryology begins with an example of method  {113}  four — the branchial circulation in young bird and mammalian embryos as indications of a “community of descent” with an aquatic past. This common condition in embryonic frogs, birds, and mammals cannot reflect design for current function: “We can not, for instance, suppose that in the embryos of the vertebrata the peculiar loop-like course of the arteries near the branchial slits are related to similar conditions, — in the young mammal which is nour­ished in the womb of his mother, in the egg of the bird which is hatched in a nest, and in the spawn of a frog under water” (p. 440).

The key argument of the section on taxonomy makes the same point in a different form: if animals had experienced no history of change, and were cre­ated in accord with current needs and functions, then why should similar an­atomical designs include creatures of such widely divergent styles of life? Dar­win writes, in the opening paragraph of his discussion on taxonomy: “The existence of groups would have been of simple signification, if one group had been exclusively fitted to inhabit the land, and another the water; one to feed on flesh, another on vegetable matter, and so on; but the case is widely differ­ent in nature; for it is notorious how commonly members of even the same subgroup have different habits” (p. 411).

These arguments strike us as most familiar when based on organic form, but fewer evolutionists recognize that method four also under girds Darwin's two chapters on biogeography (11 and 12). Darwin uses dissonance between organism and dwelling place as the coordinating theme of these chapters: the geographic distributions of organisms do not primarily suit their current cli­mates and topographies, but seem to record more closely a history of oppor­tunities for movement. Again, Darwin presents the basic argument in his first paragraph (p. 346): “In considering the distribution of organic beings over the face of the globe, the first great fact which strikes us is, that neither the similarity nor the dissimilarity of the inhabitants of various regions can be ac­counted for by their climatal and other physical conditions.”

Example tumbles upon example throughout these two chapters. Darwin notes that northern hemisphere organisms of subarctic and north temperate climes maintain far closer taxonomic similarity than the current geographic separation of their continents would imply. He therefore interprets these like­nesses as vestiges of history — preserved expressions of the glacial age, when these climatic bands stood further to the north, near the Arctic Circle where all northern continents virtually touch (p. 370). He also finds too much organic similarity for the modern range of climatic differences along lines of longi­tude from north to south poles, and he again implicates the climax of glacial ages as a time of formation (with modern persistence as a vestige), when even a subarctic species might migrate in comfort, on a cold earth, across the equa­tor from north to south along a single line of longitude. Invoking a complex and graphic metaphor for history, Darwin writes of disjunct distributions on opposite hemispheres, and of geographic refugia at high altitudes of lower latitudes between these endpoints:

The living waters may be said to have flowed during one short period from the north and from the south, and to have crossed the equator; but  {114}  to have flowed with greater force from the north so as to have freely in­undated the south. As the tide leaves its drift in horizontal lines, ... so have the living waters left their living drift on our mountain summits, in a line gently rising from the arctic lowlands to a great height under the equator. The various beings thus left stranded may be compared with savage races of man, driven up and surviving in the mountain fastnesses of almost every land, which serve as a record, full of interest to us, of the former inhabitants of the surrounding lowlands (p. 382).

Everyone cites the Galapagos in a virtual catechism about Darwin's evi­dence for evolution, but few biologists can state how he invokes these islands in the Origin. Most textbooks talk about a diversity of finches, each beauti­fully adapted to available resources on different islands, or of variation in tor­toise carapaces from place to place. Both these stories exemplify both diversi­fication and current adaptive value — but Darwin speaks not a word about either case in the Origin!

In fact, Darwin invokes the Galapagos primarily as an extended example of method four applied to biogeography: These islands house many endemic species, necessarily created in situ according to his opponents. But why then should all these endemics bear close relationship with species on the nearby American mainland? A creationist might say that God fits creatures to im­mediate circumstances, and that the Galapagos Islands, located so near America, must resemble America in environment, and therefore be best suited to house species of the same basic design. But now we grasp the beauty of the Galapagos as an almost uncannily decisive natural experiment for the influ­ence of history. These islands do lie close to America, but could scarcely resemble the mainland less in climate, geology and topography — for the Galapagos are volcanic islands in the wake of a cool current that even permits access to the northernmost species of penguin! Therefore, if the Galapagos endemics resemble American species, they must be recording a history of acci­dental transport and subsequent evolutionary change — not similar creations for similar environments. Darwin's brilliant argument deserves citation in extenso:

Here almost every product of the land and water bears the unmistakable stamp of the American continent. There are 26 land birds, and 25 of these are ranked by Mr. Gould as distinct species, supposed to have been created here; yet the close affinity of most of these birds to American spe­cies in every character, in their habits, gestures, and tones of voice, was manifest…  why should this be so? Why should the species, which are supposed to have been created in the Galapagos Archipelago, and no­where else, bear so plain a stamp of affinity to those created in America? There is nothing in the conditions of life, in the geological nature of the islands, in their height or climate, or in the proportions in which the sev­eral classes are associated together, which resembles closely the condi­tions of the South American coast: in fact there is considerable dissimi­larity in these respects. On the other hand, there is a considerable degree  {115}  of resemblance in the volcanic nature of the soil, in climate, height, and size of islands, between the Galapagos and Cape de Verde Archipelagos: but what an entire and absolute difference in their inhabitants! The in­habitants of the Cape de Verde Islands are related to those of Africa, like those of the Galapagos to America. I believe this grand fact can receive no sort of explanation on the ordinary view of independent creation; whereas on the view here maintained, it is obvious that the Galapagos Is­lands would be likely to receive colonists . . . from America; and the Cape de Verde Islands from Africa; and that such colonists would be lia­ble to modifications — the principle of inheritance still betraying their original birth place (pp. 397-399).

Finally, in rereading the Origin, I was struck by another, quite different, use of the argument from imperfection — one that had entirely escaped my notice before. Darwin showed little sympathy for our traditional and venerable at­tempts to read moral messages from nature. He almost delighted in noting that natural selection unleashes a reign of terror that would threaten our moral values if we tried — as we most emphatically should not — to find ethical guidelines for human life in the affairs of nature. But I hadn't realized that he sometimes presents the apparent cruelties of nature as imperfections pointing to evolution by natural selection — imperfections relative to an inappropriate argument about morality to be sure, but imperfections that trouble our souls nonetheless, and may therefore operate with special force as suggestive argu­ments for evolution:

Nor ought we to marvel if all the contrivances in nature be not, as far as we can judge, absolutely perfect; and if some of them be abhorrent to our ideas of fitness. We need not marvel at the sting of the bee causing the bee's own death; at drones being produced in such vast numbers for one single act, and being then slaughtered by their sterile sisters; at the astonishing waste of pollen by our fir trees; at the instinctive hatred of the queen bee for her own fertile daughters; at ichneumonidae feeding within the live bodies of caterpillars; and at other such cases. The won­der indeed is, on the theory of natural selection, that more cases of the want of absolute perfection have not been observed (p. 472).

I may have burdened readers with too much detail about Darwin's argu­ments for inferring history, but method inheres in this extended madness. My general argument holds that the Origin should be understood as a book en­compassing two opposite, but complementary, poles of science at its best and most revolutionary — first, as a methodological treatise proving by example that evolution can be tested and studied fruitfully; and second, as an intellec­tual manifesto for a new view of life and nature. As a methodological treatise, the Origin focuses upon the palpable and the small — arguing that uniformitarian extrapolation into geological scales can render all evolution. We may therefore avoid any appeal to “higher” forces that cannot be studied directly because they work only in the untestable immensity of deep time, or occur so  {116}  rarely that we can entertain little hope for direct observation during the short span of human history. The disabling Lamarckian paradox — what is impor­tant can't be studied; and what can be studied isn't important — therefore dis­appears, and evolution becomes, under Darwin's system, a working science for the first time. These features of methodology potentiate Darwin's theoreti­cal overview (as we shall see in the next section), and therefore contribute in­dispensably to what may legitimately be called the essence of Darwinism, the sine quibus non for a Darwinian view of nature. This book argues that we can define such a set of basic commitments, but then maintains that these commitments have become inadequate in our times.

Darwin as a Philosophical Revolutionary


Darwin and William Paley

In November 1859, just a week before the official publication date of the Ori­gin, Darwin wrote to his neighbor John Lubbock* “I do not think I hardly ever admired a book more than Paley's 'Natural Theology.' I could almost formerly have said it by heart” (in F. Darwin, 1887, volume 2, p. 219).

The Reverend James McCosh receives my vote for the most interesting among a largely forgotten group of late 19th century thinkers who played a vital role in their own time — liberal theologians friendly to evolution (though not usually to Darwin's philosophy), and who prove that if any warring camps can be designated in this realm, the combatants surely cannot be la­beled as science vs. religion (see Gould, 1999b), but rather as expressions of a much deeper struggle between tradition and reform, or dogmatics and open­ness to change. McCosh doesn't even merit a line in the Encyclopedia Britannica, though he did serve as president of Princeton University, where he had a major influence on the career of Henry Fairfield Osborn and other important American evolutionists.

In 1851, McCosh published an article entitled “Typical Forms” in the North British Review. Hugh Miller, the self-taught Scottish geologist and gen­eral thinker, called this article “at once the most suggestive and ingenious which we have almost ever perused,” and urged McCosh to expand his argu­ment to an entire volume. McCosh accepted this advice and, in collaboration with George Dickie, published Typical Forms and Species Ends in Creation in 1869. The Greek inscription on the title page — typos kai telos (type and purpose)  {117}  — epitomizes the argument. McCosh holds that God's order and benev­olence may be inferred from two almost contradictory properties that reside in tension within all natural objects — “the principle of order” and “the prin­ciple of special adaptation.” (These two principles persist in Darwin's formu­lation under the names “Unity of Type” and “Conditions of Existence” — 1859, p. 206, for example (see my extensive treatment of this passage on pp. 251260), where their fundamental character merits upper case designa­tions from Darwin.) McCosh defines his first principle as “a general plan, pattern, or type, to which every given object is made to conform”; and his second as a “particular end, by which each object, while constructed after a general model is, at the same time, accommodated to the situation which it has to occupy, and a purpose which it is intended to serve” (1869, p. 1). (If we call these two principles “anatomical ground plan” and “adaptation” we will be able to make the appropriate evolutionary translation without difficulty.)

McCosh argues that God's existence and benevolence can be inferred from either principle — from the first by the order of taxonomy, and the abstract beauty of bodily symmetry and structure; and from the second, by “adapta­tion,”* or the exquisite fit of form to function. McCosh also notes that the second, or functional, argument constitutes the “national signature” of Brit­ish thought: “The arguments and illustrations adduced by British writers for the last age or two in behalf of the Divine existence, have been taken almost exclusively from the indications in nature of special adaptation of parts” (1869, p. 6).

The main lineage of this national tradition for “natural theology” based on the “argument from design” runs from Robert Boyle's Disquisition About the Final Causes of Natural Things (1688) and John Ray's Wisdom of God Mani­fested in the Works of the Creation (1691) in Newton's generation that pro­mulgated what historians call “the scientific revolution”; to a grand culmina­tion in William Paley's Natural Theology (1802), one of the most influential books of the 19th century; to an anticlimax, during the 1830's, in the eight “Bridgewater Treatises” (including volumes by Buckland and Whewell), es­tablished by a legacy from the deceased Earl of Bridgewater for a series of volumes “on the power, wisdom, and goodness of God, as manifested in the creation.” Critics in Darwin's circle generally referred to this series as the “bilgewater treatises.”

Revolutions usually begin as replacements for older certainties, and not as pristine discoveries in uncharted terrain. In understanding the second pole of Darwin's genius as the uncompromising radicalism of his new philosophy for life and history, we must first characterize the comfortable orthodoxy  {118}  uprooted by the theory of natural selection. Darwin's essential argument begins with a definition of the dominant philosophy for natural history in his day — natural theology in the Paleyan mode.

At the outset of Chapter 4, I will say more about Paley and the alternative vision of continental natural theology (adaptationism vs. laws of form). For now, a simple statement of the two chief precepts of Paleyan biology will suffice:

Natural theology in general. The rational and harmonious con­struction of nature displays the character and benevolence of a creating God. In the last four chapters of his book, Paley tells us what we may infer about God from the works of creation. God's existence, of course, shines forth in his works, but this we know from many other sources. More specifically (and with a Paleyan chapter for each), nature instructs us about God's personality, his natural attributes, his unity, and (above all) his goodness.

Paley's particular version of natural theology. Natural theology has been expressed in two basic modes (see Chapter 4), one primar­ily continental (laws of form), the other mainly British (adaptationism). Paley held that God manifests his creating power in the exquisite design of organ­isms for their immediate function. We all know Paley's famous opening meta­phor: if I find a watch lying abandoned on an open field, I can conclude from the complex set of parts, all shaped to a common purpose and all well de­signed for a specific end, that some higher intelligence constructed the watch both directly and for a particular use. Since organisms show even more com­plexity and even more exquisite design, they must have been fashioned by an even greater intelligence. But fewer biologists know Paley's more specific ar­gument against the alternative version of natural theology (laws of form), as presented in his chapter 15 on “relations.” The parts of organisms exist in concert not because laws of form or symmetry demand one feature to balance another, but “from the relation which the parts bear to one another in the prosecution of a common purpose” (1803 edition, p. 296) — that is, to secure an optimal adaptation of the whole.

At the very outset of the Origin, Darwin tells us that his explanation of evolution will stress the Paleyan problem of exquisite adaptation. He writes, in the Introduction, that we could obtain sufficient confidence about evo­lution by “reflecting on the mutual affinities of organic beings, on their embryological relations, their geographical distribution, geological succession, and other such facts” (1859, p. 3). “Nevertheless,” he continues, “such a conclusion, even if well founded, would be unsatisfactory, until it could be shown how the innumerable species inhabiting this world have been modi­fied, so as to acquire that perfection of structure and coadaptation which most justly excites our admiration” (1859, p. 3). The explanation of adapta­tion therefore stands forth as the primary problem of evolution. Many lines of evidence prove that evolution occurred. But if we wish to learn how evolu­tion works, we must study adaptation.

This basic Darwinian argument operates as a close copy of Paley's defense, recast in evolutionary language, for the English alternative in natural theology.  {119}  We can infer, Paley often states, that God exists from innumerable as­pects of nature. But if we wish to know any more about the creator — his nature, his attributes, his intentions — we must study the excellence of adapta­tion via the “argument from design.” Paley writes (1803, p. 60): “When we are enquiring simply after the existence of an intelligent Creator, imperfec­tion, inaccuracy, liability to disorder, occasional irregularities, may subsist, in a considerable degree, without inducing any doubt into the question.”

On the other hand, adaptation in the fashioning of contrivances for definite ends reveals God's nature. Paley invokes this theme as a litany in developing his initial parable of the watch and watchmaker. He cites other possible ex­planations for the origin of the watch, and then intones, after each: “Contriv­ance is still unaccounted for. We still want a contriver” (“want,” that is, in the old sense of “lack,” not the modern “desire” — p. 13). “Contrivance must have had a contriver, design, a designer” (p. 14). Later, he tells us explicitly that nature can testify to God's character and goodness only by the phenome­non of adaptation (pp. 42-43): “It is only by the display of contrivance, that the existence, the agency, the wisdom of the Deity, could be testified to his ra­tional creatures. This is the scale by which we ascend to all the knowledge of our Creator which we possess, so far as it depends upon the phenomena, or the works of nature ... It is in the construction of instruments, in the choice and adaptation of means, that a creative intelligence is seen. It is this which constitutes the order and the beauty of the universe.”

I had never read Natural Theology straight through before pursuing my research for this book. In so doing, I was struck by the correspondences be­tween Paley's and Darwin's structure of argument (though Darwin, of course, inverts the explanation). Darwin did not exaggerate when stating to Lubbock that he had virtually committed Paley to memory. The style of Darwin's argu­ments, his choice of examples, even his rhythms and words, must often reflect (perhaps unconsciously) his memory of Paley. Consider just a few examples of this crucial linkage:

1. Paley, like Darwin, relies upon comparison and extrapolation from arti­ficial to natural. Darwin moves from artificial to natural selection, Paley from human to animal machines. Both rely on the central argument that a common mechanism works much more powerfully in nature. Paley's words recall Dar­win's argument that natural selection, working on all parts for so much time, must trump artificial selection, which only affects the few features we choose to emphasize in the short duration of human history. “For every indication of contrivance, every manifestation of design, which existed in the watch, exists in the works of nature; with the difference, on the side of nature, of be­ing greater and more, and that in a degree which exceeds all computation” (1803, p. 19).

2. Both men invoke the same examples. Paley compares the eye and telescope; Darwin lauds the eye as the finest example of complex natural design, and then presents an evolutionary explanation. Paley cites the swimbladder as an independent device created for life in water; Darwin illustrates homology with the tetrapod lung and proposes an evolutionary passage.  {120} 

3. Darwin often uses Paley's logic, sometimes against his predecessor. Paley, for example, dismisses arguments about “tendencies to order” or “principles of design” as empty verbiage, explaining nothing; a true cause must be identi­fied, namely God himself. Darwin makes the same point, but cites evolution as the true cause, while branding statements about creation ex nihilo as empty verbiage. Paley writes (p. 76): “A principle of order is the word: but what is meant by a principle of order, as different from an intelligent Creator, has not been explained either by definition or example: and, without such ex­planation, it should seem to be a mere substitution of words for reasons, names for causes.”

4. Paley discusses many themes of later and central importance to Darwin. He criticizes the major evolutionary conjectures of his day, including Buff on “interior molds,” and the idea of use and disuse. (Since I doubt that he had read Lamarck's earliest evolutionary work by 1802, Paley probably derived this aspect of Lamarck's theory from its status as folk wisdom in general cul­ture.) Paley also states the following crisp epitome of the very argument from Malthus that so struck Darwin. (I am not claiming that this passage provided a covert source for Darwin's central insight. Darwin, after all, had also read Malthus.) “The order of generation proceeds by something like a geometrical progression. The increase of provision, under circumstances even the most advantageous, can only assume the form on an arithmetic series. Whence it follows, that the population will always overtake the provision, will pass be­yond the line of plenty, and will continue to increase till checked by the dif­ficulty of procuring subsistence” (p. 540).

This influence, and this desire to overturn Paley, persisted throughout Darwin's career. Ghiselin (1969), for example, regards Darwin's orchid book as a conscious satire on Paley's terminology and argument. Darwin called this work (1862), his next book after the Origin of Species, “On the various con­trivances by which British and foreign orchids are fertilized by insects.” Paley used the word “contrivance,” as my previous quotations show, to designate an organic design obviously well-made by an intelligent designer. But Darwin argues that orchids must be explained as contraptions, not contrivances. Their vaunted adaptations are jury-rigged from ordinary parts of flowers, and must have evolved from such an ancestral source; the major adaptive fea­tures of orchids have not been expressly and uniquely designed for their cur­rent functions.

Now suppose, as a problem in abstract perversity, that one made a pledge to subvert Paley in the most radical way possible. What would one claim? I can imagine two basic refutations. One might label Paley's primary observa­tion as simply wrong — by arguing that exquisite adaptation is relatively rare, and that the world is replete with error, imperfection, misery and caprice. If God made such a world, then we might want to reassess our decision to wor­ship him. An upsetting argument indeed, but Darwin chose an even more rad­ical alternative.

With even more perversity, one might judge Paley's observation as un­doubtedly correct. Nature features exquisite adaptation at overwhelming relative  {121}  frequency. But the unkindest cut of all then holds that this order, the very basis of Paley's inference about the nature of God, arises not directly from omnipotent benevolence, but only as a side-consequence of a causal principle of entirely opposite import — namely, as the incidental effect of or­ganisms struggling for their own benefit, expressed as reproductive success. Could any argument be more subversive? One accepts the conventional ob­servation, but then offers an explanation that not only inverts orthodoxy, but seems to mock the standard interpretation in a manner that could almost be called cruel. This more radical version lies at the core of Darwin's argu­ment for natural selection. (Darwin actually employed both versions of the radical argument against Paley, but for different aspects of his full case. He in­voked oddities and imperfections as his major evidence for the factuality of evolution (see pp. 111116). But he used the more radical version — exqui­site adaptation exists in abundance, but its cause inverts Paley's world — to construct his mechanism for evolutionary change, the theory of natural selection.)

We all understand, of course, that the force of Darwin's radicalism extends well beyond the inversion of an explanatory order; he also undercut a pri­mary source of human comfort and solace. This book cannot address such a vital issue at any depth, but I must record the point — for this wrenching be­came so salient in subsequent human history. If the natural footprints of Paley's God — the source of our confidence in his character, his goodness and, incidentally, the only hint from nature that we should accept other revealed doctrines, in particular the idea of bodily resurrection (1803, pp. 580-581) — must be reconceived as epiphenomena of a struggle for personal success, then what becomes of nature's beauty, instruction and solace? What a bitter cup Darwin offers us, compared with Paley's sweet promise (1803, pp. 578-579): “The hinges in the wings of an earwig, and the joints of its antennae, are as highly wrought, as if the Creator had had nothing else to finish. We see no signs of diminution of care by multiplication of objects, or of distraction of thought by variety. We have no reason to fear, therefore, our being forgotten, or overlooked, or neglected.”

But then, the man who served as the primary focus of Paley's venera­tion had also promised that the truth would make us free; and Darwin justly argued that nature cannot provide the source of morality or comfort in any case.

Darwin and Adam Smith

Many scientists fail to recognize that all mental activity must occur in social contexts, and that a variety of cultural influences must therefore impact all scientific work. Those who do note the necessary link usually view cultural embeddedness as an invariably negative component of inquiry — a set of bi­ases that can only distort scientific conclusions, and that should be identified for combat. But cultural influences can also facilitate scientific change, for in­cidental reasons to be sure, but with crucially positive results nonetheless — the exaptive principle that evolutionists, above all, should grasp and honor!  {122}  The origin of Darwin's concept of natural selection provides my favorite ex­ample of cultural context as a promoter.

The link of Darwin to Malthus has been recognized and accorded proper importance from the start, if only because Darwin himself had explicitly noted and honored this impetus. But if Darwin required Malthus to grasp the central role of continuous and severe struggle for existence, then he needed the related school of Scottish economists — the laissez-faire theorists, centered on Adam Smith and the Wealth of Nations (first published in the auspicious revolutionary year of 1776) — to formulate the even more fundamental prin­ciple of natural selection itself. But the impact of Adam Smith's economics did not strike Darwin with the force of eureka; the concepts crept upon him in the conventional fashion of most influences upon our lives. How many of us can specify a definite parental admonition, or a particular taunt of our peers, as central to the construction of our deepest convictions?

Silvan S. Schweber (1977), a physicist and historian of science, has traced the chain of influence upon Darwin from Adam Smith's school of Scottish economists — beginning in the early 1830's, and culminating in Darwin's in­tense study of these ideas as he tried to fathom the role of individual action during the weeks just preceding his “Malthusian” insight of September 1838. I believe that Schweber has found the key to the logic of natural selection and its appeal for Darwin in the dual role of portraying everyday and palpable events as the stuff of all evolution (the methodological pole), and in overturn­ing Paley's comfortable world by invoking the most radical of possible argu­ments (the philosophical pole).

In fact, I would advance the even stronger claim that the theory of natural selection is, in essence, Adam Smith's economics transferred to nature. We must also note the delicious (and almost malicious) irony residing in such an assertion. Human beings are moral agents and we cannot abide the heca­tomb* — the death through competition of nearly all participants — incurred by allowing individual competition to work in the untrammeled manner of pure laissez-faire. Thus, Adam Smith's economics doesn't work in economics. But nature need not operate by the norms of human morality. If the adapta­tion of one requires the deaths of thousands in amoral nature, then so be it. The process may be messy and wasteful, but nature enjoys time in abun­dance, and maximal efficiency need not mark her ways. (In one of his most fa­mous letters, Darwin wrote to Joseph Hooker in 1856: “What a book a devil's chaplain might write on the clumsy, wasteful, blundering, low, and horribly cruel works of nature!”) The analog of pure laissez-faire can and does operate in nature — and Adam Smith's mechanism therefore enjoys its  {123}  finest, perhaps its only, full application in this analogous realm, not in the do­main that elicited the original theory itself.

The primary argument of laissez-faire rests upon a paradox. One might suppose that the best path to a maximally ordered economy would emerge from an analysis conducted by the greatest experts all assembled, and given full power to execute their recommendations (the closest human analog to Paley's lone Deity), followed by the passage of laws to implement these ratio­nally-derived, higher-level decisions. Yet Adam Smith argued that a society should follow the opposite path as a best approach to this desired end: law makers and regulators should step aside and allow each individual to struggle for personal profit in an untrammeled way — a procedure that would seem to guarantee the opposite result of chaos and disorder. In allowing the mecha­nism of personal struggle to run freely, good performers eliminate the less ef­ficient and strike a dynamic balance among themselves. The “fallout,” for so­ciety, yields a maximally ordered and prosperous economy (plus a hecatomb of dead businesses). The mechanism works by unbridled struggle for personal reward among individuals.

Schweber documents numerous sources in Darwin's wide readings for this central theme of political economy. In May 1840, for example, Darwin encountered the following passages in J. R. McCulloch's Principles of Political Economy (2nd edition of 1830 — see Schweber, 1980, p. 268):

Every individual is constantly exerting himself to find out the most ad­vantageous methods of employing his capital and labor. It is true, that it is his own advantage, and not that of society, which he has in view; but a society being nothing more than a collection of individuals, it is plain that each, in steadily pursuing his own aggrandisement, is following that precise line of conduct which is most for the public advantage (p. 149). The true line of policy is to leave individuals to pursue their own inter­ests in their own way, and never to lose sight of the maxim pas trop gouverner [not to govern too much]. It is by this spontaneous and un­constrained . . . effort of individuals to improve their conditions . . . and by them only, that nations become rich and powerful (p. 537).

The theory of natural selection lifts this entire explanatory structure, virgo intacta, and then applies the same causal scheme to nature — a tough cus­tomer who can bear the hecatomb of deaths required to produce the best pol­ity as an epiphenomenon. Individual organisms engaged in the “struggle for existence” act as the analog of firms in competition. Reproductive success be­comes the analog of profit — for, even more than in human economies, you truly cannot take it with you in nature.

Finally, continuing the analogy, Paley's dethronement follows the most rad­ical path of supreme irony. For, in the ideal laissez-faire economy, all firms (purified in the unforgiving fires of competition) become sleek and well-de­signed, while the entire polity achieves optimal balance and coordination. But no laws explicitly operate to impose good design or overall balance by fiat — none at all. The struggle among firms represents the only causal process at  {124}  work. Moreover, this cause operates at a lower level, and solely for the benefit of individual firms. Only as an incidental result, a side-consequence, does good design and overall balance emerge. Adam Smith, in coining one of the most memorable metaphors in our language, ascribed this process to the ac­tion of an “invisible hand.” In the modern terms of hierarchy theory, we might say that overall order arises as an effect of upward causation from indi­vidual struggle. We may thus gain some clarity in definition, but we can't match the original prose. In his most famous words, Smith wrote in the Wealth of Nations (Book 4, Chapter 2): “He intends only his own gain, and he is in this, as in many other cases, led by an invisible hand to promote an end which was no part of his intention ... I have never known much good done by those who affected to trade for the public good.”

But Paley had assured us, in 500 closely-argued pages, that the analogous features of the natural world — good design of organisms and harmony of ecosystems — not only prove the existence of God, but also illustrate his na­ture, his personality, and his benevolence. In Darwin's importation of Adam Smith's argument, these features of nature become epiphenomena only, with no direct cause at all. The very observations that Paley had revered as the most glorious handiwork of God, the unquestionable proof of his benevolent concern, “just happen” as a consequence of causes operating at a lower level among struggling individuals. And, as the cruelest twist of all, this lower-level cause of pattern seems to suggest a moral reading exactly opposite to Paley's lofty hopes for the meaning of comprehensive order — for nature's in­dividuals struggle for their own personal benefit, and nothing else! Paley's ob­servations could not be faulted — organisms are well designed and ecosystems are harmonious. But his interpretations could not have been more askew — for these features do not arise as direct products of divine benevolence, but only as epiphenomena of an opposite process both in level of action and in­tent of outcome: individuals struggling for themselves alone.

I write this chapter with two aims in mind: first, to explicate the major sources and content of Darwin's argument; and second, to identify the truly essential claims of Darwinism, in order to separate them from a larger set of more peripheral assertions and misunderstandings — so that we can rank and evaluate the role of modern proposals and debates by the depth of their chal­lenge to the central logic of our profession's orthodoxy. To fulfill this second goal, I try to identify a set of minimal commitments required of those who would call themselves “Darwinians.” I argue that this minimal account fea­tures a set of three broad claims and their (quite extensive) corollaries. I then use this framework to organize the rest of this book, for I devote the histori­cal chapters of this first part to pre- and post-Darwinian discussions of the three claims. Then, following a chapter on the construction of the Modern Synthesis as a Darwinian orthodoxy for the twentieth century, I revisit the three claims in the second part, this time by examining modern challenges to their exclusive sway.

By interpreting Darwin's radical theory as a response to Paley (actually an inversion), based on an importation of the central argument from Adam  {125}  Smith's laissez-faire economics, I believe that we achieve our best insight into the essential claims of Darwinism and natural selection. First, and foremost, we grasp the theoretical centrality of Darwin's conclusion that natural selec­tion works through a struggle among individual organisms for reproductive success. Darwin's choice of levels, and his attempted restriction of causality to one level alone, then becomes neither capricious nor idiosyncratic, but, rather, central to the logic of an argument that renders the former “proof” of God's direct benevolence as an epiphenomenon of causal processes acting for apparently contrary reasons at a lower level. Second, we recognize the focal role of adaptation as the chief phenomenon requiring causal explanation — for good design had also set the central problem for English traditions in nat­ural theology, the worldview that Darwin overturned by deriving the same re­sult with an opposite mechanism.

These two principles — the operation of selection on struggling organisms as active agents, and the creativity of selection in constructing adaptive change — suffice to validate the theory in observational and microevolutionary expression. But Darwin nurtured far more ambitious goals (as the forego­ing discussion of his methodology illustrates, see pages 97116): he wished to promote natural selection, by extrapolation, as the preeminent source of evo­lutionary change at all scales and levels, from the origin of phyla to the ebb and flow of diversity through geological time. Thus, the third focal claim in the Darwinian tripod of essential postulates — the extrapolationist premise — holds that natural selection, working step by step at the organismal level, can construct the entire panoply of vast evolutionary change by cumulating its small increments through the fullness of geological time. With this third premise of extrapolation, Darwin transfers to biology the uniformitarian commitments that set the worldview of his guru, the geologist Charles Lyell.


Once the syllogistic core,* the “bare bones” mechanism of natural selection, has been elucidated, two major questions — the foci of the next two sections  {126}  of this chapter — must be resolved before we can understand the theory's basic operation: the issues of agency and efficacy. The basic historical context of selection — its discovery and utilization by Darwin as a refutation of Paleyan natural theology through the imported causal structure of Adam Smith's in­visible hand — grants primacy to the issue of agency (therefore treated here in the first of two sections on fundamental attributes). The rebuttal of the for­mer centerpiece of natural history — the belief that organic designs record the intentions of an omnipotent creative power — rests upon the radical demotion of agency to a much lower level, devoid of any prospect for conscious intent, or any “view” beyond the immediate and personal. So Darwin reduced the locus of agency to the lowest level that the science of his day could treat in a testable and operational way — the organism (for ignorance of the mechanism of heredity precluded any possibility of still further reduction to cellular or genic levels). The purely abstract statement of natural selection (the syllogistic core) leaves the key question of agency entirely unanswered. Selection may be in control, but on what does selection act? On the subcellular components of heredity? on organisms? on populations? on species? or on all these levels si­multaneously?

Darwin grasped with great clarity what most of his contemporaries never understood at all — that the question of agency, or levels of selection, lies at the heart of evolutionary causation. And he provided, from the depth of his personal convictions, the roots of his central premises, and the logic of his complete argument, a forthright answer that overturned a conceptual world — natural selection works on organisms engaged in a struggle for per­sonal success, as assessed by the differential production of surviving off­spring.

We all know that Darwin emphasized selection at the organismal level, but many evolutionists do not appreciate the centrality of this claim within his theory; nor do they recognize how actively he pursued its defense and illustration.  {127}  To explicate this issue, we must reemphasize the roles of William Paley and Adam Smith in the genesis of Darwin's system — using Smith to overturn Paley.

Adaptation and the “creativity of natural selection,” as discussed in the next section, represent Darwin's evolutionary translation of Paley's chief con­cern with excellence in organic design. But the substitution of natural selec­tion for God as creative agent, while disruptive enough to Western traditions, does not express the primary feature of Darwin's radicalism. To find this root, we must pursue a different inquiry about the locus of selection. After all, se­lection might operate at the highest level of species, even communities of spe­cies, for the direct production of order and harmony. We would then, to be sure, need to abandon God's role as an immediate creator, but what a gentle dispensation compared with Darwin's actual proposal: for if the agency of se­lection stood so high, God could be reconceptualized as the loving instigator of the rules. And the rules, by working directly for organic harmony, would then embody all that Paley sought to illustrate about God's nature.

Darwin's inversion of Paley therefore required a primary postulate about the locus of selection. Selection operates on organisms, not on any higher collectivity. Selection works directly for the benefit of organisms only, and not for any larger harmony that might embody God's benevolent intent. Ironically, through the action of Adam Smith's invisible hand, such “higher harmony” may arise as an epiphenomenal result of a process with appar­ently opposite import — the struggle of individuals for personal success. Dar­win's revolution demands that features of higher-level phenomenology be ex­plained as effects of lower-level causality — in particular, that the struggle among organisms yield order and harmony in the polity of nature.

Darwin's theory therefore presents, as the primary underpinning for its radical import in philosophy, a “reductionist” account of broadest-scale phe­nomena to a single causal locus at a low level accessible to direct observation and experimental manipulation: the struggle for existence among organisms. Moreover, this claim for organismal agency expresses Darwin's chief desider­atum at each focus of his theory — at the methodological pole for tractability, and at the theoretical pole for reversal of received wisdom. Darwinians have often acknowledged the descriptively hierarchical character of nature — and some commentators have been misled to view Darwinism, for this reason, as hierarchical in mechanism of causal action as well. But Darwinism tries to ex­plain all these levels by one locus of causality — selection among organisms. Strict Darwinism is a one-level causal theory for rendering nature's hierarchi­cal richness. The major critique of our times, in advocating hierarchical levels of causality, therefore poses a fundamental challenge to an essential postulate of Darwin's system.

Consider four aspects and demonstrations of Darwin's conviction about the exclusivity of selection on organisms:

Explicit statements. Darwin did not passively “back in” to a claim for the organismic level as a nearly exclusive locus. He knew exactly what he had asserted and why — and he said so over and over again. Statements that  {128}  selection works “for the good of individuals” recur, almost in catechistic form, throughout the Origin: “Natural selection will never produce in a be­ing anything injurious to itself, for natural selection acts solely by and for the good of each (p. 201) . . . Natural selection acts only by the accumulation of slight modifications of structure or instinct, each profitable to the individual under its conditions of life” (p. 233). Even if higher-level order arises as a re­sult, the causal locus must be recognized as individual benefit: “In social animals [natural selection] will adapt the structure of each individual for the benefit of the community; if each in consequence profits by the selected change” (p. 87).

Several other statements illustrate Darwin's emphasis on struggle among organisms, and his desire to avoid all implication that members of a species might amalgamate to collectivities functioning as units of selection in them­selves. He continually stresses, for example, that competition tends to be more intense among members of a single species than between individuals of different species — thus emphasizing the difficulty of forming such collectivi­ties. Moreover, Darwin's development of the theory of sexual selection, and his increasing reliance on this mechanism as his views matured, also forestalls any temptation to advocate group selection — as no form of intraspecific com­petition can be more intense than struggle among similar individuals for per­sonal success in mating.

Response to challenges in the Origin. The primary commit­ments of a theory lie best revealed, not so much in the initial exposition of their logic, but in their later employment to resolve difficulties and para­doxes. Darwin devotes much more of the Origin than most readers have gen­erally realized to defending his single-level theory of selection on organisms.

Darwin structured the Origin as a trilogy — a first part (4 chapters) on the exposition of natural selection, a last section (5 chapters) on the evidence for evolution, and a middle series of 5 chapters on difficulties and responses. Two chapters of this middle section treat a broad range of potential challenges to the creativity of selection and its sequelae — chapter 9 on the geological re­cord (to defend gradualism in the face of apparently contradictory evidence), and chapter 5 on laws of variation (to assert the isotropy of variation — see pp. 144146). A third (chapter 6) treats general “Difficulties on Theory,” mostly centered on gradualism.

Darwin therefore devotes only two of these five chapters, 7 on “Instinct” and 8 on “Hybridism,” to specific difficulties — that is, to issues of sufficient import in his mind to merit such extensive and exclusive treatment. Readers have not always discerned the common thread between these two chapters — Darwin's defense of struggle among organisms as the locus of selection. The chapter on hybridism presents, as its central theme, an argument against spe­cies selection as the cause of sterility in interspecific crosses. The chapter on instinct treats the more general subject of selection's application to behavior as well as to form, but Darwin devotes more than half of this chapter to social insects, and he presents his primary examples of differentiation among castes and sterility of workers as threats to the principle of selection on organisms.  {129} 

Darwin raises two separate challenges to natural selection for the case of sterile castes in the Hymenoptera. How, first of all, can sterile castes evolve adaptive differences from queens (and from each other), when individuals of these castes cannot reproduce? If non-reproductive organisms can evolve ad­aptations, mustn't selection then be working at the higher level of colonies as wholes? Darwin answers, by analogy to domesticated animals once again, that differential survival of non-reproductives may still record selection on fertile members of the population. After all, a breeder can improve the dis­tinct form of castrated animals (raised for food or labor), by mating only those fertile individuals that sire non-reproductives with the most advanta­geous traits (as recognized by the correlation of selectable features in parents with different traits in their castrated offspring):

I have such faith in the powers of selection, that I do not doubt that a breed of cattle, always yielding oxen with extraordinarily long horns, could be slowly formed by carefully watching which individual bulls and cows, when matched, produced oxen with the longest horns; and yet no one ox could ever have propagated its kind. Thus I believe it has been with social insects: a slight modification of structure, or instinct, corre­lated with the sterile condition of certain members of the community, has been advantageous to the community: consequently the fertile males and females of the same community flourished, and transmitted to their fer­tile offspring a tendency to produce sterile members having the same modification. And I believe that this process has been repeated, until that prodigious amount of difference between the fertile and sterile females of the same species has been produced, which we see in many social insects (p. 238).

(This quotation illustrates a common source of misunderstanding. Darwin does often use such phrases as “advantageous to the community.” By our later linguistic conventions, such a statement might seem to signify a leaning to group selectionist arguments. But these conventions did not exist in Dar­win's generation. Note how he uses this phrase only as a description of a re­sult. Darwin identifies the causal process yielding this result, in this case and almost every other time he invokes such language, as selection on organisms, with benefit to communities as an epiphenomenal effect.)

The second challenge, the origin of sterility itself, seems more serious — for how could selection, especially in its necessarily gradualistic mode, promote the diminution of reproductive power in individuals? Clearly, the increasingly sterile workers cannot be promoting their own fitness; but their labor may aid their entire nest or hive. Must not the evolution of sterility therefore provide prima facie evidence for group selection, and for the failure of Darwin's argu­ment about the exclusivity of selection on organisms?

Darwin does indeed refer to sterility as “one special difficulty, which at first appeared to me insuperable, and actually fatal to my whole theory” (p. 236). He then offers an explanation, based exclusively on organismal selection and similar to his argument about differences in form between workers and  {130}  reproductives (p. 236): “How the workers have been rendered sterile is a dif­ficulty; but not much greater than that of any other striking modification of structure; for it can be shown that some insects and other articulate animals in a state of nature occasionally become sterile; and if such insects had been social, and it had been profitable to the community that a number should have been annually born capable of work, but incapable of procreation, I can see no very great difficulty in this being effected by natural selection.”

The phrase “profitable to the community” seems to imply group selection but, as argued above, this modern interpretation need not reflect Darwin's in­tent. He did not, after all, know about haplodiploidy, different degrees of relatedness, or parent-offspring conflict. He does not argue here at the locus classicus for modern theories of group selection — altruism defined as the ren­dering of aid (at personal peril or expense) to non-relatives. Rather, he views the hive as a group of cooperating bodies, all tightly related and all generated by the queen. Anything beneficial to the hive fosters the reproductive success of the queen in ordinary natural selection upon her as an individual. The ste­rility of a worker does not differ in principle from the horns of an ox — a trait not found in parents, but produced by selection on parents. A queen that can generate more sterile workers might be favored by selection just as a breeder picks cows that yield castrated oxen with longer horns.

At most, one might hold that Darwin treats the entire hive as an entity — a statement about higher-level selection on the “superorganism” model (see D. S. Wilson and Sober, 1989, and Sober and Wilson, 1998). But here we meet an issue that must be regarded as more linguistic than substantive. Just as Janzen (1977) wishes to identify a clone as a single El (for “evolutionary individual”), and to treat single bodies of rotifers or aphids as parts, so too might Darwin view the bodies in a hive as iterated organs of the whole. Nonetheless, selection acts on the queen as an individual reproducer. The de­terminants of her success undoubtedly include the form and function of her sterile offspring. Natural selection can “get at” a beaver through the form of its dam, or at a bird through the shape of its nest — and we do not talk about selection on the higher-level entity of organism plus product. Why should se­lection not “get at” the queen ant or bee through the conformation of the hive and the function of its members? (See Ruse, 1980, for a parallel argu­ment, in agreement with mine, on Darwin's explanation of hymenopteran castes by organismic selection.)

Darwin takes up a different challenge to the exclusivity of organismic selection in the next chapter on “Hybridism.” Crosses between varieties of a species are usually fertile, but crosses between species are generally sterile, or at least greatly impaired in fecundity. Under the guiding precepts of gradualism and uniformitarian methodology, we must view species as former varieties promoted by selection to the greater difference of true distinctness. But natu­ral selection could not have built sterility in gradual degrees from an original fertility between parent and offspring — for sterility cannot benefit the hybrid individual: “On the theory of natural selection the case is especially impor­tant, inasmuch as the sterility of hybrids could not possibly be of any advantage  {131}  to them, and therefore could not have been acquired by the continued preservation of successive profitable degrees of sterility. I hope, however, to be able to show that sterility is not a specially acquired or endowed quality but is incidental on other acquired differences” (p. 245).

Darwin considers two possible explanations. He constructs his entire chap­ter on hybridism as a defense of natural selection in its ordinary, organismal mode through the rejection of one explanation based on species selection and the advocacy of another rooted in selection on organisms with an interesting twist. Darwin admits that species selection, at first glance, seems to provide a simple and attractive solution: interspecific sterility must originate as an ad­aptation of species, built and promoted to preserve integrity by preventing introgression and subsequent dissolution. (A. R. Wallace strongly promoted this view. Darwin's firm rejection led to a protracted argument that strongly colored their relationship — see Kottler, 1985; Ruse, 1980.)

But Darwin rejected this explanation because he could not conceive how a species might act as an entity in this manner. Nonetheless, he could not pos­sibly argue in response that hybrid sterility arose by direct selection for the trait itself. He therefore proposed a subtle argument, almost surely correct in our current judgment, for the origin of hybrid sterility as an incidental consequence of other differences established by organismal selection. A. R. Wallace, in striking contrast, remained so committed to viewing every natural phenomenon as a direct adaptation that he willingly roamed up and down among levels of selection (quite unaware of the logical difficulties thus en­tailed) until he found a locus that could support a direct adaptive expla­nation.

Darwin argued that any population, in diverging far enough from an an­cestor to rank as a separate species, must undergo a series of changes (usually extensive), mediated by natural selection and leading to a set of unique fea­tures. Any two species will therefore come to differ in a series of traits directly built by natural selection. These disparities will probably render the two spe­cies sufficiently unlike, particularly in rates and modes of reproduction and development, that any hybrids between them will probably be stunted or infertile — not because selection acted directly for sterility, but only as an in­cidental effect of differences evolved by natural selection for other reasons. Although interspecific sterility cannot be built directly by selection for its ad­vantages to organisms, this feature can and will originate as a consequence of ordinary selection on organisms. Darwin contrasts his proposal with Wallace's alternative based on direct adaptation via species selection:

Now do these complex and singular rules indicate that species have been endowed with sterility simply to prevent their becoming confounded in nature? I think not. For why should the sterility be so extremely different in degree, when various species are crossed, all of which we must sup­pose it would be equally important to keep from blending together?  …The foregoing rules and facts, on the other hand, appear to me clearly to indicate that the sterility both of first crosses and of hybrids is simply  {132}  incidental or dependent on unknown differences, chiefly in the reproduc­tive systems, of the species which are crossed (p. 260).

In what I regard as Darwin's most brilliant use of his favorite device — argument by analogy — he then compares hybrid sterility with incompatibility in hybrid grafts (whereas grafts between varieties of the same species usually “take”). I find this comparison particularly compelling because we would not be tempted to construct an argument about species selection to explain the in­compatibility of grafts — as no advantage for the integrity of species accrues thereby, especially since the “experiment” of grafting between two species al­most never occurs in nature. Yet the logical structures of these two arguments about grafting and sterility, as well as the attendant results, share an identical logic — joining within species, and maintenance of separation between spe­cies, based upon incidental effects wrought by increasing degrees of differ­ence evolved for other reasons:

It will be advisable to explain a little more fully by an example what I mean by sterility being incidental on other differences, and not a spe­cially endowed quality. As the capacity of one plant to be grafted or bud­ded on another is so entirely unimportant for its welfare in a state of na­ture, I presume that no one will suppose that this capacity is a specially endowed quality, but will admit that it is incidental on differences in the laws of growth of the two plants . . . The facts by no means seem to me to indicate that the greater or lesser difficulty of either grafting or cross­ing together various species has been a special endowment; although in the case of crossing, the difficulty is as important for the endurance and stability of specific forms, as in the case of grafting it is unimportant for their welfare (pp. 261-263).

Darwin then drives the point home with a lovely prose flourish (and a memo­rable visual image!) in explicitly rejecting an appeal to supraorganismal selec­tion. Nature knows no explicit principle of higher-level order. “There is no more reason to think that species have been specially endowed with various degrees of sterility to prevent them crossing and blending in nature, than to think that trees have been specially endowed with various and somewhat analogous degrees of difficulty in being grafted together in order to prevent them becoming inarched in our forests” (p. 276).

The development of Darwin's views on organismic selec­tion. If the first edition of the Origin only marked a waystation in fluctua­tion or degree of commitment, then Darwin's stand on organismic selection, however strongly expressed in this initiating volume, might not be deemed so central to his worldview. But Ruse (1980) has documented Darwin's continu­ing and increasing attention to this issue — particularly as he argued with Wallace (see also Kottler, 1985) about the principle of incidental effects to ex­plain hybrid sterility as a side consequence of natural selection rather than a direct product of species selection. Ruse writes (1980, p. 620): “By the end of the decade [the 1860's] with respect to the animal and plant worlds, there was  {133}  nothing implicit about Darwin's commitment to individual selection. He had looked long and hard at group selection and rejected it.”

How Darwin struggles with, and “walls off,” exceptions. The exegetical literature on Darwin usually states that he allowed only two exceptions, in the entire corpus of his writing, to the exclusivity of natural se­lection on organisms — first, in permitting some form of group selection for the neuter castes of social insects, and second, for the origin of human moral behavior. I agree with Ruse (see point 2 just above) that Darwin did not stray from his orthodoxy for social insects, though some of his terminological choices invite misinterpretation today. For human morality, on the other hand, Darwin did throw in the towel after long struggle — for he could not render altruism towards non-relatives by organismal selection. Nonetheless, a theory often becomes sharpened (not destroyed or even much compromised in a world of relative frequencies) by specifying a domain of exceptions — pro­vided that the exceptions be rare in occurrence, and peculiar in form. As hu­mans, we surely have a legitimate personal interest in our moral behavior, but we cannot enshrine this property as occupying more than a tiny corner of na­ture (whatever its eventual impact upon our planet, and whatever our paro­chial concern for its uniqueness).

In the Descent of Man, Darwin presents his most interesting and extensive discussion of supraorganismal selection. As an example of his clarity on the issue of levels of selection, consider the following passage on why natural se­lection could not foster altruistic behavior within a tribe — with an explicit final statement that differential success among distinct tribes should not be called natural selection:

But it may be asked, how within the limits of the same tribe did a large number of members first become endowed with these social and moral qualities, and how was the standard of excellence raised? It is extremely doubtful whether the offspring of the more sympathetic and benevolent parents, or of those who were the most faithful to their comrades, would be reared in greater number than the children of selfish and treacherous parents of the same tribe. He who was ready to sacrifice his life, as many a savage has been, rather than betray his comrades, would often leave no offspring to inherit his noble nature ... Therefore it seems scarcely possi­ble (bearing in mind that we are not here speaking of one tribe being vic­torious over another) that the number of men gifted with such virtues, or that the standard of their excellence, would be increased through natural selection, that is, by the survival of the fittest (1871, vol. 1, p. 163).

In the light of this conundrum, and as part of his resolution, Darwin does allow for selection at the tribal level defined as differential success of groups with more altruists: “It must not be forgotten that although a high standard of morality gives but a slight or no advantage to each individual man and his children over the other men of the same tribe, yet that an advancement in the standard of morality, and an increase in the number of well-endowed men  {134}  will certainly give an immense advantage to one tribe over another” (1871, p. 166).

This passage has often been quoted, but without its surrounding context of contrary alternatives and restrictive caveats, as a clean example of Darwin's move to a higher level of selection when required. But such an interpretation seriously misrepresents Darwin's motives and logic. He did make the move, but only as one factor in a surrounding context of mitigation. I regard these mitigations and restrictions to hold the line of organismal selection (ex­pressed in three distinct arguments, discussed below) as far more interesting than the move itself, for Darwin's extreme reluctance to address selection at any level other than the organismic lies so well exposed in the totality.

1. The Descent, as a whole, rests upon the strongest mode of argument for organismal selection. Darwin did not write a separate book on human evo­lution; his ideas (mostly speculative) on this subject occupy the first, and shorter, part of a two volume treatise entitled, in full: The Descent of Man, and Selection in Relation to Sex. In other words, Darwin wrote the Descent as an introduction to his general exposition of sexual selection. We might re­gard the two parts as oddly juxtaposed until we realize that many of Darwin's major arguments about human evolution — in the establishment of secondary sexual characters, and in differentiation among races, for example — invoke sexual selection by intraspecific competition, rather than ordinary natural se­lection as adaptation to external environments. As Ruse (1980) notes, Dar­win viewed sexual selection as the strongest general argument against group selection, for its theme of relentless struggle in mating among members of a population guarantees that individualism must reign, largely by precluding the formation of alliances that higher-level selection could exploit. (Modern notions of sexual selection do envision the formation of such alliances, so the argument may strike us as incorrect today — but Darwin conceived sexual selection as a hyperindividual mode.)

2. Darwin does not present his argument for tribal selection as a happy solution to the problem of morality, but only as one potential factor among oth­ers. He also devises an argument based on organismal selection — in the form that would be called “reciprocal altruism” today: “As the reasoning powers and foresight of the members became improved, each man would soon learn from experience that if he aided his fellow-men, he would commonly receive aid in return. From this low motive he might acquire the habit of aiding his fellows” (1871, p. 163).

3. Darwin presents tribal selection as a peculiarity based on the uniqueness of human consciousness, and thus as a strictly circumscribed exception to the generality of organismal selection throughout living nature. As conscious be­ings, we become especially sensitive to the “praise and blame” of our fellows. If altruistic behavior gains a status as virtuous, then we might be persuaded — against our deeper biological drive for seeking personal advantage — to en­gage in such behaviors in order to foster praise or avoid calumny. In other words, a form of “cultural evolution,” rooted in our unique level of con­sciousness, could overcome the behaviors driven by organismal selection, and  {135}  could establish a preference for altruistic acts that might then serve as a basis for tribal selection. But such an argument cannot enjoy wide application in nature, as all other species lack this special mental mechanism for spreading abstract ideas against the thrust of natural selection:

We may therefore conclude that primeval man, at a very remote period, would have been influenced by the praise and blame of his fellows. It is obvious, that the members of the same tribe would approve of conduct, which appeared to them to be for the general good, and would reprobate that which appeared evil... A man who was not impelled by any deep, instinctive feeling, to sacrifice his life for the good of others, yet was roused to such actions by a sense of glory, would by his example excite the same wish for glory in other men, and would strengthen by exercise the noble feeling of admiration. He might thus do far more good to his tribe than by begetting offspring with a tendency to inherit his own high character (1871, p. 165).

Note also how Darwin, in this passage, explicitly limits within tribal bound­aries the extent of such spread against organismal selection. If some form of group selection had to be acknowledged for a special case, Darwin sought to confine its operation to the smallest aggregation within the species — and then to let these small collectivities struggle with others in a minimal context of groupiness.

Thus, in permitting a true exception to organismal selection, Darwin's pri­mary attitude exudes extreme reluctance — restriction to minimal groupiness, provision of other explanations in the ordinary organismal mode, limitation to a unique circumstance in a single species (human consciousness for the spread of an idea against the force of organismal selection), and placement within a more general argument for sexual selection, the strongest form of the orthodox mode.

In my researches for this book, I made a discovery that strongly supports this view of Darwin's attitude towards supraorganismal selection. I found that the traditional sources (Ruse, Kottler and others) did not identify Dar­win's major, explicit struggle to contain an apparent need for higher-level se­lection, and to assert exclusivity for the organismal mode. He fought a far more important battle with himself on an issue well beyond particular prob­lems raised by single taxa (sterility of worker castes or human morality): the explanation of the principle that he ranked second only to natural selection itself as a component of evolutionary theory — the “principle of divergence.” (Evolutionists have not recognized this important component of Darwin's de­veloping ideas about selection because he excised this discussion as he ab­stracted his longer work to compose the Origin. But the full version exists in the uncompleted manuscript of his intended larger work — edited and pub­lished by Stauffer, 1975, but not widely read by practicing biologists.) More­over, in his long version, Darwin wrestles not with the lowest interdemic level of tribal selection, but with species selection itself. I will present a full exposi­tion in Chapter 3 (pp. 224250), but should mention for now that Darwin's  {136}  tactic closely follows his argument about human morality, and therefore emphasizes his extreme reluctance to embrace supraorganismal selection, and his almost desperate effort to confine explanation to the organismal mode. The recognition that Darwin, despite such strong reluctance, could not avoid some role for species selection, builds a strong historical argument for the ineluctability of a hierarchical theory of selection. (I shall show in Chapter 3 that none of the few 19th century scientists who truly grasped the full range and subtlety of selectionist theory could avoid important roles for levels other than the organismic.)

As with the next topic of creativity for natural selection (pp. 137-159), the issue of levels in selection has resounded through the entire history of evolu­tionary theory, and continues to set a major part of the agenda for modern debate — as it must, for the subject lies (with only a few others) at the very heart of Darwinian logic. Wallace never comprehended the question of levels at all, as he searched for adaptation wherever he could find it, oblivious to any problems raised by the locus of its action; Kropotkin, in asserting mutual aid, never grasped the problem either; Weismann shared Darwin's insight about the problem's fundamental nature, but also came to understand, after a long and explicit intellectual struggle with his own strong reluctance, that ex­clusivity must yield to hierarchy (pp. 197-224).

In our generation, Wynne-Edwards (1962) riled an entire profession by defending the classical form of group selection as a generality, while Williams (1966) penned a powerful rebuttal, urging us all to toe the Darwinian line (see Chapter 7 for a full account). The classical ethologists invoked various forms of group selection (often by default); the sociobiologists proclaimed a revolution by reaction and return to the pure Darwinism of individual advan­tage. Dawkins (1976) attempted an even stronger reduction to exclusivity for genic selection, but his false argument rests on a confusion of bookkeeping with causality, and his own later work (1982) negates his original claim, though Dawkins seems unaware of his own contradictions (see Chapter 8). Supporters of hierarchy theory — I am one, and this is a partisan book — are revising Darwinism into a multilevel theory of selection.

This issue will not go away, and must excite both interest and passion. Nothing else lies so close to the raw nerve of Darwin's radicalism. The exclu­sivity of organismal selection, after all, provides the punch line that allowed the vision of Adam Smith to destroy the explicit beauty and harmony of Wil­liam Paley's world.

Viewed in this light, the Origin's very few statements about solace become particularly revealing. Darwin had just overturned a system that provided the philosophical basis of human comfort for millennia. What could he supply in return, as we continue to yearn for solace in this vale of tears? One might be tempted to read the few Darwinian statements about solace as peculiar, exceptional, even “soft” or illogical. But we should note another feature of these statements as well: they yield no ground whatever on the key issue of organismal struggle. Solace must be found in other guises; the linchpin of se­lection as struggle among organisms cannot be compromised.  {137} 

Darwin offers two sources for solace. First, the struggle, however fierce, usually brings no pain or distress to organisms (humans, with their intrusive consciousness, have introduced a tragic exception into nature). “When we re­flect on this struggle, we may console ourselves with the full belief, that the war of nature is not incessant, that no fear is felt, that death is generally prompt, and that the vigorous, the healthy, and the happy survive and multi­ply” (p. 79).

Second, this struggle does lead to general improvement, if only as an epiphenomenon, and whatever the cost: “As natural selection works solely by and for the good of each being, all corporeal and mental endowments will tend to progress towards perfection” (p. 489). Darwin could never compro­mise his central logic; for even this “softest” of all his statements explicitly as­serts that selection can only work on organisms — “for the good of each be­ing.” And why not? The logic of organismal struggle includes both fierce beauty and empirical adequacy — whatever the psychic costs. And, since roses by other names smell just as sweet, then beauty, even as an epiphenomenon, becomes no less pleasing, and no less a balm for the soul.


The following kind of incident has occurred over and over again, ever since Darwin. An evolutionist, browsing through some pre-Darwinian tome in nat­ural history, comes upon a description of natural selection. Aha, he says; I have found something important, a proof that Darwin wasn't original. Per­haps I have even discovered a source of direct and nefarious pilfering by Dar­win! In the most notorious of these claims, the great anthropologist and writer Loren Eiseley thought that he had detected such an anticipation in the writ­ings of Edward Blyth. Eiseley laboriously worked through the evidence that Darwin had read (and used) Blyth's work and, making a crucial etymological mistake along the way (Gould, 1987c), finally charged that Darwin may have pinched the central idea for his theory from Blyth. He published his case in a long article (Eiseley, 1959), later expanded by his executors into a posthu­mous volume entitled “Darwin and the Mysterious Mr. X” (1979).

Yes, Blyth had discussed natural selection, but Eiseley didn't realize — thus committing the usual and fateful error in this common line of argument — that all good biologists did so in the generations before Darwin. Natural selection ranked as a standard item in biological discourse — but with a crucial differ­ence from Darwin's version: the usual interpretation invoked natural selec­tion as part of a larger argument for created permanency.* Natural selection,  {139}  in this negative formulation, acted only to preserve the type, constant and in­violate, by eliminating extreme variants and unfit individuals who threatened to degrade the essence of created form. Paley himself presents the following variant of this argument, doing so to refute (in later pages) a claim that mod­ern species preserve the good designs winnowed from a much broader range of initial creations after natural selection had eliminated the less viable forms: “The hypothesis teaches, that every possible variety of being hath, at one time or other, found its way into existence (by what cause or in what manner is not said), and that those which were badly formed, perished” (Paley, 1803, pp. 70-71).

Darwin's theory therefore cannot be equated with the simple claim that natural selection operates. Nearly all his colleagues and predecessors ac­cepted this postulate. Darwin, in his characteristic and radical way, grasped that this standard mechanism for preserving the type could be inverted, and then converted into the primary cause of evolutionary change. Natural selec­tion obviously lies at the center of Darwin's theory, but we must recognize, as Darwin's second key postulate, the claim that natural selection acts as the cre­ative force of evolutionary change. The essence of Darwinism cannot reside in the mere observation that natural selection operates — for everyone had long accepted a negative role for natural selection in eliminating the unfit and preserving the type.

We have lost this context and distinction today, and our current perspective often hampers an understanding of the late 19th century literature and its preoccupations. Anyone who has read deeply in this literature knows that no argument inspired more discussion, while no Darwinian claim seemed more vulnerable to critics, than the proposition that natural selection should be viewed as a positive force, and therefore as the primary cause of evolutionary change. The “creativity of natural selection” — the phrase generally used in Darwin's time as a shorthand description of the problem — set the cardinal subject for debate about evolutionary mechanisms during Darwin's lifetime and throughout the late 19th century.

Non-Darwinian evolutionists did not deny the reality, or the operationality, of natural selection as a genuine cause stated in the most basic or abstract manner — in the form that I called the “syllogistic core” on page 125 (still used as the standard pedagogical device for teaching the “bare bones” logic of Darwinism in general and introductory college courses). They held, rather, that natural selection, as a headsman or executioner, could only eliminate the unfit, while some other cause must play the positive role of constructing the fit.

For example, Charles Lyell — whom Darwin convinced about the factuality of evolution but who never (much to Darwin's sadness and frustration) accepted the mechanism of natural selection — admitted that he had be­come stymied on the issue of creativity. He could understand, he wrote in his fifth journal on the “species question” in March 1860, how natural se­lection might act like two members of the “Hindoo Triad” — like Vishnu the preserver and Siva the destroyer, but he simply could not grasp how  {140}  such a force could also work like Brahma, the creator (in Wilson, 1970, p. 369).

E. D. Cope, chief American critic and exponent of neo-Lamarckism, chose a sardonic title to highlight Darwin's supposedly fatal weakness in claiming a creative role for natural selection. He called his book The Origin of the Fittest (1887) — a parody on Darwin's “survival of the fittest,” and a motto for what natural selection could not accomplish. Cope wrote: “The doctrines of  'selec­tion' and 'survival' plainly do not reach the kernel of evolution, which is, as I have long since pointed out, the question of 'the origin of the fittest.' This omission of this problem from the discussion of evolution is to leave Hamlet out of the play to which he has given the name. The law by which structures originate is one thing; those by which they are restricted, directed, or de­stroyed, is another thing” (1887, p. 226).

We can understand the trouble that Darwin's contemporaries experienced in comprehending how selection could work as a creative force when we confront the central paradox of Darwin's crucial argument: natural selec­tion makes nothing; it can only choose among variants originating by other means. How then can selection possibly be conceived as a “progressive,” or “creative,” or “positive” force?

In resolving this paradox, Darwin recognized his logical need, within the basic structure of his argument, to explicate the three main requirements and implications of an argument for selection's creativity: (1) the nature of varia­tion; (2) the rate and continuity of change; (3) the meaning of adaptation. This interrelated set of assertions promotes natural selection from mere exis­tence as a genuine, but secondary and negative, mechanism to domination as the primary cause of evolutionary change and pattern. This set of defenses for selection's creativity therefore ranks as the second of three essential postu­lates, or “minimal commitments” of Darwinian logic.

As the epitome of his own solution, Darwin admitted that his favored mechanism “made” nothing, but held that natural selection must be deemed “creative” (in any acceptable vernacular sense of the term) if its focal action of differential preservation and death could be construed as the primary cause for imparting direction to the process of evolutionary change. Darwin reasoned that natural selection can only play such a role if evolution obeys two crucial conditions: (1) if nothing about the provision of raw materials — that is, the sources of variation — imparts direction to evolutionary change; and (2) if change occurs by a long and insensible series of intermediary steps, each superintended by natural selection — so that “creativity” or “direction” can arise by the summation of increments.

Under these provisos, variation becomes raw material only — an isotropic sphere of potential about the modal form of a species. Natural selection, by superintending the differential preservation of a biassed region from this sphere in each generation, and by summing up (over countless repetitions) the tiny changes thus produced in each episode, can manufacture substantial, directional change. What else but natural selection could be called “creative,” or direction-giving, in such a process? As long as variation only supplies raw  {141}  material; as long as change accretes in an insensibly gradual manner; and as long as the reproductive advantages of certain individuals provide the statisti­cal source of change; then natural selection must be construed as the direc­tional cause of evolutionary modification.

These conditions are stringent; and they cannot be construed as vague, unconstraining, or too far in the distance to matter. In fact, I would argue that the single most brilliant (and daring) stroke in Darwin's entire theory lay in his willingness to assert a set of precise and stringent requirements for varia­tion — all in complete ignorance of the actual mechanics of heredity. Darwin understood that if any of these claims failed, natural selection could not be a creative force, and the theory of natural selection would collapse. We pay our highest tribute to the power of natural selection in recognizing how Darwin used the theory to deduce a set of necessary properties for variation, well be­fore science understood the mechanism of heredity — and in noting that these properties then turned out to be both basically correct and also entailed by the causes later discovered!

The requirements for variation

In order to act as raw material only, variation must walk a tightrope between two unacceptable alternatives. First and foremost, variation must exist in suf­ficient amounts, for natural selection can make nothing, and must rely upon the bounty thus provided; but variation must not be too florid or showy ei­ther, lest it become the creative agent of change all by itself. Variation, in short, must be copious, small in extent, and undirected. A full taxonomy of non-Darwinian evolutionary theories may be elaborated by their denials of one or more of these central assumptions.

COPIOUS. Since natural selection makes nothing and can only work with raw material presented to its stringent review, variation must be generated in copious and dependable amounts (especially given the hecatomb of selec­tive deaths accompanying the establishment of each favorable feature). Dar­win's scenario for selective modification always includes the postulate, usu­ally stated explicitly, that all structures vary, and can therefore evolve. He argues, for example, that if a short beak were favored on a full-grown pigeon “for the bird's own advantage” (p. 87), then selection would also work within the egg for sufficient beak strength to break the shell despite diminu­tion in overall size of the beak — unless evolution followed an alternate route of selection for thinner shells, “the thickness of the shell being known to vary like any other structure” (p. 87).

Darwin's faith in the copiousness of variation can be gauged most clearly by his response to the two most serious potential challenges of his time. First, he acknowledges the folk wisdom that some domestic species (dogs, for ex­ample) have developed great variety, while others (cats, for example) differ far less among populations. If these universally recognized distinctions arise as consequences of differences in the intrinsic capacity of species to vary, then Darwin's key postulate of copiousness would be compromised — for failure of  {142}  sufficient raw material would then be setting a primary limit upon the rate and style of evolutionary change, and selection would not occupy the driver's seat.

Darwin responds by denying this interpretation, and arguing that differing intensities of selection, rather than intrinsically distinct capacities for varia­tion, generally cause the greater or lesser differentiation observed among do­mestic species. I regard this argument as among the most forced and uncom­fortable in the Origin — a rare example of Darwinian special pleading. But Darwin realizes the centrality of copiousness to his argument for the creativ­ity of natural selection, and he must therefore face the issue directly:

Although I do not doubt that some domestic animals vary less than oth­ers, yet the rarity or absence of distinct breeds of the cat, the donkey, peacock, goose, etc., may be attributed in main part to selection not hav­ing been brought into play: in cats, from the difficulty in pairing them; in donkeys, from only a few being kept by poor people and little attention paid to their breeding; in peacocks, from not being very easily reared and a large stock not kept; in geese, from being valuable only for two pur­poses, food and feathers, and more especially from no pleasure having been felt in the display of distinct breeds (p. 42).

Second, copiousness must also be asserted in the face of a powerful argu­ment about limits to variation following modal departure from “type.” To use Fleeming Jenkin's (1867) famous analogy: a species may be compared to a rigid sphere, with modal morphology of individuals at the center, and limits to variation defined by the surface. So long as individuals lie near the center, variation will be copious in all directions. But if selection brings the mode to the surface, then further variation in the same direction will cease — and evo­lution will be stymied by an intrinsic limitation upon raw material, even when selection would favor further movement. Evolution, in other words, might consume its own fuel and bring itself to an eventual halt thereby. This potential refutation stood out as especially serious — not only for threaten­ing the creativity of natural selection, but also for challenging the validity of uniformitarian extrapolation as a methodology of research. Darwin re­sponded, as required by logical necessity, that such limits do not exist, and that new spheres of equal radius can be reconstituted around new modes: “No case is on record of a variable being ceasing to be variable under cultiva­tion. Our oldest cultivated plants, such as wheat, still often yield new varie­ties: our oldest domesticated animals are still capable of rapid improvement or modification” (p. 8).

I cannot here provide a full history for the subsequent odysseys of these key Darwinian precepts. But a few cursory comments indicate how these claims have remained central and contentious throughout the history of post-Dar­winian thought, and how they continue to underlie important debates within Darwinism today.

The argument about copiousness, particularly as expressed in the claim for limits to further variability after intense selection, dogged the 19th century  {143}  literature and emerged as a key issue in the biometrician vs. Mendelian debates early in our century (see Provine, 1971). Castle (1916,1919) pursued his famous experiments on selection in hooded rats in order to test the hy­pothesis of limits imposed by variability upon continued change. One of the most appealing features of Mendelism — and a strong reason for accep­tance following its “rediscovery” in 1900 — lay in the argument that mutation could restore variation “used up” by selection. Nor has the issue abated today. In another form, copiousness underlay the great debate between Dobzhansky and Muller (see Lewontin, 1974) — the classical vs. the balance view in Dobzhansky's terminology. Kimura's (1963, 1983) modern theory of neutralism may be invoked to acknowledge the fact of copiousness while avoiding the pitfalls of genetic load — and therefore becomes “neoclassical” in Lewontin's terminology.

SMALL IN EXTENT. If the variations that yielded evolutionary change were large — producing new major features, or even new taxa in a single step — then natural selection would not disappear as an evolutionary force. Selection would still function in an auxiliary and negative role as headsman — to heap up the hecatomb of the unfit, permit the new saltation to spread among or­ganisms in subsequent generations, and eventually to take over the popula­tion. But Darwinism, as a theory of evolutionary change, would perish — for selection would become both subsidiary and negative, and variation itself would emerge as the primary, and truly creative, force of evolution, the source of occasionally lucky saltation. For this reason, and quite properly, saltationist (or macromutational) theories have always been viewed as anti-Darwinian — despite the protestations of de Vries (see Chapter 5), who tried to retain the Darwinian label for his continued support of selection as a nega­tive force. The unthinking, knee-jerk response of many orthodox Darwinians whenever they hear the word “rapid” or the name “Goldschmidt,” testifies to the conceptual power of saltation as a cardinal danger to an entire theoretical edifice.

Darwin held firmly to the credo of small-scale variability as raw material because both poles of his great accomplishment required this proviso. At the methodological pole of using the present and palpable as a basis, by extrapo­lation, for all evolution, Darwin longed to locate the source of all change in the most ordinary and pervasive phenomenon of small-scale variation among members of a population — Lyell's fundamental uniformitarian principle, re­cast for biology, that all scales of history must be explained by currently ob­servable causes acting within their current ranges of magnitude and intensity. “I believe mere individual differences suffice for the work,” Darwin writes (p. 102). At the theoretical pole, natural selection can only operate in a cre­ative manner if its cumulating force builds adaptation step by step from an isotropic pool of small-scale variability. If the primary source of evolutionary innovation must be sought in the occasional luck of fortuitous saltations, then internal forces of variation become the creative agents of change, and natural selection can only help to eliminate the unfit after the fit arise by some  {144}  other process. Darwin, again using domestication as an analog, passionately defends the central role of variation so small as to pass beneath nearly every­one's notice (p. 32):

If selection consisted merely in separating some very distinct variety, and breeding from it, the principle would be so obvious as hardly to be worth notice; but its importance consists in the great effect produced by the ac­cumulation in one direction, during successive generations, of differ­ences absolutely inappreciable by an uneducated eye — differences which I for one have vainly attempted to appreciate. Not one man in a thou­sand has accuracy of eye and judgment sufficient to become an eminent breeder. If gifted with these qualities, and he studies his subject for years, and devotes his lifetime to it with indomitable perseverance, he will suc­ceed, and may make great improvements; if he wants [that is, lacks] any of these qualities, he will assuredly fail.

Saltational variation has always served as a rallying point for non-Darwin­ian evolutionary argument (see Chapters 4 and 5 for a full discussion). T. H. Huxley centered his own doubts about natural selection firmly upon Dar­win's preference for change by insensible steps. Bateson (1894), in developing the concept of homeosis, and D'Arcy Thompson (1917), in his ideas on non-continuity in certain geometrical transformations, advanced saltation as an explicitly anti-Darwinian argument. The early mutationists read Mendel as a warrant for discontinuous change, and a disproof of strict Darwinism as es­poused by the “biometricians.” Goldschmidt (1940; see Gould, 1982a) joined some interesting views on developmental discontinuity to an untenable genetic theory, all the better to espouse a saltationist view that made him the chief whipping boy of the Modern Synthesis.

Reciprocally, Darwinians countered with strong and explicit support. R. A. Fisher began his great book (1930) by rooting a defense of Darwin in a link­age of copiousness with small-scale variation — specifically, by arguing for an inverse correlation of frequency and effect, and then claiming that varia­tions of large effect therefore become too rare to serve as evolution's raw material.

UNDIRECTED. Textbooks of evolution still often refer to variation as “ran­dom.” We all recognize this designation as a misnomer, but continue to use the phrase by force of habit. Darwinians have never argued for “random” mutation in the restricted and technical sense of “equally likely in all direc­tions,” as in tossing a die. But our sloppy use of “random” (see Eble, 1999) does capture, at least in a vernacular sense, the essence of the important claim that we do wish to convey — namely, that variation must be unrelated to the direction of evolutionary change; or, more strongly, that nothing about the process of creating raw material biases the pathway of subsequent change in adaptive directions. This fundamental postulate gives Darwinism its “two step” character, the “chance” and “necessity” of Monod's famous formula­tion — the separation of a source of raw material (mutation, recombination, etc.) from a force of change (natural selection).  {145} 

In a sense, the specter of directed variability threatens Darwinism even more seriously than any putative failure of the other two postulates. Insuf­ficient variation stalls natural selection; saltation deprives selection of a cre­ative role but still calls upon Darwin's mechanism as a negative force. With directed variation, however, natural selection can be bypassed entirely. If adaptive pressures automatically trigger heritable variation in favored direc­tions, then trends can proceed under regimes of random mortality; natural se­lection, acting as a negative force, can, at most, accelerate the change.

Lamarckism (defined in the modern sense of “soft” heredity) represents the quintessential theory of directed variability. Variation arises with intrinsic bias in adaptive directions either because organisms respond creatively to “felt needs” and pass acquired features directly to their offspring, or because environments induce heritable variation along favored pathways. Other di­rectional theories differ in viewing intrinsic variation as unrelated to adap­tation, but still capable of overwhelming any counteracting selection, and therefore setting the path of evolutionary change. Historically important the­ories in this mode include various notions of orthogenesis that postulate the inevitable origin of hypertrophied and inadaptive structures; and theories of “racial life cycles” that envision an ineluctably aging protoplasm doomed to extinction despite any effort at “rejuvenation” by natural selection. (I shall discuss such ideas in Chapter 5.)

Darwin clearly understood the threat of directed variability to his cardi­nal postulate of creativity for natural selection. He explicitly restricted the sources of variation to auxiliary roles as providers of raw material, and granted all power over the direction of evolutionary change to natural selec­tion. Drawing his customary analogy to artificial selection, Darwin writes (p. 30): “The key is man's power of accumulative selection: nature gives suc­cessive variations; man adds them up in certain directions useful to him. In this sense he may be said to make for himself useful breeds.”

Darwin also understood that variation could not be construed as truly ran­dom in the mathematical sense — and that history did not imply or require this strict form of randomness. He recognized biased tendencies to certain states of variation, particularly reversions toward ancestral features. But he viewed such tendencies as weak and easily overcome by selection. Thus, by the proper criterion of relative power and frequency, selection controls the di­rection of change: “When under nature the conditions of life do change, vari­ations and reversions of character probably do occur; but natural selection, as will hereafter be explained, will determine how far the new characters thus arising shall be preserved” (p. 15).

We may summarize Darwin's third requirement for variation under the ru­bric of isotropy, a common term in mineralogy (and other sciences) for the concept of a structure or system that exhibits no preferred pathway as a con­sequence of construction with equal properties in all directions. Darwinian variation must be copious in amount, small in extent, and effectively isotropic. (Think again of a dynamic sphere, with all radii accessible. The modal form lies at the center and may move by selection along any radius. At any new location, a sphere of comparable size may be reconstituted about the altered  {146}  modal form.) Only under these stringent conditions can natural selec­tion — a force that makes nothing directly, and must rely upon variation for all raw material — be legitimately regarded as creative.


Darwinism, like most comprehensive and complex concepts, defies easy definition. Darwinism cannot be analogized to an object, like the Parthenon, with a clear criterion of membership for each potential slab (whether now resident in the British Museum or in Athens). Moreover, the various proposi­tions of Darwinism cannot be regarded as either independent or of equal force. Darwinism cannot be construed as a deductive system, with some de­fining axioms and a set of logical entailments tied together like a classical proof in plane geometry. But neither can Darwinism be viewed as a set of sep­arate stones, all of similar size, and each ejectable from a bag without great disturbance to the others.

As discussed at length in Chapter 1 (pp. 1224), I view the conceptual structure of Darwinism much like the metaphor that Darwin himself first used (see Barrett et al., 1987) for depicting evolution (in the “B Notebook” on transmutation kept during the 1830's) — the “coral of life” (later super­seded, in Chapter 4 of the Origin, and in other writings, by the tree of life). The central trunk (the theory of natural selection) cannot be severed, or the creature dies (see Fig. 1-4, p. 18). The first-order branches are also so funda­mental that any severing of a complete branch converts the theory into some­thing essentially different that must be newly named. (I have suggested that the theory of natural selection includes three major branches, discussed in sections B-D of this subchapter.) Each major branch then divides into smaller sub-branches. (In the present section C, I argue that the second major branch, the claim for “creativity of natural selection,” divides into three important sub-branches of “requirements for variation,” “gradualism,” and “the adaptationist program.”)

As further argued in Chapter 1, this model allows us to address the impor­tant question of dispensability. At some level above the base, we may excise a sub branch, deny its premises, and still consider ourselves Darwinians. I envi­sion the central trunk and first-order branches as indispensable. Along the continuum from necessary to avoidable, we may begin to make selective ne­gations at the level of sub-branches, but not without severe stress to the entire structure. Thus, T. H. Huxley could oppose gradualism and still consider himself a supporter of natural selection (though his approbation remained ambiguous and indifferent at best, and his role as “Darwin's bulldog” rested upon his defense of evolution itself, not his explication of natural selection). And a modern developmental saltationist might call himself a Darwinian, though not without an array of “buts” and qualifications.

One other feature of the model requires explicit commentary. I have chosen a coral in preference to the more conventional tree, because the branches of many corals form a network by lateral anastomoses (while each limb of a tree stands free, and may be chopped off without necessarily affecting the others).  {147}  The premises of Darwin's theory (the branches and sub-branches of the coral model) are organically connected. One might be able to excise a single branch without killing the others, but some pain and readjustment will certainly be felt throughout the entire structure. The three sub-branches of the “creativ­ity” limb, for example, are strongly conjoined in this manner. If variation forms an isotropic sphere (the expectation of sub-branch one), then change by natural selection can only occur a short step at a time (as predicted by the gradualism of sub-branch two). And if variation imposes no constraint upon the direction of change (an inference from isotropy), then natural selection works freely and adaptation prevails (as required by sub-branch three).

Finally, as so often emphasized throughout this book, we must recognize and embrace natural history as a science of relative frequencies. None of these basic Darwinian premises operates without exception throughout na­ture. Darwin insisted* — explicitly and vociferously — that natural selection only enjoyed a predominant relative frequency, not exclusivity: “the main but not exclusive means of modification,” as he writes at the close of the in­troduction (p. 6). Darwin then extended his claim for a predominant relative frequency, but not for exclusivity, to all other sub-branches of his essential argument as well. Failure of raw material might occasionally explain a puz­zling absence of evolutionary modification — but lack of selective pressure for change surely represents the more likely explanation for stasis by far. Sub­stantial change might occur as a very rare event, but most alteration must be insensible, even on geological scales: “We see nothing of these slow changes in progress, until the hand of time has marked the long lapse of ages” (p. 84).

Understanding Darwin's mode of justification by relative frequency  {148}  becomes vitally important because selective quotation represents the most com­mon error made by evolutionists in interpreting his work and theory. The Or­igin, as a volume of single authorship, maintains a stronger plot line and features fewer inconsistencies than the Bible; but Darwin and the Good Lord do share the common trait of saying something about nearly everything. Wrenched from context and divorced from a crucial assessment by relative frequency, a Darwinian statement can be found to support almost any posi­tion, even the most un-Darwinian.

Since Darwin prevails as the patron saint of our profession, and since everyone wants such a preeminent authority on his side, a lamentable tradition has arisen for appropriating single Darwinian statements as defenses for par­ticular views that either bear no relation to Darwin's own concerns, or that even confute the general tenor of his work. Thus, for example, Darwin wrote extensively about variational constraint, and he maintained great interest in this topic (see Chapter 4). But the logic of his work entails adaptive control of evolutionary change and isotropy of variation as generally prevalent — and Darwin ultimately comes down (as he must) on the side of these necessary un­derpinnings for natural selection. Proper textual analysis requires that gen­eral tenor, not selective statement, be presented. Two basic procedural modes, each with distinctive criteria, set the framework for such textual analysis. The empirical mode makes its judgments of importance by relative frequency and interconnectedness of statements. Meanwhile, and simultaneously, the logical mode employs theoretical consistency as an arbiter for judging the validity and power of the structure of argument. We revere Darwin because he unfail­ingly manifested the two key traits of brilliance and honesty. He knew where his arguments led, and he followed them relentlessly, however unpleasant the consequences. We do him the greatest possible disservice when we approach his work as a superficial grazer, searching for some particular item of per­sonal sustenance, while ignoring the beauty and power of general tenor and logical entailment.

I raise this point here because abuse of selective quotation has been particularly notable in discussions of Darwin's views on gradualism. Of course Dar­win acknowledged great variation in rates of change, and even episodes of rapidity that might be labelled catastrophic (at least on a local scale); for how could such an excellent naturalist deny nature's multifariousness on such a key issue as the character of change itself? But these occasional statements do not make Darwin the godfather of punctuated equilibrium, or a cryptic sup­porter of saltation (as de Vries actually claimed, thus earning a unique and of­ficial rebuke from the organizers of the Darwinian centenary celebration at Cambridge — see p. 416).

Gradualism may represent the most central conviction residing both within and behind all Darwin's thought. Gradualism far antedates natural selection among his guiding concerns, and casts a far wider net over his choice of subjects for study. Gradualism sets the explanatory framework for his first substantive book on coral reefs (1842) and for his last on the formation of to­pography and topsoil by earthworms (1881) — two works largely devoid of  {149}  reference to natural selection. Gradualism had been equated with rationality itself by Darwin's chief guru, Charles Lyell (see Chapter 6). All scholars have noted the centrality of gradualism, both in the ontogeny (Gruber and Barrett, 1974) and logic (Mayr, 1991) of Darwin's thought.

I will not play “duelling quotations” with “citation grazers,” though a full tabulation of relative frequencies could easily bury their claims under a mountain of statements. For the present assessment of branch two (“creativ­ity of natural selection”) on the coral of essential Darwinian logic, the neces­sity of gradualism will suffice. Selection becomes creative only if it can impart direction to evolution by superintending the slow and steady accumulation of favored subsets from an isotropic pool of variation. If gradualism does not accompany this process of change, selection must relinquish this creative role and Darwinism then fails as a creative source of evolutionary novelty. If im­portant new features, or entire new taxa, arise as large and discontinuous variations, then creativity lies in production of the variation itself. Natural se­lection no longer causes evolution, and can only act as a headsman for the unfit, thus promoting changes that originated in other ways. Gradualism therefore becomes a logical consequence of the operation of natural selection in Darwin's creative mode. Gradualism also pervades the methodological pole of Darwin's greatness because the uniformitarian argument of extrapola­tion will not work unless change at the grandest scale arises by the summa­tion through time of small, immediate, and palpable variations.

Gradualism, for Darwin, represents a complex doctrine with several layers of meaning, all interconnected, while remaining independent in some impor­tant senses. I shall consider three increasing levels of specificity, arguing, on the Goldilocks model, that one meaning is too nebulous, another overly wrought, but the third (in the middle) “just right” as the crucial validator of natural selection (whereas the other two meanings play equally crucial roles for other aspects of Darwin's view of life).

Historical continuity of stuff and information. At the broadest level, gradualism merely asserts unbroken historical connectedness between putative ancestor and descendant, without characterizing the mode or rate of transition. If new species originate as creations ex nihilo by a divine power, then connectivity fails. The assertion of gradualism in this broadest meaning encapsulates the chief defense for the factuality of evolution. Such a contention could not be more vital to Darwin's revolution of course, but this sense of gradualism only asserts that evolution occurred, while telling us nothing about how evolution happens; the logical tie of gradualism to natural selection cannot reside here.* Thus, this first, or “too big,” sense of gradualism  {150}  validates evolution itself (vs. creationism), but not Darwin's, or anyone else's, proposed mechanism of evolutionary change.

Insensibility of intermediacy. We now come to the heart of what natural selection requires. This second, “just right,” statement does not ad­vance a claim about how much time a transition must take, or how variable a rate of change might be. The second meaning simply asserts that, in going from A to a substantially different B, evolution must pass through a long and insensible sequence of intermediary steps — in other words, that ancestor and descendant must be linked by a series of changes, each within the range of what natural selection might construct from ordinary variability. Without gradualism in this form, large variations of discontinuous morphological im­port — rather than natural selection — might provide the creative force of evo­lutionary change. But if the tiny increment of each step remains inconsequen­tial in itself, then creativity must reside in the summation of these steps into something substantial — and natural selection, in Darwin's theory, acts as the agent of accumulation.

This meaning of gradualism underlies Darwin's frequent invocation of the old Leibnizian and Linnaean aphorism, Natura non facit saltum (nature does not proceed by leaps). Darwin's commitment to this postulate can only strike us as fierce and, by modern standards, overly drawn. Thus, Darwin writes (p. 189): “If it could be demonstrated that any complex organ existed, which could not possibly have been formed by numerous, successive, slight modi­fications, my theory would absolutely break down.” And lest we doubt that “my theory” refers specifically to the mechanism of natural selection (and not simply to the assertion of evolution), Darwin often draws an explicit link be­tween selection as a creative force and gradualism as an implied necessity: “Undoubtedly nothing can be effected through Natural Selection except by the addition of infinitesimally small changes; and if it could be shown that... transitional states were impossible, the theory would be overthrown” (in Natural Selection — see Stauffer, 1975, p. 250). And in the concluding chapter of the Origin: “As natural selection acts solely by accumulating slight, succes­sive, favorable variations, it can produce no great or sudden modification; it can act only by very short and slow steps. Hence the canon of 'Natura non facit saltum'... is on this theory simply intelligible” (p. 471).

But would the theory of natural selection “absolutely break down” if even a single organ — not to mention an entire organism — could arise by large and discontinuous changes? Does Darwinism truly require the following extreme  {151}  formulation: “Natural selection can only act by the preservation and accu­mulation of infinitesimally small inherited modifications” (p. 95). At some level of discontinuity, of course, Darwin's strong statement must prevail. If the altered morphology of new species often arose in single steps by fortu­itous macromutation, then selection would lose its creative role and could act only as a secondary and auxiliary force to spread the sudden blessing through a population. But can we justify Darwin's application of the same claim to single organs? Suppose (as must often happen) that developmental heterochrony produces a major shift in form and function by two or three steps without intermediary stages. The size of these steps may lie outside the “normal” variation of most populations at most moments, but not beyond the potential of an inherited developmental program. (Incidentally, these types of changes represent the concept that Goldschmidt embodied in the le­gitimate meaning of “hopeful monster,” before he made his unfortunate deci­sion to tie this interesting concept to his fallacious genetics of “systemic mu­tation” — see Chapter 5 and Gould, 1982a.)

Would natural selection perish if change in this mode were common? I don't think so. Darwinian theory would require some adjustments and com­promises — particularly a toning down of assertions about the isotropy of variation, and a more vigorous study of internal constraint in genetics and de­velopment (see Chapter 10 for advocacy of this theoretical shift) — but natu­ral selection would still enjoy a status far higher than that of a mere execu­tioner. A new organ does not make a new species; and a new morphology must be brought into functional integration — a process that requires second­ary adaptation and fine tuning, presumably by natural selection, whatever the extent of the initial step.

I believe, therefore, that Darwin's strong, even pugnacious, defense of strict gradualism reflects a much more pervasive commitment, extending far be­yond the simple recognition of a logical entailment implied by natural selec­tion — and that this stronger conviction must record such general influences as Darwin's attraction to Lyell's conflation of gradualism with rationality itself, and the cultural appeal of gradualism during Britain's greatest age of indus­trial expansion and imperial conquest (Gould, 1984a). Huxley's savvy assess­ment of the Origin still rings true, for while he offered, in his famous letter to Darwin, written just as the Origin rolled off the presses, to “go to the stake” for Darwin's view, he also stated his major criticism: “You have loaded your­self with an unnecessary difficulty in adopting Natura non facit saltum so un­reservedly” (in L. Huxley, 1901, p. 189).

Darwin persevered nonetheless. We often fail to recognize how much of the Origin presents an exposition of gradualism, rather than a defense of natural selection. As a striking example, the famous (and virtually only) statement about human evolution asserts the pedagogical value of gradualism — not nat­ural selection — in our Socratic quest to know ourselves: “Psychology will be based on a new foundation, that of the necessary acquirement of each mental power and capacity by gradation. Light will be thrown on the origin of man and his history” (p. 488).  {152} 

Chapter 9 on geological evidence, where the uninitiated might expect to find a strong defense for evolution from the most direct source of evidence in the fossil record, reads instead as a long (and legitimate) apologia for a threatening discordance between data and logical entailment — a fossil record dominated by gaps and discontinuities when read literally vs. the insensible transitions required by natural selection as a creative agent. Darwin, with his characteristic honesty, states the dilemma baldly in succinct deference to his methodological need for equating temporal steps of change with differences noted among varieties of contemporary species: “By the theory of natural se­lection all living species have been connected with the parent-species of each genus, by differences not greater than we see between the varieties of the same species at the present day” (p. 281).

Darwin, as we all know, resolved this discordance by branding the fossil record as so imperfect — like a book with few pages present and only a few let­ters preserved on each page — that truly insensible continuity becomes de­graded to a series of abrupt leaps in surviving evidence:

Why then is not every geological formation and every stratum full of such intermediate links? Geology assuredly does not reveal any such finely graduated organic chain; and this, perhaps, is the most obvious and gravest objection which can be urged against my theory. This expla­nation lies, as I believe, in the extreme imperfection of the geological rec­ord (p. 280).

He, who rejects these views on the nature of the geological record, will rightly reject my whole theory (p. 342).

Slowness and smoothness (but not constancy) of rate. Darwin also championed the most stringent version of gradualism — not mere continuity of information, and not just insensibility of innumerable transi­tional steps; but also the additional claim that change must be insensibly gradual even at the broadest temporal scale of geological durations, and that continuous flux (at variable rates to be sure) represents the usual state of nature.

This broadest version of gradualism does not hold strong logical ties to natural selection as an evolutionary mechanism. Change might be episodic and abrupt in geological perspective, but still proceed by insensible intermediacy at a generational perspective — given the crucial scaling principle that thousands of generations make a geological moment. For this reason, Eldredge and I have never viewed punctuated equilibrium, which does refute Darwinian gradualism in this third sense, as an attack on the creativity of nat­ural selection itself (Eldredge and Gould, 1972; Gould and Eldredge, 1977, 1993). The challenge of punctuated equilibrium to natural selection rests upon two entirely different issues of support provided by punctuational ge­ometry for the explanation of cladal trends by differential species success and not by extrapolated anagenesis, and for the high relative frequency of species selection, as opposed to the exclusivity of Darwinian selection on organisms (see Chapters 8 and 9).  {153} 

Some fidei defensores of the Darwinian citadel have sensed the weakness of this third version of gradualism, and have either pointed out that the creativ­ity of natural selection cannot be compromised thereby (quite correct, but then no one ever raised such a challenge, at least within the legitimate de­bate on punctuated equilibrium); or have argued either that Darwin meant no such thing, or that, if he really did, the claim has no importance (see Dawkins, 1986). This last effort in apologetics provides a striking illustration of the retrospective fallacy in historiography. Whatever the current status of this third formulation within modern Darwinism, this broadest style of grad­ualism was vitally important to Darwin; for belief in slow change in geologi­cal perspective lies at the heart of his more inclusive view about nature and science, an issue even larger than the mechanics of natural selection.

Darwin often states his convictions about extreme slowness and contin­uous flux in geological time — as something quite apart from gradualism's second meaning of insensible intermediacy in microevolutionary perspec­tive. Evolutionary change, Darwin asserts, usually occurs so slowly that even the immense length of an average geological formation may not reach the mean time of transformation between species. Thus, apparent stasis may ac­tually represent change at average rates, but to an imperceptible degree even through such an extensive stretch of geological time! “Although each forma­tion may mark a very long lapse of years, each perhaps is short compared with the period requisite to change one species into another” (p. 293).

Change not only occurs with geological slowness on this largest scale; but most transformations also proceed in sufficient continuity and limited varia­tion in rate that elapsed time may be roughly measured by degree of accumu­lated difference: “The amount of organic change in the fossils of consecutive formations probably serves as a fair measure of the lapse of actual time” (p. 488).

Darwin presents his credo in crisp epitome: “Nature acts uniformly and slowly during vast periods of time on the whole organization, in any way which may be for each creature's own good” (p. 269). Note how Darwin concentrates so many of his central beliefs into so few words: gradualism, adaptationism, locus of selection on organisms.

But the most striking testimony to Darwin's conviction about gradualism in this third sense of slow and continuous flux lies in several errors promi­nently highlighted in the Origin — all based on convictions about steady rate (gradualism in the third sense), not on the insensible intermediacy genuinely demanded by natural selection (gradualism in the second sense), or on the simple continuity of historical information required to validate the factuality of evolution itself (gradualism in the first sense). For example, Darwin makes a famous calculation (dropped from later editions) on the “denudation of the Weald” — the erosion of the anticlinal valley located between the North and South Chalk Downs of southern England (pp. 285-287). He tries to deter­mine an average value for yearly erosion of seacliffs today, and then extrapo­lates his figure as a constant rate into the past. His date of some 300 million years for the denudation of the Weald overestimated the true duration by five  {154}  times or more. (The deposition of the Chalk, an Upper Cretaceous formation, persisted nearly to the period's end 65 million years ago.)

Moving to a biological example that underscores Darwin's hostility to episodes of “explosive” evolutionary diversification (he used his usual argument about the imperfection of the fossil record to deny their literal appearance and to spread them out in time), Darwin predicted that the Cambrian explo­sion would be exposed as an artifact, and that complex multicellular crea­tures must have thrived for vast Precambrian durations, gradually reaching the complexity of basal Cambrian forms. (When Darwin published in 1859, the Cambrian had not yet been recognized, and his text therefore speaks of the base of the Silurian, meaning lower Cambrian in modern terminology): “If my theory be true, it is indisputable that before the lowest Silurian stra­tum was deposited, long periods elapsed, as long as, or probably far longer than, the whole interval from the Silurian age to the present day; and that during these vast, yet quite unknown periods of time, the world swarmed with living creatures” (p. 307).

Paleontologists have now established a good record of Precambrian life. The world did swarm indeed, but only with single-celled forms and multi-cellular algae, until the latest Precambrian fauna of the Ediacara beds (begin­ning about 600 million years ago). The explosion of multicellular life now seems as abrupt as ever — even more so since the argument now rests on copi­ous documentation of Precambrian life, rather than a paucity of evidence that could be attributed to imperfections of the geological record (see Chapter 10, pp. 11551161). Darwin on the other hand, predicted that complex, multi­cellular creatures must extend far into the Precambrian. He wrote: “I cannot doubt that all the Silurian [= Cambrian] trilobites have descended from some one crustacean, which must have lived long before the Silurian [= Cambrian] age” (p. 306). Darwin also conjectured, again incorrectly, that the ancestral verterbrate, an animal with an adult phenotype resembling the common embryological Bauplan of all modern vertebrates, must have lived long before the dawn of Cambrian times: “It would be vain to look for [adult] animals having the common embryological character of the Vertebrata, until beds far beneath the lowest Silurian strata are discovered” (p. 338).

Darwin struggled for clarity and consistency. He did not always succeed. (How can an honest person so prevail in our complex and confusing world? I shall, for example, examine Darwin's ambivalences on progress in Chapter 6.) Darwin did not always keep the different senses of gradualism distinct. He frequently conflated meanings, arguing (for example) that the validity of nat­ural selection (sense 2) required an acceptance of slow and continuous flux (sense 3). Consider once again the following familiar passage: “It may be said that natural selection is daily and hourly scrutinizing, throughout the world, every variation, even the slightest... We see nothing of these slow changes in progress, until the hand of time has marked the long lapse of ages” (p. 84).

This conflation came easily (and probably unconsciously) to Darwin, in large part because gradualism stood prior to natural selection in the core of his beliefs about the nature of things. Natural selection exemplified gradualism,  {155}  not vice versa — and the various forms of gradualism converged to a sin­gle, coordinated view of life that extended its compass far beyond natural se­lection and even evolution itself. This situation inspired Huxley's frustration as he remonstrated with Darwin (see the famous quote on p. 151): you will have enough trouble convincing people about natural selection; why do you insist upon uniting this theory with an unnecessary and, by the way, false claim for gradualism?

We can best sense this overarching Darwinian conviction in a lovely pas­sage that conflates all three senses of gradualism — the rationalist argument against creationism, the validation of natural selection by insensible intermediacy, and the slow pace of change at geological scales — all in the context of Darwin's homage to his guru Lyell, and his aesthetic and ethical convic­tions about the superiority of these “noble views” about natural causation and the nature of change:

I am well aware that this doctrine of natural selection ... is open to the same objections which were at first urged against Sir Charles Lyell's no­ble views on “the modern changes of the earth, as illustrative of geol­ogy;” but we now very seldom hear the action, for instance, of the coast-waves, called a trifling and insignificant cause, when applied to the exca­vation of gigantic valleys or to the formation of the longest lines of in­land cliffs. Natural selection can act only by the preservation and accu­mulation of infinitesimally small inherited modifications, each profitable to the preserved being; and as modern geology has almost banished such views as the excavation of a great valley by a single diluvial wave, so will natural selection, if it be a true principle, banish the belief in the contin­ued creation of new organic beings, or of any great and sudden modifica­tion in their structure (pp. 95-96).

The adaptationist program

Darwin's three constraints on the nature of variation form a single conceptual thrust: variation only serves as a prerequisite, a source of raw material in­capable of imparting direction or generating evolutionary change by itself. Gradualism, in the second meaning of insensible intermediacy, then guaran­tees that the positive force of modification proceeds step by tiny step. There­fore, the explanation of evolution must reside in specifying the causes of change under two conditions that logically entail a primary focus on adapta­tion as a canonical result: we know the general nature of change (gradual­ism), and we have eliminated an internal source from variation itself (the ar­gument for isotropy). Change must therefore arise by interaction between external conditions (both biotic and abiotic) and the equipotential raw mate­rial of variation. Such gradual adjustment of one to the other must yield ad­aptation as a primary outcome.

Adaptational results flow logically from the mechanisms defining all other subbranches on the limb of Darwinism designated here as the “creativity of natural selection.” But Darwin constructed this limb in reverse order in the  {156}  psychological development of his theory. For he had long viewed an explana­tion of adaptation as the chief requirement of evolutionary theory. He sought the causes of evolution within his patrimony — the English tradition in natural theology — and he attempted to subvert this patrimony from within by ac­cepting its chief empirical postulate of good design and then providing an in­verted theoretical explanation (see p. 125).

When Darwin permits himself to make one of his rare forays into lyrical prose, we can grasp more fully (and dramatically) the extent of his feelings and the depth of his conviction. Consider the following passage on why the basic results of evolution and variation teach us so little about the origin of species, and why an understanding of mechanism requires an explanation of adaptation:

But the mere existence of individual variability and of some few well-marked varieties, though necessary as the foundation for the work, helps us but little in understanding how species arise in nature. How have all those exquisite adaptations of one part of the organization to another part, and to the conditions of life, and of one distinct organic being to another being, been perfected? We see these beautiful co-adaptations most plainly in the woodpecker and missletoe; and only a little less plainly in the humblest parasite which clings to the hairs of a quadruped or feathers of a bird; in the structure of the beetle which dives through the water; in the plumed seed which is wafted by the gentlest breeze; in short, we see beautiful adaptations everywhere and in every part of the organic world (pp. 60-61).

Pursuing the theme of rare Darwinian lyricism as a guide to what he viewed as essential, consider his convictions about the overwhelming power of natural selection — a point that he usually conveyed by comparison with the limitations of artificial selection in breeding and agriculture:

Man can act only on external and visible characters: nature cares noth­ing for appearances, except in so far as they may be useful to any being. She can act on every internal organ, on every shade of constitutional dif­ference, on the whole machinery of life. Man selects only for his own good; Nature only for that of the being which she tends. Every selected character is fully exercised by her; and the being is placed under well-suited conditions of life. Man keeps the natives of many climates in the same country; he seldom exercises each selected character in some pecu­liar and fitting manner; he feeds a long and a short beaked pigeon on the same food; he does not exercise a long-backed or long-legged quadruped in any peculiar manner; he exposes sheep with long and short wool to the same climate. He does not allow the most vigorous males to struggle for the females. He does not rigidly destroy all inferior animals, but pro­tects during each varying season, as far as lies in his power, all his pro­ductions... Under nature, the slightest difference of structure or consti­tution may well turn the nicely balanced scale in the struggle for life  {157}  and so be preserved. How fleeting are the wishes and efforts of man! how short his time! and consequently how poor will his products be, compared with those accumulated by nature during whole geological pe­riods. Can we wonder, then, that nature's productions should be far “truer” in character than man's productions; that they should be in­finitely better adapted to the most complex conditions of life, and should plainly bear the stamp of far higher workmanship? (pp. 83-84).

But Darwin's world also differs strongly from Paley's, and the outcome of natural selection, however great the power of Darwin's mechanism, cannot be perfection, but only improvement to a point of competitive superiority in local circumstances. Natural selection operates as a principle of “better than,” not as a doctrine of perfection: “Natural selection tends only to make each organic being as perfect as, or slightly more perfect than, the other in­habitants of the same country with which it has to struggle for existence” (p. 201). Thus, the signs of history will not be erased; creatures will retain sig­natures of their past as quirks, oddities and imperfections (see pp. 111-116 on methodology). Natural selection will fashion the organic world, while leaving enough signs of her previous handiwork to reveal a forming presence.

I have called this section “the adaptationist program,” rather than, simply, “adaptation” because Darwin presents a protocol for actual research, not just an abstract conceptual structure. The relevant arguments may be ordered in various ways, but consider this sequence:

• Adaptation is the central phenomenon of evolution, and the key to any understanding of mechanisms.

• Natural selection crafts adaptation.

• Natural selection maintains an overwhelmingly predominant relative frequency as a cause of adaptation. Variation only provides raw material, and cannot do the work unaided.

Adaptation may be viewed as a problem of transforming environmental (external) information into internal changes of form, physiology and behav­ior. Two forces other than natural selection might play such a role — the cre­ative response of organisms to felt needs with inheritance of acquired charac­ters (Lamarck's system), or direct impress of environments upon organisms, also with inheritance of traits thus acquired (a system often associated with Geoffroy Saint-Hilaire). Darwin regards both alternatives as true causes, and he explicitly contrasts them with natural selection in several passages within the Origin. But, in these statements and elsewhere, he always grants natural selection the cardinal role by virtue of relative frequency — “by far the pre­dominant Power,” he writes on page 43, in upper case for emphasis. “Over all other causes of change, I am convinced that Natural Selection is para­mount” (in Natural Selection, 1975 edition, p. 223).

In this light, how should evolutionists proceed if they wish to discover the mechanisms of change? Should they study the causes of variation (a vitally important issue, but unresolvable in Darwin's time, and not the cause of  {158}  change in any case)? Or should they examine the large-scale phenomena of taxonomic order or geographic distribution (issues of great import again, but lying too far from immediate causation)? Instead, the best strategy, Darwin asserts, lies in the study of adaptation, for adaptation is the direct and pri­mary result of natural selection; and the relative frequency of selection stands so high that almost any adaptation will record its forming power.

Adaptation therefore becomes, for Darwin, the primary subject for practi­cal study of evolutionary mechanisms. Recall the basic methodological prob­lem of a science of history (see p. 102): science aims, above all, to understand causal processes; past processes cannot be observed in principle; we must therefore learn about past causes by making inferences from preserved re­sults. Adaptation is the common and coordinating result of nearly any epi­sode of non-trivial evolutionary change. Adaptation not only pervades nature with an overwhelming relative frequency, but also embodies the immediate action of the primary cause of change — natural selection. The adaptations of organisms therefore constitute the bread and butter objects of study in evolu­tionary biology. Our first order approach to change must pose the following question in any particular case: what adaptive value can we assign; how did natural selection work in this instance? In a revealing statement, Darwin rolls all exceptions, all ifs and buts, into a set of subsidiaries to adaptation forged by natural selection — as either consequences of adaptation, inherited marks of older adaptations, or rare products of other processes: “Hence every detail of structure in every living creature (making some little allowance for the di­rect action of physical conditions) may be viewed, either as having been of special use to some ancestral form or as being now of special use to the descendants of this form — either directly, or indirectly through the complex laws of growth” (p. 200).

The primary anti-Darwinian argument of late 19th century biology pro­ceeded by denying a creative role to natural selection — but Darwin countered with a strong riposte. If adaptation pervades nature and must be constructed by natural selection, and if the steps of evolutionary sequences are generally so tiny that we may seek their source in palpable events subject to our direct view and manipulation, then we not only gain a theoretical explanation for evolutionary change. We also obtain the practical gift of a workable research program rooted in the observable and the resolvable.

But nothing so precious comes without a price, or without consequences. Darwin's argument works; no logical holes remain. But the research program thus entailed must embody attitudes and assumptions not necessarily true — or at least not necessarily valid at sufficiently high relative frequency to make the world exclusively, or even primarily, Darwinian. To accept Darwin's full argument about the creativity of natural selection, one must buy into an en­tire conceptual world — a world where externalities direct, and internalities supply raw material but impose no serious constraint upon change; a world where the functional impetus for change comes first and the structural alter­ation of form can only follow. The creativity of natural selection makes adap­tation central, isotropy of variation necessary, and gradualism pervasive.  {159} 

But suppose these precepts do not govern a commanding relative frequency of cases? What if adaptation does not always record the primacy of natural selection, but often arises as secondary fine tuning of structures arising in other ways? What if variation imposes strong constraints and supplies pow­erful channels of preferred direction for change? What if the nature of varia­tion (particularly as expressed in development) often produces change with­out insensible intermediacy?

All these arguments merge into a structuralist critique that seriously challenges the predominant functionalism of classical Darwinism. As a common thread, these challenges deny exclusivity to natural selection as the agent of creativity, and claim a high relative frequency of control by internal factors. McCosh was right in establishing his pre-evolutionary contrast of a “princi­ple of order” and a “principle of special adaptation” (see p. 116). Darwin was right in translating this distinction into evolutionary terms as “Unity of Type” and “Conditions of Existence,” though he was probably wrong in his fateful decision — the basis of Darwinian functionalism — to yoke the two cat­egories together under a common cause by defining unity of type as the his­torical legacy of previous adaptation, thus asserting the domination of natu­ral selection (1859, p. 206 — see extensive commentary in chapter 4). And E. S. Russell (1916) was also right in contrasting the “formal or transcenden­tal” with the “functional or synthetic” approach to morphology.

We are children of Darwin, and an English school of adaptation and functionalism far older than evolutionary theory. Darwin's key claim for the creativity of natural selection — and the resulting sequelae of gradualism, adaptationism, and the isotropy of variation — builds the main line of defense for this powerful and venerable attitude towards nature and change. For many of us, these claims lie too close to the core of our deeply assimilated and now largely unconscious beliefs to be challenged, or even overtly recognized as something potentially disputable. Yet a coherent alternative has been pro­posed, and now provides one of the three most trenchant modern critiques of strict Darwinism. I believe that these critiques, taken together, will reorient evolutionary theory into a richer structure with a Darwinian core. But we cannot appreciate the alternatives until we grasp the basis of orthodoxy as an argument of compelling brilliance and power. Important critiques can only operate against great orthodoxies.


The first two themes — causal focus on organisms as agents of selection and creativity of selection in crafting adaptation — establish the biological core of Darwinian theory. That is, they perform the biological “work” needed to as­sure the third and last essential component of a Darwinian worldview: the uniformitarian argument for full application in extrapolation to all scales and times in the history of life. Mere operation in the microevolutionary here and now cannot suffice. Natural selection must also assert a vigorous claim for  {160}  preeminence throughout the 3.5 billion years of phylogeny, lest the theory be reduced to an ornamental device, imposing only a fillip of immediate adap­tive detail upon a grand pageant generated by other causes and forces.

Darwin, who fledged professionally as a geologist (the subject of his first three scientific books in the 1840's, on coral reefs, volcanic islands, and the geology of South America), and who regarded Charles Lyell as his intellectual hero, while embracing his mentor's doctrine of uniformitarianism as the core of his own philosophy as well, fully understood that his revolution would succeed only if he could show how natural selection might act as architect for the full panoply of life's history throughout geological time. The “method­ological pole,” one of the two foci of Darwin's revolution (see Section II of this chapter), brilliantly develops a set of procedures for defending extrapola­tion in various contexts of limited evidence.

The link of the first two themes (agency and efficacy) to this third theme of extended scope or capacity — thus forming in their threefold ensemble a minimally complete statement of revolution — received succinct expression in Ernst Mayr's (1963, p. 586) epitome of Darwinism as preached by the Mod­ern Synthesis: “All evolution is due to the accumulation of small genetic changes, guided by natural selection [the first two themes of agency and efficacy], and that transpecific evolution [the third theme of scope, or uniformitarian extension] is nothing but an extrapolation and magnification of the events that take place within populations and species.”

In this book, my explicit discussion for this third theme of extrapolation (Chapters 6 and 12) shall be shorter than my treatment of the first theme of agency (Chapters 3 and 89), leading from Darwin's nearly exclusive focus on the organismal level to the modern revision of hierarchical selection the­ory, and the second theme of efficacy (Chapters 45 and 1011) on older and modern critiques of panadaptationism, with an emphasis on structural prin­ciples and constraints. I allocate my attention in this unequal manner because the first two themes already include, within themselves, the biological argu­ments for extrapolation, as embodied in Darwin's uniformitarian beliefs and practices. For my explicit and separate treatment of the crucial extrapolationist theme in this work, I therefore follow a different strategy, if only to avoid redundancy in a book that we all undoubtedly regard, author and read­ers alike, as quite long enough already! I will not rehearse Darwin's biological arguments for extrapolation, but will rather, as a “place holder” of sorts, concentrate upon the nature of the geological stage that must welcome Dar­win's biological play.

I proceed in this way for a principled reason, and not merely as a convenience. All major evolutionary theories before Darwin, and nearly all important versions that followed his enunciation of natural selection as well, retained fealty to an ancient Western tradition, dating to Plato and other clas­sical authors, by presenting a fundamentally “internalist” account, based upon intrinsic and predictable patterns set by the nature of living systems, for development or “unfolding” through time. (Ironically, such internalist theo­ries follow the literal meaning of “evolution” (unfolding) far better than the Darwinian system that  {161}  eventually absorbed the term. Darwin understood this etymological point perfectly well, and he initially declined to use the word “evolution” — preferring “descent with modification” — probably because he recognized the difference between the literal meaning of “evolution” and his own concept of life's history and change by natural selection — see Gould, 2000a.)

Darwin's theory, in strong and revolutionary contrast, presents a first “externalist” account of evolution, in which contingent change (the summation of unpredictable local adaptations rather than a deterministic unfolding of in­herent potential under internal, biological principles) proceeds by an interac­tion between organic raw material (undirected variation) and environmental guidance (natural selection). Darwin overturned all previous traditions by thus granting the external environment a causal and controlling role in the di­rection of evolutionary change (with “environment” construed as the ensem­ble of biotic and abiotic factors of course, but still external to the organism, however intrinsically locked to, and even largely defined by, the presence of the organism itself). Thus, and finally, in considering the validity of extrapo­lation to complete the roster of essential Darwinian claims, the role of the geological stage becomes an appropriate focus as a surrogate for more overtly biological discussion.

If the uniqueness of Darwinism, and its revolutionary character as well, inheres largely in the formulation of natural selection as a theory of interaction between biological insides and environmental outsides — and not as a theory of evolutio, or intrinsic unfolding — then “outsides” must receive explicit dis­cussion as well, a need best fulfilled within this treatment of extrapolation. Under internalist theories of evolution, environment, at most, holds power to derail the process by not behaving properly — drying up, as on Mars, or freez­ing over, as nearly occurred on Earth more than once during our planet's geo­logical history. Under Darwinian functionalism, however, environment be­comes an active partner in both the modes and directions of evolutionary change.

As the Utopian tradition recognizes, we can often devise lovely and optimal systems in abstract principle, but then be utterly unable to apply them in practice because an imperfect world precludes their operation. The central logic of Darwinism faces an issue of this kind. The two essential biological postulates of natural selection — its operation at the organismal level, and its creativity in crafting adaptations — build a sufficient theoretical apparatus to fuel the system. The play of evolution can run with such a minimal cast, but we do not know whether the drama can actually unfold on our planet until we also examine and specify the character of the theater — the geological and environmental stage for the play of natural selection. The geological stage therefore becomes a major actor in the drama set on its own premises.

Moreover, and reinforcing my argument that Darwin's strength lies in his brave specificity, Darwin places a great burden on geology and environment by devising such stringent conditions for the nature of this external setting. Again, we encounter the Goldilocks problem — environment cannot impose too much or provide too little, but must be “just right” in the middle.

Environment, as an active Darwinian agent, cannot under perform. In particular,  {162}  an absence of environmental change would probably bring evolution to an eventual halt, as selective pressures for adaptive alteration diminished (see Stenseth and Maynard Smith, 1984). Purely biotic interaction might drive evolution for some time following a cessation of environmental change, but probably not indefinitely.

The possibility of too little change has rarely been viewed as a threat to Darwinism, largely because the geological record seems so clearly to empha­size potential dangers in the other direction (though see pp. 492502 on Lord Kelvin). The specter of “too much” change, on the other hand, has haunted Darwinism from the start. In particular, if the theory of geological catastrophism were generally true, or even just sufficiently important in relative frequency, then Darwinism would be compromised as the primary agent of pattern in the history of life.

By catastrophism, I mean to designate the classical theory of global parox­ysm as a primary agent of geological change — in particular, the idea that mass extinctions thus engendered might lie largely outside the domain of tradi­tional Darwinism. Of course, mass extinctions cannot be construed as “undarwinian” per se. If environment changes so rapidly that organisms cannot adapt fast enough by natural selection, then many species will die. But, in a conceptual world of relative frequency, where Darwinism must not only op­erate, but also dominate as the creator of change, such formative power for mass extinction constitutes a serious challenge. If we survey the entire history of life, and find that catastrophic mass extinction, with non-Darwinian fortu­ity in causes of change (on either the “random” or the “different rules” model — see Chapter 12, and Gould, 1985a, 1993c), establishes more features of overall pattern than the ordinary interplay of taxa during normal times (between such episodes of coordinated death) can build and maintain, then Darwin's view of life lacks the generality once accorded. In particular, the key uniformitarian argument will then fail. The adaptive struggles of immediate moments will not extrapolate to explain the patterns of life's history. More­over, if these undarwinian components of fortuity in extinction, and success for reasons unrelated to the original adaptive basis of traits, also maintain strong influence at lesser scales of smaller mass extinctions (Raup and Sepkoski, 1984), and even, in a fractal manner, for some ordinary extinctions in normal times (Raup, 1991), then the challenge may become truly pervasive.

These characterizations of Darwinian requirements cannot be dismissed or downgraded as conjectures or reconstructions, only inferentially based on deductions from premises stated by Darwin for different reasons. Darwin de­voted an entire chapter of the Origin, number 10 “on the geological succes­sion of organic beings,” to an exploration of the geological stage and its requirements for natural selection. He argues that biotic competition, gradualistically expressed through time as coordinated waxing and waning of in­teracting clades, marks the overall pattern of life — and that the apparent fos­sil evidence for more rapid change, set by physical environments and leading to mass extinctions, must generally be read as artifacts of an imperfect record (see Chapter 12 for detailed exegesis of Darwin's arguments on this subject).  {163}  This issue exposes another essential Darwinian theme not yet discussed (but receiving full treatment in Chapter 6) — the nature of competition; the preva­lence of biotic over abiotic effects; the metaphor of the wedge; and the funda­mental role of Darwinian ecology as a validator of progress (in the absence of any available defense from the bare-bones mechanism of natural selection it­self). Thus, the argument for uniformitarian change in geology undergirds a central conviction of the Darwinian corpus.

We cannot overestimate the depth of Darwin's debt to his intellectual hero, Charles Lyell. The uniformitarianism of his mentor not only provided, by transfer into biology, a theory of evolutionary change. The doctrine of unifor­mity also supplied, on its original geological turf, a world that could grant enough slow and continuous environmental change to fuel natural selec­tion — but not so much, or so quickly, that selection would be overcome, and the rein of pattern seized by environment in its own right. In natural selec­tion, environment proposes and organisms dispose; this subtle balance of in­side and outside must be maintained. But in a world of too much environ­mental change, the external component does not only propose, but can also dispose of organisms and species without much backtalk. Darwinism does not run well on such a one-way street.

Judgments of Importance

In the difficult genre of comprehensive historical reviews, a few special books stand out as so fair in their judgments and so lucid in their characterizations that they set the conceptual boundaries of disciplines for generations. In mor­phology, E. S. Russell's Form and Function (1916) occupies this role for the brilliance and justice of its characterizations, even though Russell, as an avowed Lamarckian, made no secret about his own preferences (and made the wrong choice by modern standards). In evolutionary biology, similar plaudits may be granted to Vernon L. Kellogg's Darwinism Today (1907). Kellogg, a great educator and entomologist from Stanford, had collaborated with David Starr Jordan on the best textbooks of his generation. He also played an ironic role in the history of evolution by serving a term (while America maintained her early neutrality) as chief agent for Belgian relief, posted to the German General Staff in Berlin during World War I. There, he listened in horror to German leaders perverting Darwinism as a justification for war and conquest — and he exposed these distortions in his fascinating volume, Headquarters Nights (1917). William Jennings Bryan read this book and, understanding the abuse but blaming the victims of misinterpretation rather than the perpetrators, launched his campaign to ban the teaching of evolution as a result (see Gould, 1991b).

As the Darwinian centennial of 1909 neared, Kellogg decided to write a volume providing a fair hearing for all varieties of Darwinism, and all alter­native views in a decade of maximal agnosticism and diversity in evolution­ary theories. Kellogg's book adopts the same premise as this treatise — that  {164}  Darwinism embodies a meaningful central logic, or “essence,” and that other proposals about evolutionary mechanisms can be classified with reference to their consonance or dissonance with these basic Darwinian commitments.

I was particularly pleased to learn that Kellogg's categories, though differ­ently named and parsed, are identical with those recognized here. He divides the plethora of proposals under discussion in his time into those “auxiliary to” and those “alternative to” natural selection. Among auxiliaries that aid, expand, improve, or lie within the spirit of Darwinism, Kellogg highlights two principal themes: studies of Wagner, Jordan, and Gulick on the role of isolation in the formation of species; and hierarchical models of selection as espoused by Roux and Weismann (discussed in detail in Chapter 3). I noted with special gratification that Kellogg recognized hierarchy as an auxiliary, not a confutation, to Darwinism, for this same contention sets a principal theme of this book.

In his second category of confutations, Kellogg identified “three general theories, or groups of theories, which are offered more as alternative and substitionary theories for natural selection than as auxiliary or supporting theories” (1907, p. 262): Lamarckism (inheritance of acquired characters in the form advocated by late 19th century neo-Lamarckians), orthogenesis, and heterogenesis (Kellogg's designation for saltationism).

Kellogg's taxonomy works particularly well in evaluating the central principles of Darwinism. His “auxiliaries” aid selection (by addition of other principles that do not challenge or diminish selection, or by expansion of se­lection to other levels); but his “alternatives” confute particular maxims of the minimal commitments for Darwinian logic. The Kelloggian “alterna­tives” all deny the fundamental postulate of creativity for selection by desig­nating other causes as originators of evolutionary novelties, and by relegating selection to a diminished status as a negative force. Each alternative rejects a necessary Darwinian postulate about the nature of variation (see pp. 141146): Lamarckism and orthogenesis deny the principle of undirected variabil­ity; saltationism refutes the claim that variation must be small in extent.

I warmly endorse Kellogg's approach. As practicing scientists, we often do not pay enough respect to the logical structure of an argument — to its rigors and its entailments. We tend to assume that conclusions flow unambiguously from data, and that if we observe nature closely enough, and experiment with sufficient care and cleverness, the right ideas will somehow coalesce or flow into place by themselves. But scholars should know, from the bones and guts of their practice, that all great theories originate by intense and explicit men­tal struggle as well. We should not castigate such efforts as “speculation” or “armchair theorizing” — for mental struggle deserves this designation only when the thinker opposes or disparages our shared conviction that, ulti­mately, empirical work or testing must accompany and validate such exer­cises in thought (and then all scientists would agree to let the calumny fall). Great theories emerge by titration of this basically lonely mental struggle with the more public, empirical acts of fieldwork and bench work.

One need look no further than Charles Darwin for proper inspiration. He  {165}  rooted his theory in practical testability, and he continually devised and performed clever experiments, despite limited resources (of available equipment and personnel at Down, not of funds; for Darwin was a wealthy man and did not need to spend his time seeking patronage, his generation's equivalent of modern grant swinging). But natural selection did not flow from the external world into a tabula rasa of Darwin's mind. He carried out with himself, as re­corded in his copious notebooks (Barrett et al., 1987), one of the great men­tal struggles of human history — proposing and rejecting numerous theories along his slow and almost painful journey by inches, accompanied by lateral feints and backward plunges, towards the theory of natural selection. That theory, when fully formulated in the 1850's, emerged as an intricately devised amalgam of logically connected parts, each with a necessary function — and not as a simple message from nature. We must treat this theory, as Kellogg does, with respect for its integrity.

With the coalescence and hardening of the Modern Synthesis (Gould, 1983b), culminating in the Darwinian celebrations of 1959, orthodoxy descended over evolutionary theory, and a generation of unprecedented agreement ensued (often for reasons of complacency or authority). However, the press of new concepts and discoveries has since fractured this shaky con­sensus, and we now face a range of options and alternatives fully as broad as those available in the contentious decade of Kellogg's review. In this renewed context, I recommend Kellogg's procedure as both intellectually admirable and maximally useful — namely, to arrange and evaluate various views and challenges by classification according to their attitudes towards the minimal commitments of Darwinism. I say “admirable” because such an approach pays proper respect to the intellectual power of Darwin's synthesis, and “use­ful” because taxonomy by minimal commitments of an essential logic allows us to rank, assess, and interconnect an otherwise confusing array of proposals and counterproposals. And just as the widespread debate of Kellogg's time led to the Modern Synthesis of the next generation, I believe that the re­newed arguments of our day will pay dividends in the form of a richer and more adequate consensus for our new millennium. Kellogg's characterization of his own era therefore becomes relevant to our current situation:

The present time is one of unprecedented activity and fertility both in the discovery of facts and in attempts to perceive their significance in rela­tion to the great problems of bionomics. Both destructive criticism of old, and synthesis of new hypotheses and theories, are being so energeti­cally carried forward that the scientific layman and educated reader, if he stand but ever so little outside of the actual working ranks of biology, is likely to lose his orientation as to the trends of evolutionary advance. Precisely at the present moment is this modification of the general point of view and attitude of philosophical biologists unusually important and far-reaching in its relation to certain long-held general conceptions of bi­ology and evolution (1907, p. ii).

I have therefore followed Kellogg's lead and attempted, in this introductory chapter, to characterize the central logic and minimal commitments of Darwinism — an essence, if you will, to invoke a good word and concept that  {166}  has become taboo in our profession. I will then use this characterization as a foundation for classifying various challenges and controversies — just as Kellogg did — according to their stance towards the essential concepts of Darwinism. The most interesting and far-reaching challenges directly engage these essential concepts, either as alternatives to refute them in part, or as auxiliaries to expand and reinterpret them in fundamental ways. This book presents, as its primary thesis, the notion that (i) Darwinism may be viewed as a platform with a tripod of essential support; (ii) each leg of the tripod now faces a serious reforming critique acting more as an auxiliary than an alterna­tive formulation; and (iii) the three critiques hold strong elements in com­mon, and may lead to a fundamentally revised evolutionary theory with a re­tained Darwinian core.

We must rank challenges by their degree of engagement with the Darwin­ian core; we cannot follow a strategy of mindless “raw empiricism” towards the Origin and simply compile a list of Darwin's mistakes. All great works are bursting with error; how else could true creativity be achieved? Could anyone possibly reformulate a universe of thought and get every detail right the first time? We should not simply count Darwin's errors, but rather assess their im­portance relative to his essential postulates. (Consider, for example, the stan­dard rhetorical, and deeply anti-intellectual, ploy of politically motivated and destructive critics, American creationists in particular. They just list the mis­takes, envelop each in a cloud of verbal mockery, and pretend that the whole system has drowned in this tiny puddle of inconsequential error.)

I suggest that we use the list of minimal commitments to gauge the status of Darwin's errors. Very few faults of simple fact can, as individual items, be of much consequence unless they confute a core commitment. Darwin argued, for example, that swimbladders evolved into lungs (see p. 107) though ex­actly the opposite occurred — but no premise of the general theory suffers any injury by this mistake, however embarrassing. What about more important theoretical claims like Darwin's hypothesis of “pangenesis” as a mechanism of heredity (Darwin, 1868)? Again, Darwin's view of life would have been easier to vindicate if the theory had been affirmed, but none of his three essen­tial postulates about the nature of variation fell with the disproof of pangen­esis, and the core commitments remained intact, if unproven. What about the impact of major claims that turn out to be basically true, Mendelism for ex­ample? We must make our judgment by assessing their engagement with the core commitments. In the first decade of the 20th century, most evolution­ists invoked Mendelism as a saltational theory of macromutation against the Darwinian core commitment to small-scale variation (see Chapter 5). Later, largely through R. A. Fisher's analysis and the resolution of the Mendelian vs. biometrician debate, macromutations were rejected, “ordinary” small-scale variation granted a Mendelian basis, and Mendelism comfortably reinter­preted as support for the same core commitment. Again, challenges and new proposals must be judged and ranked by their engagement with the essence of a reigning theory. Darwinism embodies a definable set of minimal commit­ments; all great theories do and must.  {167} 

We should use this perspective of engagement with the core commitments to assess the relative theoretical importance of issues now commanding at­tention among evolutionists. For example, Kimura's theory of neutralism (1983) ranks as fundamental and reformative for proposing a new domain of causation at high relative frequency. I regard as unfair, and disrespectful of Darwin's clear commitments, the common rhetorical strategy of arguing, as Stebbins and Ayala did for example (1981), that selection and neutralism should be judged as competing paradigms comfortably embraced within the Modern Synthesis. The Synthesis, as an intellectual structure, has always been understood as Darwinism strengthened by modern knowledge about ge­netics and heredity. The Synthesis must therefore assert a dominant relative frequency for selection. Of course such a theory allows for neutrality — one could scarcely deny either the mathematics or the conditions of potential op­eration — but only at a low relative frequency, so that the preeminence of se­lection will remain unchallenged.

Kimura's claim for high, even dominant, relative frequency of neutral change at the nucleotide level introduces a world different from Darwin's. At most, one can say that this world, largely invisible at the organismic level, does not subvert Darwin's proposal that selection dominates the phenotypic realm of overt form, function, and physiology of organisms. But in so saying, we must admit that a large part of reality, though unaddressed by Darwin himself, cannot be explained on Darwinian principles if Kimura's theory holds. Darwinism does not fall thereby, but a new and distinct domain, pri­marily regulated by a different style of causality, has been added to evolution­ary explanation. How can one deny that evolutionary theory becomes sub­stantially reformulated and enriched thereby? Why would one want to issue such a denial, unless psychic health depends upon the continued assertion of comfortable orthodoxy, whatever the required twist of logic?

My own expertise lies in paleontology, and this book shall emphasize critiques from the attendant domain of macroevolution, descriptively defined as patterns and causes of evolution at and above the species level. (I acknowl­edge, of course, the fascination and transforming power of work at the mo­lecular level. I also recognize that macroevolution must shake hands with mo­lecular genetics in order to forge the new consensus. If this book slights the molecular side, my own ignorance stands as the only cause, and this work necessarily suffers thereby.)

Basically, I shall defend the view that each leg of Darwin's essential tripod, as explicated in this chapter, now faces a serious critique from the domain of macroevolutionary change. These critiques rank as auxiliaries to Darwin­ism in Kellogg's sense; for they either expand or add to the core commit­ments. But the expansions are large and the additions substantial — so the resulting revision can no longer be called ordinary Darwinism in any conven­tional meaning. I am convinced that the three critiques intertwine in a poten­tially unified way. But consensus is premature and we can only see the result­ing shape of the revised and unified theory through a glass darkly — though in the future, no doubt, face to face.  {168} 

Proceeding in reverse order through critiques of Darwinism's three core claims, catastrophic mass extinction, and more general views about fortu­ity in abiotically driven extinction at all levels, challenge Darwin's essential notion of a dominant relative frequency for biotic struggle in a crowded world — the third leg of the tripod, as represented by the geological stage re­quired for an evolutionary play based entirely on extrapolation of microevolutionary principles (Chapters 6 and 12). The general idea of constraint — more in the positive sense of internally biased channels for change, rather than the negative meaning of limited variation for potentially useful alter­ations (see Gould, 1989a) — rejects the key Darwinian notion of isotropy in raw material, and consequent control of evolutionary direction by natural se­lection. Constraint therefore challenges the second leg of the tripod — the “creativity of natural selection” — not by confuting the proposition that natu­ral selection acts as a creative force, but by insisting on diminished relative frequency and a sharing of control. Moreover, by reasserting the structuralist side of the old dichotomy between structure and function in biology — an is­sue far predating evolution, and inherent in the struggle between continental vs. Paleyan approaches to natural theology — the idea of constraint reengages one of the deepest issues in all the life sciences (Chapters 45 and 1011).

Most importantly, and as the best integrator of all three critiques, the hierarchical theory of natural selection, by asserting both the existence and rela­tive importance of selection at all levels from genes to species, challenges the first leg of the tripod — the insistence, so crucial to Darwin's radical overthrow of Paley via Adam Smith, that selection works almost exclusively on organ­isms (Chapters 3 and 89). I believe that this hierarchical theory provides the most fundamental, and potentially unifying, of all critiques — for I suspect that many constraints will be explained as effects of lower level selection indi­rectly expressed in phenotypes; while the contribution of mass extinction to repatterning life's history will include a crucial component of selection at lev­els above the organismic. Moreover, the attendant need to reconceptualize trends and stabilities not as optimalities of selection upon organisms alone, but as outcomes of interactions among numerous levels of selection, implies an evolutionary world sufficiently at variance from Darwin's own conception that the resulting theory, although still “selectionist” at its core, must be rec­ognized as substantially different from current orthodoxy — and not just as a dash of spice on an underflavored dish. I therefore devote the largest section of this book's second half (Chapters 8 and 9) to defining and defending this hierarchical theory of selection.

If the next generation of evolutionists follows and extends this protocol at the outset of our new millennium, as presaged by the tentative work and ex­ploration of so many scientists at the close of the last millennium, then we shall honor, all the more, the vitality of the tight definitions and firm commit­ments proposed by Darwin himself at the foundation of our discipline. Few theories hold the range of power, and the intricacy of logic, necessary to gen­erate an intellectual structure of such continuing fascination and relevance. We do not pay our proper respect to Darwin by bowing before the icons of  {169}  his central propositions, but by engaging these focal precepts as living pres­ences, ripe for reformulation, almost 150 years after their initial presentation. In Darwin's own world of continuous flux, anything that lasts so long be­comes a many-splendored thing. In a revised world of structuralism, we might say that Darwin first located and embellished one of the few brilliant and coherent positions in an intellectual universe with few nucleating places. Either formulation engenders the same result of abiding respect for Darwin's view of life — leading to proper thanks owed by all of us for the good fortune of such an interesting founder. What greater pleasure can we know than to engage Darwin in dialogue — as we can and must do, because his theory rests upon a powerful and defining essence. Darwin, in short, is the extraordinary man who, all by himself, embodied the only three beings proclaimed worthy of respect by Baudelaire — for he pulled down an old order, and came to know a large part of the new world that he created. Il n'existe que trois etres re­spectables: le pretre, le guerrier, le poete. Savoir, tuer, et creer. There exist only three beings worthy of our respect: the priest, the warrior, and the poet. Know, kill, and create.



Seeds of Hierarchy

Lamarck and the Birth of Modern Evolutionism in
Two-Factor Theories


In 1793, the French revolutionary government, having expunged the past by executing a monarch, proclaimed a new beginning of time. They renamed the months, and started the calendar all over again with the foundation of the Republic in September 1792. The old months had honored emperors and gods, but the new months would celebrate the passing of seasons by weather and activity — Brumaire (the foggy month in fall), Thermidor (the hot times of mid summer), and Nivose (for the depth of a frosty winter), for example.

Jean-Baptiste-Pierre-Antoine de Monet, Chevalier de Lamarck (1744-1829) — now redesignated, with democratic brevity, as Citoyen Lamarck — became professor of “lower” animals (the old Linnaean classes Insecta and Vermes, later renamed “invertebrates” by Lamarck himself) at the newly founded Museum d'Histoire Naturelle in 1793. (His previous work, nearly all in botany, had not prepared him for this new role, though he had long been an avid shell collector and student of conchology.) Until 1797, he had supported the conventional idea of species as fixed entities. But he then be­came an evolutionist, first expressing this new view of life in his inaugural lec­ture for the Museum course of 1800, and then in three major works — the Recherches sur I'organisation des corps vivans of 1802; his most famous work, the Philosophie zoologique of 1809; and the Histoire naturelle des animaux sans vertebres of 1815-1822.

In an ironic symbol, Lamarck first presented his evolutionary theories in an inaugural lecture pronounced on the 21st day of Floreal, year VIII (May 11, 1800) — in the month of flowering. For Lamarck's theory suffered the oppo­site fate of withering, and the scorn of inattention. We all know the image of Lamarck — an impression carefully nurtured, for different reasons, by friends and foes alike — as a lonely man (a prophet before his time to some, a kook to others), penniless, friendless, and, finally, blind; living out the last days of a long and sad life, supported only by his devoted daughters.

This image of a forgotten failure was fostered by the two greatest figures of 19th century natural history — first by Cuvier, and later by Darwin. Darwin said little about Lamarck (see pp. 192197), but his derision still permeates  {171}  our view. Cuvier did far more damage. I don't know what lip service Cuvier gave to the ancient maxim de mortuis nil nisi bonum (say only good of the dead), but he violated this precept with avidity in writing eloges (eulogies) of deceased colleagues. Cuvier, the consummate politician, understood the power thus granted to shape history in his own favor. For what forum could be less subject to rebuttal, and therefore more suited for easy passage into re­ceived truth. As master of eloges, Cuvier held enormous power over his col­leagues, as long as he could outlive them! (see pp. 309312 on Geoffroy's re­venge for the same reason). His official eloge of Lamarck is a masterful, if repugnant, document of propaganda directed against a close colleague and former friend who had (in Cuvier's view) gone beyond the pale in both meth­odology of research and content of belief. Cuvier used his eloge as an oppor­tunity to castigate Lamarck, and thus provide a lesson in proper procedure for aspiring scientists.

Cuvier began with cloying praise, and then portrayed his need to criticize as a sad duty: “In sketching the life of one of our most celebrated naturalists, we have conceived it to be our duty, while bestowing the commendation they deserve on the great and useful works which science owes to him, likewise to give prominence to such of his productions in which too great indulgence of a lively imagination had led to results of a more questionable kind, and to indi­cate, as far as we can, the cause or, if it may be so expressed, the genealogy of his deviations” (1832, 1984 edition, p. 435).

Cuvier then dismembered Lamarck on two grounds. First, with justice in the claim (however unkind the rhetoric), he castigated Lamarck for reaching too far without foundation, and for building all-encompassing systems in the speculative mode. (This criticism reflected Cuvier's main unhappiness with Lamarck's science. Cuvier viewed himself as a modernist, committed to rigor­ous empirical documentation, and no extension beyond direct evidence in the search for explanations — as opposed to Lamarck's unfruitful, comprehensive speculation in the antiquated esprit de systeme, or spirit of system): “He had meditated on the general laws of physics and chemistry, on the phenomena of the atmosphere, on those of living bodies, and on the origin of the globe and its revolutions. Psychology, and the higher branches of metaphysics, were not beyond the range of his contemplations; and on all these subjects he had formed a number of definite ideas . . . calculated to place every branch of knowledge on a new foundation” (1832, 1984 edition, p. 442).

Cuvier acknowledged Lamarck's excellent efforts in morphology and taxonomy, but then damned him for denigrating this admirable work as a trifle compared with all-embracing and useless theories. What a sorry spectacle: Lamarck in his armchair, challenging the great Lavoisier, the icon and martyr of true science. (Lavoisier was beheaded during the Reign of Terror.) “So inti­mately did he identify himself with his systems, and such was his desire that they should be propagated, that all other objects seemed to him subordinate, and even his greatest and most useful works appeared in his own eyes merely as the slight accessories of lofty speculations. Thus, while Lavoisier was creat­ing in his laboratory a new chemistry, founded on a beautiful and methodical  {172}  series of experiments, M. de Lamarck, without attempting experiment, and destitute of the means of doing so, imagined that he had discovered another” (1832, 1984 edition, p. 442).

After ridiculing Lamarck's general method of system building, Cuvier mounted his second attack and dismembered the particular content of Lamarck's system, especially his evolutionary views. Cuvier did his former colleague a lasting disservice by caricaturing Lamarckian evolution, as the outcome of organic will, based on desires, translated into phyletic progress. Cuvier's rhetoric was brilliant, his characterization grossly distorted:

Wants and desires, produced by circumstances, will lead to other efforts, which will produce other organs. ... It is the desire and the attempt to swim that produces membranes in the feet of aquatic birds; wading in the water, and at the same time the desire to avoid wet, has lengthened the legs of such as frequent the sides of rivers . . . These principles once admitted, it will easily be perceived that nothing is wanting but time and circumstances to enable a monad or a polypus gradually and indiffer­ently to transform themselves into a frog, a stork, or an elephant (1832, 1984 edition, p. 446).

Finally, in an ultimate dismissal from a “hard” scientist (and with the tone of the Yahoo), Cuvier concludes: “A system established on such foundations may amuse the imagination of a poet; a metaphysician may derive from it an entirely new series of systems; but it cannot for a moment bear the examina­tion of anyone who has dissected a hand ... or even a feather” (1832, 1984 edition, p. 447).

Cuvier's caricature remains potent in our worst modern misunderstanding of Lamarck as a mystical vitalist, advancing the idea of an ineffable organic will against the ordinary physical causality of science. (Tit for tat, and how­ever unfairly, Lyell hurt Cuvier even more in return by caricaturing him as a theologically tainted, antiscientific catastrophist in geology.) But Lamarck, schooled (along with Cuvier) in the ideals of the French Enlightenment, was an ardent materialist. His idiosyncratic and unfruitful views about the nature of matter (arising primarily from his anti-Lavoisierian chemistry) led to pre­dictions of odd behavior for living bodies, but his basic notions of reduction and causality remained in the scientific mainstream. In his last great work, and in the context of his evolutionary theory, Lamarck defended a conven­tional view of mechanistic causality, and derided all teleological interpreta­tions. Goals, he argued, are false appearances reflecting an underlying causal necessity:

It is chiefly among the living, and most notably among Animals, that some have claimed to glimpse a purpose in nature's operations. Even in this case the purpose is mere appearance, not reality. Indeed, in every type of animal organism, there subsists an order of things ... whose only effect is to lead to what seems to us to be a goal, but is essentially a neces­sity. The order achieves this necessity through the progressive development  {173}  of parts, which are [also] shaped by environmental conditions (1815, in Corsi, 1988, p. 190).

Since watchdogs tend to be more vigilant than publicists, Lamarck's opponents among the natural theologians often noted (and deplored) his materialism. The pious Reverend William Kirby, one of Britain's greatest entomolo­gists, made a statement that I regard as both trenchant and descriptively accurate (in Burkhardt, 1977, p. 189): “Lamarck's great error, and that of many other of his compatriots, is materialism; he seems to have no faith in anything but body, attributing every thing to a physical, and scarcely any­thing to a metaphysical cause. Even when, in words, he admits the being of a God, he employs the whole strength of his intellect to prove that he had noth­ing to do with the works of creation. Thus he excludes the Deity from the government of the world that he has created, putting nature in his place.”

Curiously, each generation of historians and biological commentators has to discover, anew and for itself — and by reading original sources rather than imbibing mythology — this general and mainstream scientific position held by a man with such idiosyncratic views on specific subjects (see Mayr, 1972, and Simpson, 1961a, for the scientists; Gillispie, 1959; Burkhardt, 1977; and Corsi, 1988, for historians). For example, Gillispie wrote in his classic article for the Darwinian centenary (1959, p. 275): “Life is a purely physical phe­nomenon in Lamarck, and it is only because science has (quite rightly) left be­hind his conception of the physical that he has been systematically misunder­stood and assimilated to a theistic or vitalistic tradition which in fact he held in abhorrence.”

This correction allows us to see Lamarck as a key figure in and of his time — an age as rife with intellectual, as with political, ferment — and not as a painfully peripheral, and actively marginalized, oddball. In a meticu­lous analysis of French scientific thought, Corsi (1988) has placed Lamarck's views firmly amidst the debates of his age, and also demonstrated that his the­ories were not so ignored or ridiculed as tradition maintains.

I am not arguing that Lamarck was popular in his day, only that he was contemporary. In many ways, Lamarck became his own worst enemy, and he owed his fall from favor towards obscurity as much to his own unfortunate habits as to the peculiarity of his ideas. He possessed no political skills, and could only fare badly in any match with the masterful Cuvier (in an age that must rank as the best and the worst of all political times). He continued to practice the old style of speculative system building in an increasingly empiri­cal climate. He was combative, and so self-assured, that affirmation without any documentation became his principal style of argument. Consider this claim for use and disuse from the Philosophie zoologique (1809, 1984 edi­tion, p. 108): “Nothing of all this can be considered as hypothesis or private opinion; on the contrary, they are truths which, in order to be made clear, only require attention and the observation of facts.” Lamarck's certainty ex­tended even to the maximally dubious subject of weather forecasting: “I am not submitting an opinion, but announcing a fact. I am indicating an order of  {174}  things that anyone can verify through observation” (in Corsi, 1988, p. 59). The old story that Napoleon refused a copy of Lamarck's Philosophie zoologique is apparently true (unlike most legends in the history of science). But Napoleon's motive has not generally been recognized: he mistook the nature of the gift, thinking that he had been offered one of Lamarck's discredited volumes of weather predictions for the coming year!


The preceding section on Lamarck as a man of his time may seem peripheral, if not wholly out of place, in a chapter on hierarchical causation in evolution­ary theory, but this theme holds a definite place in the logic of my presen­tation. Such a diffuse and comprehensive idea as evolution can claim no sin­gle initiator or unique starting point. The search for precursors in ancient Greece, while overextended (Osborn, 1894), rests upon a legitimate founda­tion. But Lamarck holds a special place as the first to transcend footnote, pe­ripheral commentary, and partial commitment, and to formulate a consistent and comprehensive evolutionary theory — in Corsi's words (1988, p. xi) “the first major evolutionist synthesis in modern biology.”

Moreover, even in a book parochially skewed to British and American evolutionary theory, Lamarck still merits the status of an ultimate source. Ger­man and French biologists could cite a variety of references from their in­digenous movements of Naturphilosophie (Oken, Meckel, or Goethe himself, for example) and the revolutionary times of the Age of Reason (Buffon, Maupertuis, Diderot, and a host of largely forgotten Enlightenment figures). England could boast a few precursors (including Darwin's grandfather Eras­mus), but no strong movements. Ironically, as Darwin, Wallace, and all the great mid-century evolutionists acknowledged, Lamarck instigated both ma­jor treatments of evolutionary thought in English before 1859: first, the accu­rate and extensive, if negative, presentation of Lamarck's system by Charles Lyell in the first four chapters of The Principles of Geology, Volume 2 (1832); and second, the anonymously published (1844) Vestiges of the Natural His­tory of Creation. The author of that scandalous and widely debated book, the Scottish publisher Robert Chambers, acknowledged Lamarck, via Lyell, as his major source of inspiration.

I have (see Chapter 1) rejected Hull's genealogical approach to the defini­tion of theories, but I certainly defend this criterion (almost tautologically, I suppose) for the tracing of influences. Of Lamarck's foundational impact on English evolutionary thought, Hull (1985, p. 803) writes: “Darwin first con­fronted a detailed explication of the species problem in the context of Lyell's refutation of Lamarck in his Principles of Geology . . . Others like Spencer and Chambers were converted by reading Lyell's refutation of Lamarck; still others like Wallace and Powell were led to entertain the possibility of evolu­tion by reading Chambers.” We cannot, in short, view Lamarck as an oddity, cast aside by his own contemporaries, and irrelevant except as a whipping boy ever since. And we must acknowledge that Lamarckism, properly de­fined, forms a coherent and innovative system in the context of its own time.  {175} 

Lamarck's active centrality provides a foundation to my historical argu­ment because his theory, as presented in the next section, rests upon the con­cept of hierarchy, with distinct causes at two primary levels. Lamarck's hier­archy differs radically in form and logic from any acceptable modern version; indeed, I shall reject the basis of Lamarck's notion as an important compo­nent in developing the modern interpretation.

Lamarck's concept became Darwin's context. In perhaps the most widely quoted of all his letters, Darwin wrote to Hooker on January 11, 1844 (in F. Darwin, 1887, volume 2, p. 23): “I am almost convinced (quite contrary to opinion I started with) that species are not (it is like confessing a murder) im­mutable. Heaven forfend me from Lamarck nonsense of a 'tendency to pro­gression,' 'adaptations from the slow willing of animals,' etc.! But the conclu­sions I am led to are not widely different from his; though the means of change are wholly so. I think I have found out (here's presumption!) the sim­ple way by which species become exquisitely adapted to various ends.”

Hierarchy has resided at the heart of evolutionary theory from' the very beginning, despite a temporary eclipse during the rally-round-the-flag pe­riod of strict Darwinism at the middle of the 20th century. When the Beagle docked at Montevideo, Darwin received his most precious item of mail — volume two of Lyell`s Principles of Geology. His joy at this gift, and his careful study of the contents, are well attested. This volume began with a long and careful exposition of Lamarck's theory, fairly but negatively described by Lyell. Darwin formulated his focal concept of small-scale change, based on organismal advantage as the mechanism (by extrapolation) for all evolu­tion, as an explicit denial of Lamarck's hierarchy of causes. I believe that Darwin correctly rejected an untenable theory of hierarchy based on distinct causes for different levels, but that (in a historically portentous example of the cliche about babies and bathwater) he carried a good thing too far by dismissing the general concept entirely. I conceived this book — The Structure of Evolutionary Theory — both as a celebration of Darwin's exemplary tough­ness, and as a call for the reinstitution of causal hierarchy, properly refor­mulated.


In the short period of 1797 to 1800, beginning with the Directory in power and culminating in Napoleon's coup d'etat of 18th Brumaire year VIII (No­vember 9,1799), Lamarck became an evolutionist and constructed the major features of his theory. Scholars have identified many sources as Lamarck's pri­mary impetus — his developing views on spontaneous generation, his work on living and fossil shells (Burkhardt, 1977), the implications of his unconven­tional theories in physics and chemistry (Corsi, 1988). But I wish to present the logic, rather than the ontogeny, of his final and completed argument.

Lamarck's evolutionary system attempts to marry two sets of ideas, each embodying a primary module of his conceptual world. These two sets com­mingle at their edges, but their distinction establishes the basis of Lamarckism,  {176}  and provides a hierarchical context for this first comprehensive attempt to formulate an evolutionary theory.

The first set: environment and adaptation

The first set focuses on adaptation, and links this key attribute of organisms to the history of environments, the general pace of change, and the intimate relationship between physical and biological worlds through time. (Corsi, 1988, grants primacy to this set in the ontogeny of Lamarck's developing ideas; I accept this assertion but note that the same set, curiously, becomes secondary in the logic of Lamarck's fully formulated argument.) The frame­work can be entered in several places, with the rest of the edifice following by implication from a few basic premises. Lamarck's views on extinction provide a good beginning.

In opposition to his colleague Cuvier, and acting as a major source of their estrangement, Lamarck denied that true extinction (defined as termination of genealogical lines) could occur — though he allowed an exception for large quadrupeds wiped out by human predation. (Cuvier, on the other hand, embraced extinction both as the foundation of geological ordering, and as a cardinal indication that animals cannot evolve to match changing environ­ments.) Yet, as a molluscan paleontologist, Lamarck knew that the morphol­ogies of organisms within major taxonomic groups changed in an orderly manner through time. Evolution of outward form, with consequent preserva­tion of lineages from extinction, represents the only alternative to termina­tion of lineages followed by creation of new and different morphologies.

Lamarck far out-Lyelled Lyell in his commitment to uniformitarian geol­ogy (an ironic fact, given Lyell's lambasting of Lamarckian biology in his own treatise on geological uniformity). Lamarck's geological volume, the Hydrogeologie of Year X (1802), may strike us as bizarre in several particular as­sertions; but his general commitment to uniformity cannot be denied as a primary feature. Lamarck would admit no causes not now observable in op­eration; in particular, no paroxysms or catastrophes beyond the range of modern effects. He adopted Hutton's rigidly ahistorical vision (see Gould, 1987b) and postulated a geological history ruled by aqueous erosion (hence the title of his work). Cycles of construction and erosion unfold so many times, and in so similar a manner, that individual moments lose any distinct­ness, given past and future repetition of their features. Ocean waters carve mountains and continents (though Lamarck made an exception for volcanoes built by magmas). Currents tend to flow from east to west, and continents therefore erode on their eastern borders and accrete by deposition at their western edges. In a sense, therefore, continents undergo a slow westward march around the globe. This curious circumnavigation has occurred several times during the earth's extended history. But why doesn't the process yield directionality as erosion wears continents down to flat plains permanently be­low water? Lamarck countered with his distinctive mineralogical thesis: all rocks arise as ultimate products of organic deposition. Erosion may break continents into tiny comminuted grains, bits, and dissolved material; but as  {177}  long as organisms maintain their steady state of abundance, these raw materi­als will be taken up again and redeposited as new rocks fashioned from the products of life.

In summary, if lineages could not become extinct, if climate and geology changed in a continuous and insensibly gradual manner throughout geologi­cal time, and if the forms and functions of organisms always matched the fea­tures of their local environments, then gradual, adaptive evolution becomes a logical necessity.

But by what mechanism will this ineluctable evolution occur? In particular, since steady, continuous alteration of environment provides the impetus for organic change, how does information flow from new environments to mod­ify the old forms of organisms? Lamarck's answer to this riddle — building only one corner of his complete system — invokes the familiar ideas that later generations would call “Lamarckism” when the rest of his edifice had been forgotten. Lamarck begins by formulating the central principle of his func­tionalist credo — the counterintuitive statement, later embraced by Darwin as well, that form follows function as the order of life's history. When we con­template any adaptation of an organism, and consider the intricate correla­tion of form with function, we naturally assume (or so Lamarck asserts) that form comes first, and that function can only follow. (God makes a wing, and a bird can then fly, to cite a nonevolutionary example.) But Lamarck ad­vanced the paradoxical reverse order as his key premise: new habits lead to altered structures.*

In Lamarck's proposed mechanisms, environment changes first. Indeed, environment changes slowly and continuously on our uniformitarian planet. “Every locality,” Lamarck writes (1809, p. Ill), “itself changes in time as to exposure, climate, character and quality, although with such extreme slow­ness, according to our notions, that we ascribe it to complete stability.” Or­ganisms must accommodate to these changes by alterations in their habits — chewing with greater strength if the food gets tougher, moving more vigor­ously if the temperature gets colder. These altered habits, if long sustained, must feed back upon the organism in the guise of altered morphology or physiology — a thicker beak to crack the harder seeds, longer hair on a tougher skin to resist the cold.

At this point in the argument, the famous “Lamarckian” theory of inheri­tance finally enters. As many scholars have documented, “soft,” or “La­marckian” inheritance represented the folk wisdom of Lamarck's time, and cannot be regarded as an innovation of the Philosophic zoologique. There­fore, the restriction of “Lamarckism” to this relatively small and non-distinc­tive corner of Lamarck's thought must be labeled as more than a misnomer,  {178}  and truly a discredit to the memory of a man and his much more compre­hensive system. In any case, the changes wrought by new habits during an organism's lifetime can be passed directly to offspring in the form of altered heredity. Soft inheritance may have been the standard belief of the time, but Lamarck certainly recognized its crucial and particular role in his system. He wrote with his characteristic lack of doubt (1815, in Burkhardt, 1984, p. xxix): “The law of nature by which new individuals receive all that has been acquired in organization during the lifetime of their parents is so true, so striking, so much attested by the facts, that there is no observer who has been unable to convince himself of its reality.” Lamarck abstracts his idea of inher­itance as two principles, usually printed in italics in his texts to emphasize their importance, and known ever since as:

• use and disuse

• the inheritance of acquired characters (1809, volume 1, p. 113)

Even if this theory of inheritance ranked as folk wisdom of the day, Lamarck's revolutionary statement, one of the great transforming insights in the history of human thought, resides in the preceding principle that trans­lates this mode of inheritance into a theory of evolution — the triggering of change in form by prior alterations in behavior. Lamarck clearly recognized the central role of this claim, for he always cited this counterintuitive se­quence of causality — from altered environments to changed habits to modi­fied form — as the linchpin of his entire system. In the Philosophie zoologique, he quotes his own earlier statement from the Recherches of 1802: “It is not the organs, that is to say, the nature and shape of the parts of an animal's body, that have given rise to its special habits and faculties; but it is, on the contrary, its habits, mode of life and environment that have in course of time controlled the shape of its body, the number and state of its organs and, lastly, the faculties which it possesses” (1809, p. 114). Lamarck then makes his threefold causal chain — environment to habits to form — even more explicit (1809, p. 126): “This is a fact that can never be disputed; since nature shows us in innumerable other instances the power of environment over habit and that of habit over the shape, arrangement and proportions of the parts of animals.”

Causality might run from altered environment to changed organism, but Lamarck insisted that he did not view organisms as passive writing slates, ripe for inscription by the modifying hand of environment. Environmental change translates to adaptation of form only through the intermediary of or­ganic action expressed, in higher creatures at least, as altered habits: “What­ever the environment may do, it does not work any direct modification what­ever in the shape and organization of animals. But great alterations in the environment of animals lead to great alterations in their needs, and these al­terations in their needs necessarily lead to others in their activities. Now if the new needs become permanent, the animals then adopt new habits which last as long as the needs that evoked them” (1809, p. 107).

These statements about the responses of animals to “felt needs” (Lamarck  {179}  used the word besoins) left Lamarck open to charges of mystical vitalism when distorted for rhetorical purposes by Cuvier, or approached with over­wrought caution by Darwin. One might caricature this part of Lamarck's sys­tem by saying that a giraffe felt a need for a long neck, stretched ever so hard, and then passed the results of these successful efforts directly to offspring. But a fair assessment of Lamarck's actual words shows that he advocated no inef­fable willing, but only the commonplace idea that a change in environment can, in an almost mechanical way, elicit an organic response in terms of al­tered habits: “Variations in the environment induce changes in the needs, habits, and mode of life of living beings . . . these changes give rise to modi­fications or developments in their organs and the shape of their parts” (1809, p. 45).

This first set of Lamarckian ideas contains nothing that should have of­fended Darwin, while several points embody the deeper functionalist and adaptationist spirit of the Darwinian view of life. Darwin did not grant such crucial emphasis to soft inheritance, although he accepted the principles of use and disuse and inheritance of acquired characters, and he awarded them a subsidiary role in his own theory. But two key items in this first set might be designated as decidedly Darwinian in spirit, if only because they advance and presage two of the half dozen most important ideas in Darwin's theory: the uniformity of environmental change,* and the functionalist first principle that change of habit sets the pathway to altered form. The mechanisms of change differ to be sure — altered habits establish new selection pressures for Darwin, but induce heritable modifications more directly for Lamarck — but both thinkers share a functionalist commitment.

I would argue that the structuralist-functionalist dichotomy precedes any particular theory of mechanism within either camp. Thus, we may view La­marck and Darwin as occupying the common ground of functionalism, with their differing mechanisms of natural selection and soft inheritance as ver­sions of the same deeper commitment. Therefore, if Lamarckism only encom­passed this first set of ideas, we might interpret Lamarck as the inception of a smooth transition to Darwin. But Lamarckism also includes a second set of concepts, which, when combined with the first set into Lamarck's full system, builds an evolutionary theory truly opposed to Darwin's chief theoretical concept and operational principle as well.

The second set: progress and taxonomy

The first set, by itself, leads to a logical dilemma for Lamarck's view of life and his professional commitments. Adaptation to changing local environ­ments may be well explained, but Lamarck's truly ahistorical uniformitarianism implies that life can manifest no progress, or no linear order at all, if ad­aptation matches creatures to an environmental history without direction.  {180}  (This issue arose for many environmental determinists in both creationist and evolutionist camps. Buckland, and most of his catastrophist colleagues, maintained their allegiance to life's increasing perfection by positing a direc­tional history of environmental change — increasing inclemency, for exam­ple, requiring improvement in organic design to meet the growing challenge. This option was not open to Lamarck, who espoused a steady state, non-directionalist geology.)

Yet Lamarck firmly advocated a taxonomic ordering of organisms by the conventional scheme of increasing perfection in organization. This subject greeted him on a daily basis, for he held the post of curator for invertebrates at the Museum in Paris, and his yearly courses featured this organizing theme of linear order. (As a pedagogic device, Lamarck usually started with humans, as the “highest” creature, and then discussed the rest of nature as degrada­tion from maximal complexity. He defended this procedure, even in his evo­lutionary writings, as a method for teaching, even though historical order had actually moved from simple to complex — for he argued that one must under­stand the full and final possibilities before grasping the imperfect and incipi­ent beginnings.)

Lamarck argued that a second set of forces, distinct from the causal flow of environment to organism, produced nature's other primary pattern of advancing complexity. But this claim for an efficient and universal cause of progress engendered another dilemma: why, on our present and ancient earth, do some organisms still maintain the simplest anatomies? Why were these forms not pushed up the ladder of complexity ages ago? Lamarck resolved this problem with the last major argument of his full system — con­tinuous spontaneous generation. New life continues to arise from chemical constituents; these simple forms begin their march up the ladder, while re­placements at their lowly status continue to form anew. (Thus, in a curious sense, as Simpson and others have noted, Lamarck's evolutionary system op­erates as a grand steady state, even as any particular bit of protoplasm moves on a historical path up the sequence. The ladder of life really operates as a continuous escalator, with all steps occupied at all moments. The simplest forms continue to arise by spontaneous generation from chemical constitu­ents formed by the breakdown of higher creatures upon their individual deaths.)

Lamarck argued that his unconventional chemistry, emphasizing the role of fire and the motions of subtle fluids, engendered these two central phenom­ena — spontaneous generation and progress up the ladder — as consequences of deeper physical principles. Lavoisier had destroyed the old quadripartite taxonomy of air, water, earth, and fire in developing his theory of chemical el­ements. Lamarck opposed the “new chemistry” by asserting the old taxon­omy, and his own distinctive claim for the primacy of fire. Much of Cuvier's disdain focused not on Lamarck's biology, but on his allegiance to this anti­quated chemistry.

Lamarck, who excelled in crisp assertion but not in clear exposition, never fully specified why chemicals should aggregate to life, or what subtle motions  {181}  of physical fluids would build the increasing complexity of anatomy. He held that the products of spontaneous generation arose as small, soft and un-differentiated primal forms. The complexifying force — which Lamarck usu­ally called le pouvoir de la vie or la force qui tend sans cesse a composer I'organisation — resides in the motion of fluids and their inevitable tendency to carve channels, sacs, and passageways in soft tissues. This process, ex­tended through time, gradually builds ever greater complexity. In the Histoire naturelle of 1815, Lamarck offers his most explicit statement about this pro­cess: “As the movement of fluids . . . accelerates, the vital forces would grow proportionately, and so will their power. The rapid motion of fluids will etch canals between delicate tissues. Soon their flow will begin to vary, leading to the emergence of distinct organs. The fluids themselves, now more elaborate, will become more complex, engendering a greater variety of secretions and substances composing the organs” (1815, in Corsi, 1988, p. 189).

Lamarck did clearly assert that these internal carvings of complexity main­tained a relentless and intrinsic causal basis separate from the apparatus of response to “felt needs” used in building adaptations to changing local environments. He contrasted the two sets of forces in writing: “There exists a variety of environmental factors which induces a corresponding variety in the shapes and structure of animals, independent of that special variety which necessarily results from the progress of the complexity of organization in each animal” (1809, p. 112). He also stated that the entire escalator of com­plexity could run a full course in a constant environment: “If nature had given existence to none but aquatic animals and if all these animals had al­ways lived in the same climate, the same kind of water, the same depth, etc., etc., we should then no doubt have found a regular and even continuous gra­dation in the organization of these animals” (1809, p. 69).

Lamarck therefore proposes two distinct sets of forces to construct what he regarded as the two preeminent features of life — progress in linear order, and adaptation to environment. The interactions of these sets — not the causes or properties of either one — establish the foundation of Lamarckism, properly defined in his own expansive terms.

Distinctness of the two sets

I shall argue in the next section that these two sets of concepts must be regarded as both logically distinct and opposed in Lamarck's system. My basis for regarding Lamarckism as a theory of hierarchy lies in this division. Lamarck, as we shall see, always presents the two sets as separate in his later evolutionary writing, and scholars of Lamarckism have accepted this contrast as crucial (Burkhardt, 1977; Mayr, 1972; Simpson, 1961). But Lamarck, as noted several times above, remains a frustrating figure for historians. His as­sertions are bold, even dogmatic; but his arguments tend to be sketchy, full of elisions, or even self-contradictory. These frustrations become most apparent in Lamarck's treatment of his two primary forces (as Corsi, 1988, has dis­cussed with great insight). The explicit assertions of his later works rank the two forces as distinct and opposed, but both the ontogeny and logic of  {182}  Lamarck's argument shows more “leakage” than his words would suggest. Consider the following, as stressed by Corsi and Burkhardt:

Ontogeny. Although Lamarck presents the forces of adaptation as devi­ations from, and therefore secondary to, the primary causes that build com­plexity, he apparently developed his mechanism for progress from his previ­ously formulated ideas about adaptation (Corsi, 1988; and newly discovered evidence in Gould, 2000d). Still, the psychological source of a theory needn't map its eventual logical structure, and this point, while interesting, scarcely compromises the distinctness and ranking of the two sets.

Causation. At several points, Lamarck breaches the boundaries be­tween his sets in discussing causation.

(1) Soft inheritance works in both sets. Whether an organism becomes more complex because fluids carve channels by intrinsic chemistry, or be­comes better adapted because habits change in response to altered environ­ments, the acquired features must still be passed to offspring by direct inheri­tance. Still, a common mechanism may work in two modes, and this linkage does not compromise Lamarck's claim for distinctness.

(2) The style of action for soft inheritance in adaptation depends upon the state of complexity engendered by opposing forces of progress. Lamarck di­vided organisms into three ascending groups designated, in the old Aristote­lian terms, as insensitive, sensitive, and rational. The first group, too simple to mount a creative response to external change, reacts to environment not by altered habits, but by direct influence. The capacity for active response, Lamarck's famous sentiment interieur, only arises in the second group and unleashes the tripartite causal sequence of changed environment to altered habit to modified form.

(3) The real blurring occurs when we try to make sense of Lamarck's claim that forces of progress can build the entire sequence from infusorian to com­plex vertebrate without any environmental change. Lamarck surely makes this assertion explicitly, without hesitation (see citation on p. 187), and the distinctness of his two forces depends upon this potential independence. But Lamarck does not work out a consistent justification, and several frustrated historians have even argued that he could not have done so without con­tradiction — that his system, in other words, suffers from a true defect in logic of argument on this point.

The simplest organisms, Lamarck states, are carved out and complexified by “subtle” and “imponderable” fluids — caloric and electricity in his system. These fluids work in their intrinsic way to produce increased complexity. But as animals differentiate and harden, fluids must flow in preset channels; the weak imponderables then lose their power to mold, and the body's own pon­derable fluids must assume this role. (Lamarck locates this transition at about the echinoderm grade of organization.) At this level, the “power of life” should become inoperable without an impetus from environmental change — and the two sets of forces should therefore commingle. Protected inside a rigid body, and constrained to flow in preexisting channels, how can the ponderable  {183}  fluids produce further advances in complexity unless changed en­vironments elicit altered habits, thus modifying form and permitting the pon­derables to flow in new ways (see Corsi, 1988, p. 200; and Burkhardt, 1977, p. 147).

Pattern. The “pure” distinction of progress and adaptation should pro­duce a single linear chain (for progress) with lateral deflections (for adapta­tion). Lamarck tried to construct such a topology, but could not carry his scheme to completion at two important places — the top and bottom of the ladder — where environment intruded upon the chain to blur the distinction of forces.

(1) Two sequences of spontaneous generation. Lamarck first proposed a single linear series of animals, starting with the spontaneous generation of infusorians (protistans) as free-living creatures in water. These unicells then aggregated to polyps and their relatives, and then to simple, bilaterally symmet­rical worms (see Fig. 3-1). However, Lamarck later discovered worms (the acoelous platyhelminths in modern terminology) without nerve cords. These worms ranked “higher” than polyps on grounds of their mobility, but could not be the descendants of polyps, unless the nerve cords of polyps had degen­erated and disappeared — impossibility under the “force that tends inces­santly to complicate organization.” Thus, worms without nerve cords must represent part of a second and separate sequence of progress. Lamarck pro­posed an origin for this second sequence in the spontaneous generation of even simpler worms as parasites within the bodies of other organisms. If

3-1. Lamarck's linear series of animal organization, from volume 1 of the Philosophie zoologique of 1809. (Author's collection.)


different environments — a pond and the body of a complex creature — encour­age disparate inceptions for sequences of progress, then the two forces com­mingle (Fig. 3-2).

(2) Ramification at the top. Lamarck could not rank the vertebrates in lin­ear order. He followed the conventional path of fish to reptile, but could not convince himself that birds fell between reptiles and mammals in a genealogi­cal sense. He therefore permitted a fork, provoked by the environmental set of forces, at the very top of a ladder supposedly built by the unilinear impetus of progress (Fig. 3-3): “We cannot doubt,” he wrote with characteristic cer­tainty (1809, p. 176), “that the reptiles by means of two distinct branches,

3-2. Lamarck's later conception of two chains of being with different starting points, the first (to the left) from free-living single-celled infusorians, the second (to the right) beginning with parasitic worms spontaneously generated within the bodies of higher organisms. From Lamarck, 1815. (Author's collection.)


3-3. Lamarck allowed his linear sequence to branch at the apex of complexity (shown as the bottom in this figure because he begins with the lowest forms and works down by descent). Lamarck could not rank birds and mammals as part of the single sequence, and therefore allowed a branch after reptiles with birds on one side (left, culminating in egg-laying monotremes), and mammals to the right. From additional material added to the end of Volume 2 of the Philosophie zoologique of 1809. (Author's collection.)


caused by the environment, have given rise, on the one hand, to the formation of birds, and, on the other hand, to the . . . mammals.”

I doubt that we can take this analysis much further. Historians often err in trying to wrest consistency from great thinkers at all costs. Some issues are too difficult, too encompassing, too important, too socially embedded, or just too devoid of evidence, for resolution even by the finest scientists. Darwin never consolidated his contradictory ideas about progess (see Chapter 6), and Lamarck never found a thoroughly consistent way to fulfill his desired argu­ment for a full separation between two forces pulling evolution in orthogonal  {186}  directions — up the ladder of progress, and sideways into lateral paths of adaptation. Lamarck may never have completed his scheme with success and consistency, but he made his desires clear to the point of redundancy. Lamarck's two-factor theory holds the distinction of being both the first evo­lutionary system in modern Western thought, and a strong argument for causal hierarchy. The two levels — in strong contrast with modern theories of hierarchy — are both causally distinct and contradictory for Lamarck, thus in­spiring Darwin's legitimate disparagement. Lamarck's distinction of levels, as discussed in the next section, unites hierarchy and evolution at the starting gate of the subject's modern history.


Lamarck had separated his sets of forces in order to account for the two pri­mary attributes of natural order — features that seemed to play off against, or even to contradict, each other. First, organisms form a progressive sequence from mo»ad to man, but the sequence abounds with gaps and deviations — so some other force must be disrupting a potentially smooth gradation. Second, organisms are well adapted to their environments, but most adaptations, from the tiny eyes of moles to the legendary necks of giraffes, represent particular specializations and departures from type (with many adaptations counting as losses or degenerations); therefore, adaptation cannot account for the sequence of progress.

Lamarck joined the two sets in a discordant union that operated more like a tug of war than a harmony. This partnership made no pretense to equality. A primary and dominating force — the march of progress — struggled to order organisms in a simple and sensible way; while a secondary and disrupting force — l'influence des circonstances, or adaptation to local environments — tore this order apart by pushing individual lineages into lateral deviations from the main track, thereby making the order of life rich, messy, and replete with clumps and gaps. This clear distinction of merit — the regular vs. the de­viant, the progressive vs. the merely fit — imparts the character of hierarchy to Lamarck's uneasy marriage of forces, with a primary factor doing its inexora­ble, underlying work at a higher level, while a secondary but more immediate factor of disruption plays upon the products of this higher level, pushing some forms into the side-channels of its influence. Burkhardt (1977, p. 87) captures both the hierarchy and conflict of forces in his epitome of Lamarck's system as an attempt to explain “how organisms would develop naturally” along a chain of progress “were it not for the constraining accidents of his­tory” pushing lineages into side channels of adaptation.

Lamarck worked his way slowly towards this final system of hierarchy and relative importance. The Floreal lecture of 1800 states that the “princi­pal masses” of major taxonomic units “are almost regularly spaced” (1800, 1984 edition, p. 416), but designates some peculiarly adapted species as “lat­eral ramifications” and “truly isolated points.” But this lecture cites only environment  {187}  as a trigger of change. The Recherches of 1802 adds the theme of “organic movements” forming new organs and faculties in an intrinsic se­quence of advance. By the time of his most famous work, the Philosophie zoologique of 1809, Lamarck “was explicit in portraying the diversity of ani­mal form as a result of two separate processes” (Burkhardt, 1977, p. 145), and he had formulated the arguments of hierarchy and relative importance as well. The Histoire naturelle of 1815-1822 then consolidates and advocates the hierarchical two-factor theory even more strongly.

In the Philosophie zoologique, Lamarck begins by claiming that, in an ide­ally simple world, a single sequence of progress would regulate all taxonomic order:

It may then be truly said that in each kingdom of living bodies the groups are arranged in a single graduated series, in conformity with the increas­ing complexity of organization and the affinities of the object. This series in the animal and vegetable kingdoms should contain the simplest and least organized of living beings at its anterior extremity, and ends with those whose organization and faculties are most perfect. Such appears to be the true order of nature, and such indeed is the order clearly disclosed to us by the most careful observation and an extended study of all her modes of procedure (1809, p. 59).

But this principle of progress remains insufficient in our actual world, where environmental change elicits adaptations off the main sequence: “It does not show us why the increasing complexity of the organization of ani­mals from the most imperfect to the most perfect exhibits only an irregular gradation, in the course of which there occur numerous anomalies or devia­tions with a variety in which no order is apparent” (1809, p. 107).

These “anomalies and deviations” are produced by a second, and clearly subsidiary, force — a “special factor” that thwarts the “incessantly working” source of general progress, and riddles the chain with gaps and lateral branches: “If the factor which is incessantly working toward complicating or­ganization were the only one which had any influence on the shape and or­gans of animals, the growing complexity of organization would everywhere be very regular. But it is not; nature is forced to submit her works to the influ­ence of their environment, and this environment everywhere produces varia­tions in them. This is the special factor which occasionally produces . . . the often curious deviations that may be observed in the progression” (1809, p. 69).

This special factor may be identified as environmental adaptation, initiated by changed habits and abetted by soft inheritance in the principles of use and disuse and the hereditary passage of acquired characters: “The environment exercises a great influence over the activities of animals, and as a result of this influence the increased and sustained use or disuse of any organ are causes of modification of the organization and shape of animals and give rise to the anomalies observed in the progress of the complexity of animal organiza­tion” (1809, p. 105).  {188} 

As a primary sign of our estrangement from Lamarck's world, and our lack of understanding for his system, all “standard” textbook examples of Lamarckian evolution ignore his fundamental, higher-level principle of prog­ress, and only cite instances of lateral twigs built as highly specialized adapta­tions. We do this, I suppose, because adaptation and specialization constitute the major theme in our modern evolutionary vocabulary (in the altered guise of Darwinian causation), while the bulk of Lamarck's system has passed be­yond our notice into cognitive dissonance. In any case, every classical exam­ple — from eyeless moles, to webbed feet of water birds, to long legs of shore birds, to the blacksmith's strong right arm — ranks as a lateral deviation, not a stage on the main sequence. As for the greatest cliché and exemplar of all, the ubiquitous giraffe of our text-books, happily munching leaves at the tops of acacia trees, Lamarck provides only one paragraph of speculation — with no elaboration, no measurements, no data at all. An example can become a knee-jerk standard for many reasons, with cogent, complete documentation not always prominent among them (see Gould, 1991b, on the evolution of horses and the size of Hyracotherium). Nor does simple repetition enhance the probability of truth! Lamarck wrote this and only this about giraffes (even repeating a common error about differential lengths of fore and hind limbs):

It is interesting to observe the result of habit in the peculiar shape and size of the giraffe (Camelo-pardalis)': this animal, the tallest of the mam­mals, is known to live in the interior of Africa in places where the soil is nearly always arid and barren, so that it is obliged to browse on the leaves of trees and to make constant efforts to reach them. From this habit long maintained in all its race, it has resulted that the animal's fore­legs have become longer than its hind-legs, and that its neck is length­ened to such a degree that the giraffe, without standing up on its hind-legs, attains a height of six meters (1809, p. 122).

The final, complex order of life arises from an interplay of the two forces in conflict, with progress driving lineages up the ladder and adaptation forcing them aside into channels set by peculiarities of local environments: “The state in which we find any animal, is, on the one hand, the result of the increasing complexity of organization tending to form a regular gradation; and, on the other hand, of the influence of a multitude of very various conditions ever tending to destroy the regularity in the gradation of the increasing complexity of organization” (1809, p. 107). In his strongest characterization of the two forces as conflicting, Lamarck tells us in another passage that “nature's work [of progress] has often been modified, thwarted and even reversed by the in­fluence exercised by very different and indeed conflicting conditions of life upon animals exposed to them throughout a long succession of generations” (1809, p. 81).

Two additional statements in the Philosophie zoologique give dramatic expression to the absolute distinction of the forces, and to their hierarchical character, with progress as primary and regular, and diversity as secondary and disturbing. The first provides a vivid iconography of the two-factor theory:  {189}  progress builds the regular and rising trunk; diversity snatches some items off this upward highway and pulls them into orthogonal blind alleys — “lateral ramifications” that peter out into “isolated points”: “These irregu­larities ... are found in those organs which are the most exposed to the influ­ence of the environment; this influence involves similar irregularities in the shape and condition of the external parts, and gives rise to so great and singu­lar a diversity of species that, instead of being arranged like the main groups in a single linear series as a regularly graduated scale, these species often con­stitute lateral ramifications around the groups to which they belong, and their extremities are in reality isolated points” (1809, p. 59). The second features Lamarck's explicit statement about hierarchy, translated into differential taxonomic levels of attention. The broad forces of progress set relations among orders and classes; smaller and more immediate episodes of adaptation estab­lish species and genera:

Nature . . . has really formed a true scale ... as regards the increasing complexity of organization; but the gradations in this scale . . . are only perceptible in the main groups of the general series, and not in the spe­cies or even in the genera. This fact arises from the extreme diversity of conditions in which the various races of animals and plants exist; for these conditions have no relation to the increasing complexity of organi­zation; but they produce anomalies or deviations in the external shape and characters which could not have been brought about solely by the growing complexity of organization (1809, p. 58).

Lamarck's increasing conviction about the distinctness, hierarchical character, and conflicting nature of the two forces culminates in his last major work, the Histoire naturelle of 1815-1822. The force of progress has now be­come a “predominant prime cause,” while adaptation ranks below as occa­sional and foreign, a disturbance strong enough to disrupt but not to efface nature's deeper law:

The plan followed by nature in producing animals clearly comprises a predominant prime cause. This endows animal life with the power to make organization gradually more complex . . . Occasionally a foreign, accidental, and therefore variable cause has interfered with the execution of the plan, without, however, destroying it. This has created gaps in the  series, in the form either of terminal branches that depart from the series in several points and alter its simplicity, or of anomalies observable in specific apparatuses of various organisms (1815, in Corsi, 1988, p. 189).


We have seen how Lamarck formulated and intensified the two-factor theory from his first exposition of evolution in 1800 to his last and most comprehen­sive works. One might even say, following a developmental metaphor, that the two sets of forces differentiated from an originally more inchoate con­glomeration of evolutionary ideas, gradually becoming more different and  {190}  more sharply defined. As the two sets grew along their orthogonal pathways, they became alternate centers of nucleation for the full realm of evolutionary ideas. All major items of this new conceptual world fell to one side or the other — as though two suns had entered an originally homogeneous universe, and all particles had to enter either one or the other gravitational system. Lamarck's two-factor theory separated the universe of evolutionary concepts into a set of dichotomies best characterized as antinomies, an unfamiliar word designating contradictions between two equally binding principles (and originally used to specify differences between ecclesiastical and secular law when both vied for domination in medieval states). The precipitation of ideas about Lamarck's two axes established a long list of antinomies that, in an im­portant way, has set the agenda of evolutionary biology ever since. Darwin opposed this structure of antinomies; others have advanced strong defenses, in whole or in part. The modern theory of hierarchy depends upon a selective defense, but in a manner radically different from Lamarck's formulation. Consider a few key items:

Ideal order (real) vs. disruption (disturbing). The interpre­tation of diversification by adaptation as lateral to, and disruptive of, an un­derlying lawful regularity marks an old tradition that Darwin fought fiercely by elevating the supposed “disturbing” force to the cause, by extrapolation, of all evolution. Curiously, since old traditions die hard, this antinomy re­mains potent (even under a Darwinian rubric) in the common claim that anagenesis, or evolutionary trends in lineages, should be viewed as distinct from cladogenesis, or diversification — and that speciation is, in Julian Hux­ley's words (1942, p. 389), “a biological luxury, without bearing upon the major and continuing trends of the evolutionary process.”

Behind this issue, of course, and particularly well expressed in this first antinomy, stands the ancient credo of essentialism. Just as the essence or type never becomes fully incarnate in an actual object (because any material being must be subject to all the slings and arrows of outrageous fortune in our pal­pable world of accidents), so does the outside world of changing environ­ments deny full expression to the ideal march of progress.

Progress vs. diversity. Lamarck's expression of the fundamental vs. the disruptive; note Huxley's words above for a modern expression.

Internal vs. external, or intrinsic vs. interactive. The march of progress is intrinsically and internally generated as a consequence of the chemical properties of matter; this march represents the “essential” pro­cess of life, moving ever forward in the absence of any push or disturbance from external forces. Lamarck makes this contrast explicit in arguing that the march of progress would proceed to smooth completion even in an absolutely constant environment.

Immanence vs. unpredictability. The forces of progress, arising as consequences of chemical laws, generate a set of predictable products in­herent in the constitution of nature. Since the chain is constantly replenished by spontaneous generation, all stages exist at all times, and the entire se­quence constitutes a permanent part of nature. But the disturbing force of environmental  {191}  change introduces the oddness and unpredictability of contin­gency. We can never know exactly what climatic change will occur, which lineage will be diverted into its channel, and how the resulting adaptation will form. Thus, our actual world becomes filled up with unique particulars. Gi­raffe necks do not arise by first principles of natural law, but as a contingency of dry climates and acacia trees at a particular time and place.

Timeless vs. historical. History requires distinctive moments that tell a story as a sequence of events. The force of progress may confer history upon any particular bolus of protoplasm as it mounts the ladder. But, in a larger sense, this force also cancels the usual meaning of history. Each step becomes predictable and repeatable; and each exists at every time (since spon­taneous generation continually replenishes the base). Thus, Lamarck's per­fecting force becomes essentially ahistorical. The rungs of the ladder are per­manent and always occupied; items pass up and through, but the forms are timeless. Genuine history enters via the disturbing force of environmental ad­aptation instead. A mole without eyes, a stork with long legs, a duck with webbed feet — all originate as nonrepeatable objects of a historical moment, triggered by a particular change of environment in a unique time and place.

Higher taxa vs. species and genera. Lamarck's two-factor the­ory is hierarchical. The force of progress — paramount, primary, and underly­ing — produces patterns of nature at the broadest scale, and therefore forges the relationships among higher taxa in our classifications. The force of adap­tation is secondary, disruptive, and subsidiary. It seizes individual lineages and pulls them from the main Sequence into side channels that always peter out as dead ends. This lower-level force produces the smaller units, the spe­cies and genera of our taxonomies.

Elusive vs. palpable. This last antinomy does not form part of La­marck's scheme, but becomes important in later interpretations, particularly in Darwin's refutation. The force of progress lies deeper within and operates at a higher level; the force of adaptation works palpably at the surface of things. One can, at least in principle, observe climates getting colder and ele­phants growing thick coats of fur in direct response; but advance up the lad­der lies further from our view in an abstract distance. Lamarck might have denied that the causes of progress posed any greater difficulty in recognition and observation; in his conceptual world, these forces arose from the chemi­cal nature of matter — and therefore became just as accessible as the immedi­ate causes of adaptation. But when Lamarck's theories of physical causa­tion collapsed, the force of progress became elusive — something operating so slowly, and at such high taxonomic levels, as to be effectively invisible in the here and now of testable science. Darwin based his theory upon a reformu­lation of this seventh antinomy — by arguing that palpable and immediate forces of adaptation did not oppose an inscrutable and untestable force of progress — but rather became the source of progress as well (and hence the only primary cause of all evolution) when extrapolated, by principles of uniformitarianism, into the immensity of time. All evolution therefore en­tered the realm of the testable.  {192} 

Modern evolutionists may read this list with an odd feeling of deja vu — in the backward sense that we have already encountered all these issues in the modern debates of our own professional careers, but didn't know that our forebears had struggled over the same themes. Doesn't the late 20th cen­tury debate about micro and macroevolution raise the same questions about different causes at higher and lower taxonomic levels (the basis of Goldschmidt's argument, for example — see pp. 451466); and don't extrapolationists still charge the defenders of higher level causality with proposing untestable theories of evolutionary change?

I do not find this persistence surprising at all (see Gould, 1977a). I have already cited (p. 58) A. N. Whitehead's famous remark that all later philosophy is a footnote to Plato. He did not mean to argue, by this statement, that no one (including himself) had thought anything new for more than two thou­sand years. Rather, he wished to defend the proposition that truly deep issues are few and not o obscure. The first great thinker should be able to lay out a framework and specify the primary questions. Later history must recycle the same issues, while offering new explanations in abundance (especially in em­pirical realms where truly novel information becomes available). Lamarck, for all his stubbornness and for all the idiosyncrasies of his theory, was a great thinker, and he did find a location for all major questions within his system. His theory therefore becomes a starting point, and later debate must engage the same issues. Given the central theorem of this book, I am especially gratified that Lamarck based the initiating system of our profession upon a theory of hierarchy — in a form that did not work, based on causes that we must reject for Darwin's good reasons; but a theory of hierarchy nonetheless. Evolutionary theory therefore set its roots, and cut its teeth, in the concept of hierarchical levels of causality.

An Interlude on Darwin's Reaction

In the flood of Darwinian scholarship unleashed after the centennial celebra­tions of 1959 and continuing unabated today, I regard no reform as more im­portant than the thorough debunking of the romantic myth that Darwin, alone and at sea, separated from the constraints of his culture, apprehended evolution as an objective raw truth of nature. This Galapagos myth, rooted in tortoises and finches, is demonstrably false in Darwin's particular case, and surely bankrupt as a general statement about human psychology and the soci­ology of knowledge.

Darwin saw many wonderful things on the Beagle; nature challenged him, broadened his view, and instilled flexibility. Darwin returned to England with the tools of conversion, but still as a creationist, however suffused with doubts and questions (Sulloway, 1982a; Gruber and Barrett, 1974; Schweber, 1977; Kohn, 1980; Desmond and Moore, 1991; Browne, 1995). As for the Galapagos, he had missed the story of the finches entirely, because he had been fooled by their convergences and had not recognized the underlying taxonomic  {193}  unity (Sulloway, 1982b; Gould, 1985c). Darwin had been informed that the tortoises differed from island to island, but had failed to appreciate the significance of this claim. Naturalists then believed, falsely, that the Ga­lapagos tortoises were not indigenous but had been transported in recent memory by Spanish buccaneers and placed on the Galapagos as a source of food for revictualizing ships. Thus, Darwin must have reasoned, if the tor­toises had only been on the Galapagos for two or three centuries, differences among islands could not be consistent or meaningful. The Beagle had pro­visioned itself with several tortoises, stored live in the hold as meat on the hoof, so to speak. In a modern version of fiddling while Rome burned, Dar­win partook of the feasts but made no plea for conservation when his ship­mates then pitched the carapaces overboard.

Darwin became an evolutionist by returning to England and immersing himself in the scientific culture of London — by arguing with colleagues, by reading and pondering (mostly in the library of the Atheneum Club), by seek­ing good advice (learning from ornithologist John Gould, for example, that those diverse Galapagos birds were all finches). He exploited a broad range of humanistic Western culture in pursuing his struggle for intellectual reform in natural history. He read Plato, Milton, and Wordsworth. He constructed the theory of natural selection, as argued in the last chapter, in conscious analogy with the laissez-faire theories of Adam Smith and the Scottish economic school. Darwin, without the impetus and challenge of this intellectual envi­ronment, might have become a country parson, with a beetle collection main­tained by an ecclesiastical sinecure as the remnant of a childhood passion for natural history.

In this enlarged perspective on the origin of Darwin's evolutionary views, the importance of his precursors becomes greatly enhanced. Lamarck and Chambers* do not figure as irrelevancies to be ignored, or (even worse) as impediments  {194}  to be overcome; they represent an essential part of his context for study, and they played a major role in shaping the radical and different char­acter of Darwin's distinctive theory.

I believe that Lamarck had a far greater influence on Darwin than tradition has allowed (a point advanced by other historians of science as well — see Corsi, 1978; Mayr, 1972, p. 90). I base this claim on Darwin's own contact with Lamarck's works, his private reactions as recorded in letters, and the eventual content of his theory. I don't claim that Darwin devised natural selection  {195}  as a conscious point-by-point contrast or refutation of Lamarck, but I suspect that Darwin clearly recognized what he liked least in Lamarck and strove to formulate a theory of opposite import.

Darwin said little about Lamarck in his published works, with no explicit reference to Lamarck's evolutionary views in the first edition of the Origin, and grudging praise in the historical preface added to later editions. But we know that he studied Lamarck intensely, and didn't like what he read. Dar­win owned a copy of the 1830 printing of the Philosophie zoologique (see Hull, 1985, p. 802), and read the book while making heavy annotations at least twice. More important, perhaps, Lamarck had provided Darwin's intro­duction to evolution via Lyell's fair but critical exegesis in the Principles of Geology.

Lyell's characterization becomes particularly interesting because he empha­sizes, in his masterful prose, the very two points that Darwin would strive most mightily to correct. First, Lyell castigates Lamarck for making asser­tions without a shred of direct evidence. Note that Lyell directs his scorn not at the palpable forces of lateral adaptation, but at claims for the origin of new organs as increments of complexity wrought by the forces of progress:

We point out to the reader this important chasm in the chain of the evidence, because he might otherwise imagine that we had merely omitted the illustrations for the sake of brevity, but the plain truth is, that there were no examples to be found; and when Lamarck talks “of the effects of internal sentiment,” “the influence of subtle fluids,” and the “acts of organization,” as causes whereby animals and plants may acquire new organs, he gives us names for things, and with a disregard to the strict rules of induction, resorts to fictions, as ideal as the “plastic virtue,” and other phantoms of the middle ages (Lyell, 1832, p. 8).

Second, and more important for my argument, Lyell gives a crisp and accurate account of Lamarck's hierarchical view of evolutionary causality, emphasizing the contrast between the regular cause of progress, and the disrupting force of adaptation. The passage, worth quoting in extenso, probably repre­sents Darwin's first contact with this invalid style of hierarchical theory:

Nature is daily engaged in the formation of the elementary rudiments of animal and vegetable existence, which correspond to what the ancients termed spontaneous generations . . . These are gradually developed into the higher and more perfect classes by the slow, but unceasing agency of two influential principles: first, the tendency to progressive advancement in organization, accompanied by greater dignity in instinct, intelligence, etc.; secondly, the force of external circumstances, or of variations in the physical condition of the earth, or the mutual relations of plants and ani­mals ... Now, if the first of these principles, the tendency to progressive development, were left to exert itself with perfect freedom, it would give rise, says Lamarck, in the course of ages, to a graduated scale of being, where the most insensible transition might be traced from the simplest to the most compound structure, from the humblest to the most exalted degree  {196}  of intelligence. But in consequence of the perpetual interference of the external causes before mentioned, this regular order is greatly inter­fered with, and an approximation only to such a state of things is exhib­ited by the animate creation, the progress of some races being retarded by unfavorable, and that of others accelerated by favorable, combina­tions of circumstances. Hence, all kinds of anomalies interrupt the conti­nuity of the plan, and chasms, into which whole genera or families might be inserted, are seen to separate the nearest existing portions of the series (Lyell, 1832, pp. 13-14).

Darwin's public silence (or mild approbation) is belied by his consistently negative attitude towards Lamarck, as recorded in private letters extending from the 1840's to post-Origin years. In 1844, he wrote to Hooker on the dearth of available evolutionary writing: “With respect to books on this sub­ject, I do not know of any systematical ones except Lamarck's, which is veri­table rubbish” (in F. Darwin, 1887, volume 2, p. 29). The most interesting post-Origin references occur in letters to Lyell, who criticized Darwin for not giving Lamarck sufficient credit. In responding to Lyell's first reaction to the Origin, Darwin wrote on October 11, 1859 (in F. Darwin, 1887, vol­ume 2, p. 215): “You often allude to Lamarck's work; I do not know what you think about it, but it appeared to me extremely poor; I got not a fact or idea from it.” Perhaps, but I suspect that Darwin got many a concept against it.

Darwin's longest statement, a testy comment directed against Lyell's re­peated designation of Lamarck as a source (though mitigated at the end by Darwin's usual humor), conveys special insight in Darwin's stated rationale for rejecting Lamarck's theory so firmly:

Lastly, you refer repeatedly to my view as a modification of Lamarck's doctrine of development and progression. If this is your deliberate opin­ion there is nothing to be said, but it does not seem so to me. Plato, Buffon, my grandfather before Lamarck, and others, propounded the obvious view that if species were not created separately they must have descended from other species, and I can see nothing else in common be­tween the “Origin” and Lamarck. I believe this way of putting the case is very injurious to its acceptance, as it implies necessary progression, and closely connects Wallace's and my views with what I consider, after two deliberate readings as a wretched book, and one from which (I well re­member my surprise) I gained nothing. But I know you rank it higher, which is curious, as it did not in the least shake your belief. But enough, and more than enough. Please remember you have brought it down on yourself!! (in F. Darwin, 1887, volume 2, pp. 198-199).

Note the basis of Darwin's critique — “very injurious ... as it implies neces­sary progression.” In other words, Darwin dismisses the higher-level cause of Lamarck's hierarchy. Darwin's own theory, of course, rested on the com­plete sufficiency, by extrapolation, of the lower-level force of adaptation in  {197}  Lamarck's system (as produced by the different mechanism of natural se­lection).

Lamarck's hierarchical theory set a context (in opposition) for Darwin's distinctive single-level theory of extrapolation, based on uniformitarian assumptions, from the palpable and small-scale cases of adaptation that sur­round us to all evolutionary changes at all scales of time and magnitude. Nat­ural selection does not emerge from the raw observation of nature, but as a complex idea embedded both in observation and in Darwin's voracious study and trenchant analysis of contemporary ideas in biology and general culture. Lamarck's hierarchical theory formed an important, though not widely rec­ognized, part of the mix, distilled by Darwin to extract a theory that would change the world.

Darwin directed his anti-hierarchical theory against Lamarck's old and in­valid concept of hierarchy — different and opposed causes at distinct levels. Darwin labored mightily to encompass the entire domain of evolutionary causation within a single level — natural selection working on organisms. He knew what he wanted to do, and he pursued and extended the logic of his ar­gument relentlessly. Most of his supporters (including Wallace and Huxley) never understood the subtle logic of the single-level theory. Among the few who did, Weismann also made a strenuous effort to bring the system to com­pletion. I find the strongest historical support for modern versions of hierar­chy (same causes working in different ways at various levels, in direct con­trast with Lamarck's notion of disparate causes in opposition), in the intense intellectual struggle carried out by the two greatest selectionists of the 19th century — Weismann for lower levels, and Darwin himself for upper levels — to bring the nonhierarchical theory of selection to completion and sufficiency. Both men, as we shall see in the subsequent sections of this chapter, struggled valiantly, but could not prevail. (Chapter 5 will then discuss (in the saltationist context of his version) the other major hierarchical system of evolu­tionary thought before the Modern Synthesis — the fascinating and subtle early 20th century theory of de Vries on reintroducing selection at the species level after denying its central importance at Darwin's own level of the organ­ism.) Call it the bad penny that keeps cropping up, or the pearl of great price always found within, but hierarchy seems unavoidable. Could the basic rea­son for this persistence find an explanation in something so lovely, and so beautifully simple, as truth-value?

No Allmacht without Hierarchy: Weismann
on Germinal Selection


In 1893, Herbert Spencer, who had a word (many of them) and a thought for nearly everything,* published a long critique in the Contemporary Review {198}  “The Inadequacy of 'Natural Selection.'” He strongly supported the Lamarckian principle of use and disuse with inheritance of acquired charac­ters and, while not denying the importance of Darwin's principle, railed against the exclusivity claimed for natural selection by August Weismann and his school, variously labelled as “strict,” or “ultra” or “neo” Darwinism. Weismann quickly rose to the challenge, choosing for his title a phrase that would become a motto for his approach. He called his rebuttal, in German, and parrying Spencer directly: Die Allmacht der Naturzuchtung — a title ren­dered by the English translator as “The All-Sufficiency of Natural Selection” (although I would prefer “Omnipotence,” or the literal “All-Might”).

This exchange (Spencer, 1893a and b; Weismann, 1893) became the focal point and most widely cited set of documents in the great debate between “neo-Darwinism” and “neo-Lamarckism,” perhaps the hottest subject in evolutionary theory at the end of the 19th century (see Kellogg, who wrote, 1907, p. 134 — “The best known part of the general debate was that carried on directly by Weismann and Spencer in the Contemporary Review.” These terms, as so often noted, bear little relation to the chief concerns of the name-bearers. Neo-Lamarckians bypassed Lamarck's central concept of materialis­tic progress and focused on a theory of heredity that Lamarck espoused as the folk wisdom of his day, not as anything distinctive in his system. Neo-Dar­winism referred to the panselectionism of Weismann and Wallace, an attitude explicitly and pointedly rejected by Darwin, who gave selection pride of place (hence the association), but granted other forces (including “Lamarckism”) important, if lesser, roles in evolutionary change).

Passions ran high; I own Weismann's annotated copies of Spencer's articles, and his anger drips off the pages (Fig. 3-4). The two warriors thrusted and parried on both high and low roads, mixing some good arguments about the structure of evolutionary explanation with ad hominem charges of incompe­tence. Weismann (1893, p. 317) disparaged Spencer for being merely a phi­losopher, and not a true scientist: “ [I] can only explain Mr. Spencer's ignoring such cogent instances by supposing that, as a philosopher, he is unacquainted with the facts by personal observation, and that therefore they appear less weighty to him than to a naturalist; for I would not for a moment suppose that he purposely evades the difficulties which face his opinion, as is the man­ner of popular orators and advocates — and alas! even of some scientists.” Spencer, in his touché (1893, p. 23), replied, not entirely without justice as we shall see, that Weismann had hidden poor arguments under the cloak of au­thority as a practicing scientist: “Now it is doubtless true that as a naturalist he may claim for his 'opinion' a relatively great weight. Still, in pursuance of the method of science, it seems to me that something more than an opinion is required as the basis for a far-reaching theory.”  {199} 

This chapter explores the hints, inklings, and tentative formulations of hierarchical selection theory developed and published during Darwin's century. As a primary goal, I wish to bring to light the buried and forgotten discovery made by all strict Darwinians: that they could not carry through the logic of organismal selection to render all evolution without a crucial assist from se­lection at other levels. Hierarchy theory only became a major and explicit theme in evolutionary thought during the late 20th century, but a secret of history reveals that none of the great thinkers who struggled, with uncompro­mising respect for logic, to establish a general theory based on organismal se­lection alone could ever make the argument work without an appeal, some­times in frustration, to hierarchy.

Alfred Russel Wallace and August Weismann stand out as the two principal “neo-Darwinians” of the late 19th century, the men most strongly dedicated to the Allmacht of selection. They therefore become the test cases for my as­sertion that hierarchy cannot be avoided. I shall bypass Wallace, though he fits my claim that no pure selectionist could avoid hierarchy, because I find no

3-4. Weismann's personal copy (see his signature in upper right hand corner) of Spencer's reply to their first round of polemics. The two marginal comments on page 12 read (in translation of Weismann's German): “Impermissibly weak!” And (ironically): “As if that were certain!” (Author's collection.)


evidence that he ever clearly conceptualized the issue of levels in selection. Wallace felt entirely comfortable with selection on all levels (see Kottler, 1985) and never seemed to grasp either the logic of Darwin's central commit­ment to the organismal level, or the problems involved in claiming that selec­tion on other units (particularly higher “individuals”) could be effective in the face of strong selection at the organismal level. Wallace maintained such an unshakable and primary commitment to the ubiquity of good design that he unhesitatingly invoked higher levels to preserve an argument for active se­lection whenever a focus on organisms raised the specter of nonadaptation (notably in his uncritical advocacy of species selection for sterility in interspe­cific crosses, rather than accepting Darwin's argument for infertility as a side consequence of accumulated differences in two diverging populations — see pp. 131132).

But August Weismann represents the ideal test case for my assertion. Once he had declared war on “Lamarckian” inheritance, Weismann dedicated his professional life to promoting the Allmacht of selection. He grasped the logic of Darwin's argument in all its details and extensions. He recognized the centrality of selection on organisms, and he struggled to make Darwin's single-level theory work for all phenomena of evolution. His famous 1893 paper on the Allmacht of selection presents, as its central theme, an explicit defense for exclusivity of the organismal level — or “personal selection” in his terms.* Later, and largely in response to strong arguments made by Spencer, Weis­mann admitted that he could not rely on personal selection alone. He could continue to promote Allmacht only by recognizing another level of “germinal selection” for subcellular components of the germplasm.

Moreover, Weismann gradually extended the theory of germinal selection, from an ad hoc aid for personal selection (in the original formulation of 1895 and 1896) to a fully articulated theory of hierarchy replete with notions of independence and conflict between levels (1904 version). Finally, Weismann came to regard hierarchical selection as the linchpin and completion of his en­tire theory (see pp. 221224) — though we have forgotten his cogent argu­ments, and usually depict him as the champion of conventional, organismic selection. Weismann's intellectual journey, his relentless probings and frequent reformulations, leading finally (and perhaps inexorably) to a full theory of hierarchy, provide an object lesson in the logic of evolutionary argument,  {201}  and the needs imposed by completeness and coherence once we abandon the myopia of regarding “organized adaptive complexity” (Dawkins, 1986) as the only focus for evolutionary explanation (with all else arising by extrapolation therefrom).


I first learned about August Weismann in high school biology as the man who “disproved” Lamarckism by cutting off mouse tails for numerous genera­tions and noting the fully retained tails of all offspring (a good example of terrible teaching based upon the myth of crucial experiments as the source of all insight in science). Weismann did perform these experiments (1888, in 1891, pp. 431-461), but they (by his own admission) did little to combat Lamarckism, which is, as supporters parried, a theory about the inheritance of functional adaptations, not of sudden and accidental mutilations.

Weismann's strong anti-Lamarckian argument does not rest upon an experiment, or an empirical observation at all. The rejection of soft inheritance arises as a logical deduction from Weismann's most distinctive contribution — his theory of inheritance and the continuity of germ-plasm (1885, in 1891, pp. 163-256). If germ-plasm is “immortal” (by passage across generations) and soma-plasm limited in existence by the death of each multicellular organ­ism; and if germ-plasm is sequestered early in ontogeny (“locked away” as the guardian of posterity, and protected from all somatic influence); then Lamarckian inheritance becomes structurally impossible because acquired somatic adaptations cannot affect the protected germ plasm. Weismann wrote in his Allmacht paper (1893, p. 608): “Nature has carefully enclosed the germ-plasm of all germ-cells in a capsule, and it is only yielded up for the formation of daughter-cells, under most complicated precautionary con­ditions.”

Once Lamarckian inheritance becomes impossible, Weismann's argument for the Allmacht of selection proceeds in four logical steps. This fourfold de­velopment will strike most modern scientists as curious and unsatisfactory, for the sequence not only requires no empirical contribution, but actively de­nies the possibility of effective input from this conventional source of scien­tific affirmation. The argument breaks no rules of logic, but several of its premises are (to say the least) not self-evidently true.

1. Adaptation is ubiquitous in nature; explaining adaptation therefore be­comes the chief goal of evolutionary theory. As “the greatest riddle that living Nature presents to us” (1909, p. 18), Weismann identified “the purposiveness of every living form relative to the conditions of its life, and its marvelously exact adaptation to these” (loc. cit.).

I believe it can be clearly proved that the wing of a butterfly is a tablet on which Nature has inscribed everything she has deemed advantageous to the preservation and welfare of her creatures, and nothing else (1896, p. 5).  {202} 

Everything we see in animals is adaptation, whether of today, or of yesterday . . . Every kind of cell... is adapted to absolutely definite and specific functions, and every organ which is composed of these different kinds of cells contains them in the proper proportions, and in a particu­lar arrangement which best serves the function of the organ . . . The or­ganism as a whole is adapted to the conditions of its life, and it is so at every stage of its evolution” (1909, pp. 64-65. This statement comes from Weismann's contribution to the “official” centennial celebration of Darwin's birth. Thus, Weismann chose to honor Darwin by stressing panselection.)

2. Adaptation must be attributed either to some materialistic cause, or to teleology (in the classic sense of spiritually directed purpose). The validity of science depends upon our ability to supply explanations in the former mode.

3. Among materialistic proposals, only Lamarckism and natural selection can explain adaptation — for adaptation is ubiquitous and clearly too com­plex to ascribe to chance or to render as a side consequence of any process serving unrelated ends.

4. Since Lamarckism is logically impossible (under the doctrine of continu­ity of germ-plasm), selection must be correct. To assert the Allmacht of selec­tion, we need no evidence beyond the disproof of Lamarckism. In fact, given the complexities of nature, and our inability to reconstruct past conditions in detail, we probably could not supply adequate direct evidence in any sin­gle case.

We accept it, not because we are able to demonstrate the process in de­tail, not even because we can with more or less ease imagine it, but sim­ply because we must, because it is the only possible explanation that we can conceive. For there are only two possible a priori explanations of adaptations for the naturalist — namely, the transmission of functional adaptations [i.e. Lamarckism] and natural selection; but as the first of these can be excluded, only the second remains . . . We are thus able to prove by exclusion the reality of natural selection, and once that is done, the general objections which are based on our inability to demonstrate selec­tion-value in individual cases, must collapse, as being of no weight... It does not matter whether I am able to do so or not, or whether I could do it well or ill; once it is established that natural selection is the only princi­ple which has to be considered, it necessarily follows that the facts can be correctly explained by natural selection (1893, pp. 336-337).

In 1893, when he made this bold assertion to counter Spencer's claim for the “inadequacy of natural selection,” Weismann advocated a kind of double exclusivity — for natural selection over Lamarckism, and for selection upon organisms as the only mode of Darwinian action. As a terminological matter, Weismann equated the general phrase “natural selection” with selection upon organisms alone (“personal selection” in his words. For example, he wrote (1903, vol. 2, p. 126): “It is upon this that the operation of natural selection,  {203}  that is, personal selection, must depend”). However, spurred by Spencer's cri­tique, he soon expanded the boundaries of selection to include other levels of nature's hierarchy.


As discussed in the last chapter, the primary and standard refutation of Dar­winism by late 19th century evolutionists held that natural selection could eliminate, but not create — and that some other factor must therefore be iden­tified to explain the origin of adaptations and species. For example, T. H. Morgan wrote in 1905, before he became a Darwinian: “It appears that new species are born; they are not made by Darwinian methods, and the theory of natural selection has nothing to do with the origin of species, but with the survival of already formed species” (in Kellogg, 1907, p. 95).

Darwinians, of course, understood this challenge, and responded with the argument that differential survivals, long cumulated, produce gradual and substantial changes meriting the designation “creative.” Weismann himself, for example (1896, p. 1), spoke of “the opposition of our own day, which contends that selection cannot create but only reject, and which fails to see that precisely through this rejection its creative efficiency is asserted.”

On this contentious question of creativity, several standard anti-Darwinian arguments invoked the earliest stages of features easily recognized as adaptive in their perfected form, for selection can preserve and accentuate a feature fully in place, but how can an organism move from an initial “there” to a fully functional “here”? Two claims predominated (see Mivart, 1871, for the classic statement that provoked Darwin's own response in later editions of the Origin): first, that initial steps are too small to provide any conceivable benefit in selection; and second, that earliest stages cannot initiate the final function in any sense (a bird cannot fly with 5 percent of a wing).

Darwinians developed satisfactory responses to both arguments about incipient stages of useful structures — the palpable value of tiny benefits for the first, and the principle of functional shift (preadaptation) for the second (see extensive discussion in Chapter 11, pp. 12181246). But the same problem seemed far more acute for the opposite dilemma of degeneration. Incipient stages of useful structures posed enough difficulties, although ultimate adaptiveness did suggest a Darwinian solution. But what conceivable pressure of natural selection could account for gradual stages in the disappearance of a functionless organ — for loss of function should remove a structure from the domain of selection entirely, and knowledge about an eventually adaptive state could not be invoked to guide an explanation for intermediary stages along such a functionless path.

(We might designate this problem by its classic example — the complete disappearance of eyes in some cave fishes. Despite a century of adequate Darwinian explanation, this issue continues to provide a rallying point for vernacular Lamarckism. I can testify to this in a personal way. As a result of  {204}  writing more than 300 popular monthly essays on evolutionary topics during the past 25 years, I have become a statistically adequate sampling point, through thousands of letters received from lay readers, for both the fre­quency and intensity of standard confusions about our profession. I can tes­tify that three items top the list of puzzlement: (1) evolution seen as ana­genesis rather than branching (“if humans evolved from apes, why are apes still around”); (2) panselectionism (“what is the adaptive significance of male nipples”); and (3) Lamarckism and the failure of natural selection (“doesn't the blindness of cave fishes imply a necessary space for Lamarckian evolution by disuse”).)

The problem of incipiency in degeneration poses more difficulty than the opposite issue of construction — for what can mediate the sequence if selec­tion does not regulate the final outcome? Weismann struggled to encompass this issue with his favored apparatus of Allmacht for selection — and he failed. Degeneration acted as the lever that pried Weismann from his panselec­tionism, and led him through a chronological series of honorable changes that must be read, in one sense, as retreats from a former pugnacious insis­tence on Allmacht, but that also represents a complexification and strength­ening of his original views.

Consider the example that Spencer raised with such effectiveness against Weismann, and that eventually prompted the theory of germinal selection — reduction of hind limbs in some whales to tiny vestiges with no exterior ex­pression at all. Two classical explanations had been invoked by panselectionists: (1) the limbs became so reduced by ordinary negative selection, as a consequence of the hindrances they imposed upon efficient, streamlined swimming; (2) the limbs are not, in themselves, harmful, but energy invested in any useless structure must handicap a creature relative to conspecifics with fewer vestiges and neutral organs.

Weismann invoked these standard arguments, but he became convinced (long before his debate with Spencer) that only part of the puzzle could be re­solved thereby. Selection would reduce the limbs to some degree (perhaps considerably), but surely the increments of further reduction soon become too small for granting a continuing, believable role to selection. Consider the figures that Spencer presents (1893b, p. 25), based on the efforts of a Dr. Struthers of Aberdeen, who had “kindly taken much trouble in furnishing the needful data, based upon direct weighing and measuring and estimation of specific gravity.” Spencer cites a Greenland Right Whale, weight 44,800 pounds, femur weight, 3-1/2 ounces; and a Razorback at 56,000 pounds, with a femur weight of 1 ounce — “so that these vanishing remnants of hind limbs weighed but 1/896,000th part of the animal.” Could one possibly be­lieve that a profound relative, but inconsequential absolute, reduction — from a two-ounce to a one-ounce femur, for example — might materially aid stream­lining (especially since external expression had disappeared long before) or conserve meaningful energy? Weismann accepted the implausibility of such a claim and recognized that he would have to seek an explanation beyond  {205}  organismal selection for such late stages in the reduction of degenerate or­gans. “To use Herbert Spencer's striking illustration, how could the balance between life and death, in the case of a colossus like the Greenland whale, be turned one way or another by the difference of a few inches in the length of the hind-leg, as compared with his fellows, in whom the reduction of the hind-limb may not have gone quite so far? . . . Further reduction to their modern state of great degeneration and absolute concealment within the flesh of the animal cannot be referred even to negative selection” (Weismann, 1903, vol. 2, p. 114).

This example, and the general phenomenon of degeneration, deeply trou­bled Weismann because common sense seemed to demand that his Lamarckian bugbear and bogeyman — so recently and, as he thought, finally and effec­tively buried — be disinterred to explain reduction as inheritance of features shriveled by disuse. Spencer himself raised this example in order to defend a Lamarckian explanation prima facie:

Thus, the only reasonable interpretation is the inheritance of acquired characters. If the effects of use and disuse, which are known causes of change in each individual, influence succeeding individuals . . . then this reduction of the whale's hind limbs to minute rudiments is accounted for. The cause has been unceasingly operative on all individuals of the species ever since the transformation began. In one case see all. If this cause has thus operated on the limbs of the whale, it has thus operated in all crea­tures on all parts having active functions (Spencer, 1893b, p. 26).

Weismann first attempted to resolve the difficulties posed by degeneration with his hypothesis of panmixia (not the later Fisherian definition now famil­iar to evolutionists). By panmixia, Weismann referred to the effect of recom­bination in sexual reproduction (amphimixis in his vocabulary) upon organs no longer subject to selection. When selection operates, Weismann argued, organs will be actively maintained, with constant vigilance and no relaxation, at the peak of their potential size and complexity by elimination of individu­als bearing substandard parts. But as soon as selection ceases to act, formerly “substandard” attributes will no longer be eliminated; they now mix freely with “good” parts, and the organ slides, by continuous dilution, down an in­clined plane towards total elimination. In a poignant example (since poor eyesight plagued his own career), Weismann wrote (1903, pp. 114-115). “If this conservative action of natural selection secures the maintenance of the parts and organs of a species at their maximum of perfection, it follows that these will fall below this maximum as soon as the selection ceases to operate ... Those with inferior organs of vision will, ceteris paribus, produce as good offspring as those with better eyes, and the consequence of this must be that there will be a general deterioration of eyes, because the bad ones can be transmitted as well as the good, and thus the selection of good eyes is made impossible.”

By his own admission and explicit defense (see p. 201), Weismann's argument  {206}  for Allmacht, and against Lamarckian inheritance, rested upon a log­ical structure of inferences from premises, not upon observation — for an em­pirical approach, Weismann held, could not achieve resolution, given the impossibility of “seeing,” at their minute sizes, the material bearers of hered­ity. Panmixia did compromise Allmacht in a sense, for this process yielded evolutionary change without selection. But following Kellogg's key distinc­tion of auxiliary from contradictory hypotheses (see pp. 163169), panmixia worked as an adjunct and aid — a mopping-up operation for organs fallen below the purview of selection, and, more importantly, a moat to prevent the incursion of a true enemy, the antiselectionist forces of Neo-Lamarckism. (Lamarck battled against Darwin for the common ground of universal adap­tation, while panmixia only worked to finish what selection had started, and only in the limited domain of degeneration.)

But Weismann's panmixia, having no support beyond the internal logic of the argument itself, could not survive the detection and exposure of crucial flaws. Spencer was not the first writer to illustrate the weaknesses of pan­mixia, but the debate of 1893 does mark Weismann's last attempt to explain degeneration by panmixia alone, and therefore contains the seeds for his next and final attempt — the theory of germinal selection.

Spencer, referring to “the vexed question of panmixia” (1893b, p. 22), of­fered three major rebuttals. “When from the abstract statement of it we pass to a concrete test, in the case of the whale, we find that it necessitates an un­proved and improbable assumption respecting plus and minus variation; that it ignores the unceasing tendency to reversion; and that it implies an effect out of all proportion to the cause” (1893b, pp. 28-29). The second point, based on Galton's principle of regression to the mean, denies that “minus” varia­tions can continue to accumulate differentially; the third brands panmixia as too weak a force to secure the total elimination of a useless organ. The first argument, however, proved to be not only decisive in itself, but unusual in sci­entific discourse by accusing Weismann (correctly) of conflating linguistic us­age with biological reality.

Weismann continually argued that selection maintained an organ “at its highest level.” Relaxation of selection might then impel an accumulation of previously eliminated variation in the minus direction only. But, as Spencer and others protested, why should selective optimization hold an organ at the summit of its potential size and complexity. Shouldn't optimality lie some­where in the middle of a possible range, with selective elimination of both plus and minus variations? “Take the case of the tongue,” Spencer argued (1893b, pages 23-24). “Certainly there are tongues inconveniently large, and probably tongues inconveniently small. What reason have we for assuming that the inconveniently small tongues occur more frequently than the incon­veniently large ones?” Without the invalid metaphor of selective summits, panmixia cannot reduce an organ to oblivion, for release from selection does not impart an inexorably downward trend to preserved variation.

All these objections can be combined into a single claim, which Weismann found so compelling that he eventually surrendered panmixia as a fully adequate  {207}  explanation of degeneration. Panmixia is a genuine, but weak, force; it can reduce the average value of an organ to a state somewhat below its for­mer functional size. But panmixia cannot solve the central question of degen­eration: what propels a useless organ all the way down the slide and into his­tory's dumpster? Weismann admitted his failure (1896, p. 22), and later summarized this ultimately unsuccessful episode in his quest to understand degeneration:

As my doubts regarding the Lamarckian principle grew greater and greater, I was obliged to seek for some other factor in modification, which should be sufficient to effect the degeneration of a disused part, and for a time I thought I found this in panmixia, that is, in the mingling of all together, well and less well equipped alike. This factor does cer­tainly operate, but the more I thought over it the clearer it became to me that there must be some other factor at work as well, for while panmixia might explain the deterioration of an organ, it could not explain its decrease in size, its gradual wearing away, and ultimate total disappear­ance. Yet this is the path followed, slowly indeed, but quite surely, by all organs, which have become useless (1903, vol. 2, p. 115).

Weismann therefore needed another kind of auxiliary hypothesis to pre­serve the Allmacht of selection against resurgent Lamarckism. He had tried the mechanics of inheritance as expressed in the doctrine of panmixia; now he would expand the domain of selection itself. He would depart from Dar­win's distinctive focus on struggle among organisms, and attempt to identify a source of directional variation in an analogous competition among determi­nants of heredity within germ cells — a “germinal selection.” Weismann de­vised a truly ingenious argument: if natural selection can produce trends in the morphology of phenotypes, then an intracellular, germinal selection might yield directionality in the variation presented to conventional selec­tion upon organisms. If the determinants of a useless organ predictably lose in an intracellular struggle for existence, then a trend to complete elimina­tion — an apparent example of Lamarckian inheritance by the principle of dis­use — might still be attributed to selection. This new mechanism could not be equated with Darwinian selection upon struggling organisms, but “germinal selection” did represent a process of the same form and logic, but applied to replicating objects at a subcellular scale rather than to entire organisms.

Weismann first proposed the theory of germinal selection as a brief note in his last rebuttal to Herbert Spencer, thus marking Britain's Victorian pun­dit as a chief source (in reaction) to the first explicit theory of hierarchical selection. (Neue Gedanken zur Vererbungsfrage, eine Antwort an Herbert Spencer, Jena, 1895). Weismann then elaborated the theory in 1896 (pre­sented to the International Congress of Zoology at Leiden on September 16, 1895; first published in The Monist in January, 1896, then as a separate pamphlet, translated into English later that year). Weismann's fullest develop­ment, with some remarkable changes by extension, appeared in his most im­portant book, Vortrage iiber Descendenztheorie (1902), translated into English  {208}  by J. Arthur and Margaret R. Thomson as The Evolution Theory in 1903. A comparison of the original 1896 version with the fullest exposi­tion of 1902 provides a fascinating exercise in itself, and also becomes a cru­cial argument for this book — for Weismann moved from a limited hypothe­sis proposed only as an adjunct to natural selection to a fully articulated theory of hierarchy, including concepts of independence and conflict between levels.


Germinal selection certainly finds its immediate source in Weismann's war with Lamarckism, his debate with Spencer, and his severe, longstanding dif­ficulty with the problem of degeneration. But Weismann's eventual embrace of hierarchy as an ultimate argument against Lamarckism also grew from a deeper foundation in German evolutionary thought. This lineage of argument is virtually unknown to English-speaking evolutionists, for the roots lie in the two most important untranslated documents of 19th century German evolu­tionary biology — the Generelle Morphologie (1866) of Ernst Haeckel and the Jugendwerk of a man who eventually made his considerable mark in an­other area of biology, Wilhelm Roux's (1881) Der Kampf der Theile im Organismus (The Battle of Parts in the Organism). Neither Haeckel nor Roux proposed a theory of causal hierarchy across levels of selection; both, in fact, spoke in the name of reductionism. Yet by denying, in very different ways, the exclusivity, or even the privileged status, of the organism as a causal agent in evolution, and by focusing attention on a structural hierarchy of lev­els, both Haeckel and Roux provided central ingredients to Weismann's the­ory of evolutionary hierarchy.

Haeckel's descriptive hierarchy in levels of organization

Generelle Morphologie der Organismen (1866), Haeckel's first book, repre­sents an eclectic mixture of militant reductionism and old-fashioned idealis­tic morphology; all united to an evolutionary theory every bit as idiosyn­cratic. (Haeckel dedicated volume two, jointly, to Darwin, Lamarck, and Goethe — and its central argument represents an odd fusion of their disparate ideas.) Haeckel's later notoriety rested almost entirely on the second volume, with its celebrated evolutionary trees (so often reproduced in modern text­books), based largely on his “biogenetic law,” ontogeny recapitulates phylogeny (Gould, 1977b). The first volume, entitled “Allgemeine Anatomie” and dedicated to Carl Gegenbaur, has largely been forgotten. This first volume consists of two major parts, each attempting to establish a formal science for morphological study and each, following Haeckel's invariable practice, stud­ded with a baroque terminology of his own construction. (Haeckel, with a sure sense of what R. K. Merton (1965) would later call the eponymous strat­egy for renown, coined new terms shamelessly, recognizing (I suspect) that a few would probably hang on to bear his legacy (an r-selection approach to  {209}  the courting of fame). The vast majority quickly succumbed to the negative selection of incomprehensibility, but survivors include ontogeny, phylogeny, heterochrony, ecology, and Monera.)

The second science, “promorphology,” tried to establish a physical, or crystallographic basis for organic form. Haeckel created a forest of terms illustrated in two complex plates, but never established any useful connections with physical or chemical principles. (Haeckel promoted his much vaunted mechanistic reductionism more by verbal proclamation than by deed, but the influence of a well-articulated philosophy consonant with social trends of an age must never be doubted.) The first science, “tectology,” tried a different approach to reductionism — not subsumption under physical laws, but break­down to component parts.

As the “basic principle” of tectology, Haeckel stated that all organic ob­jects must be built from components in a structural hierarchy of six ascending levels. But, in applying this pronouncement to actual cases, Haeckel makes a fascinating intellectual move, proving that his allegiance lay as much with ho­listic traditions of an older idealistic morphology, as with the militant physi­cal reductionism that won his lip service and fit with many of his social and political goals (Gasman, 1971). For Haeckel did not argue, in the manner of most 19th century reductionists, that his first and lowest level stands as fun­damental and basic (also “closer” to physics), with subsequent levels only treated as amalgamations based on principles of joining. Instead, Haeckel proclaimed a form of equality among the six levels (while not denying the compositional theme that lower units join to build higher entities). He re­ferred to tectology as the “doctrine of organic individuality” (Lehre von der organischen Individuality), and insisted that the objects at each of the six levels be designated as “individuals” in their own right — “individuals of the first order,” “individuals of the second order,” etc. He placed “plastids” (cells and cell components) on the first rung, organs (including tissues and organ systems) on the second, antimeres or Gegenstiicke (symmetrical parts, includ­ing rays or body halves of bilateral creatures, literally “counterparts”) on the third, metameres or Folgestucke (body segments, literally “following pieces”) on the fourth, persons (or vernacular “individuals”) on the fifth as “morphologische Individuen funfter Ordnung,” and colonies or “corms” on the sixth and last plane.

This equalization of status prompted the interesting consequence, with reference to natural selection, of denying to organisms their privileged Darwin­ian role as exclusive evolutionary agents. Natural selection surely ascribed evolutionary change to a struggle among individuals for reproductive success. But Haeckel insisted that objects at all six levels counted as “individuals,” and that no level could claim any special status as evolutionary agent. Organ­isms represent only one waystation in the ascending hierarchy. Perched on the fifth rung, they are made of metameres and aggregate into corms — just as or­gans are made of plastids and aggregate into antimeres. In an insightful state­ment on the role of language in prejudicing thought, Haeckel wrote of his fifth level (1866, vol. 1, pp. 318-319):  {210} 

An unbiassed and more deeply probing conception of organic individu­ality shows that these “true” or absolute individuals are, in fact, only rel­ative . . . Although these “true” individuals are, in most higher plants and coelenterates, only the subordinate components of a higher-standing unity (the colony), nonetheless the individuality of humans and higher animals leads us to the erroneous conception that morphological indi­viduals of the fifth order are the “true” organic individuals. This concept has become so general, and has been so strongly fixed in both scientific and vernacular consciousness, that we must mark it as the major source of the numerous and varied interpretations and debates that prevail on the subject of organic individuality.

Haeckel's concept of structural levels and the non-distinctive status of or­ganisms entered Weismann's argument in two crucial places — first, very gen­erally, when Weismann used the same style of thinking to establish a hierar­chy of (hypothetical) entities as the physical bearers of heredity within germ cells (see p. 214); and, second, quite specifically, when Weismann invoked Haeckel's six-part hierarchy (1896, p. 42) to argue that the struggle for exis­tence starts within germ cells, but then extends up through all Haeckelian cat­egories to colonies at the top.

Roux's theory of intracorporeal struggle

Wilhelm Roux's Der Kampf der Theile im Organismus evoked a wide range of reactions. Roux's teacher, Gustav Schwalbe, warned him against ever pub­lishing such a “philosophical” book again. Haeckel, another teacher, liked the work for its consonance with his own ideas, while Darwin himself, during the last year of his life, became greatly intrigued, writing to G. J. Romanes on April 16, 1881:

Dr. Roux has sent me a book just published by him ... It is full of rea­soning, and this in German is very difficult for me, so that I have only skimmed through each page; here and there reading with a little more care. As far as I can imperfectly judge, it is the most important book on Evolution which has appeared for some time ... Roux argues that there is a struggle going on within every organism between the organic mole­cules, the cells and the organs ... If you read it, and are struck with it (but I may be wholly mistaken about its value), you would do a public service by analyzing and criticizing it in 'Nature' (in F. Darwin, 1887, vol. 3, p. 244).

(Note how, contrary to the prevalent historical myth of the aged Darwin as the reclusive “sage of Down,” he actually (and actively) kept his ears alert, and his fingers right on the pulse of evolutionary debate. Romanes repre­sented just one among several younger colleagues and supporters whom Dar­win often recruited, both overtly and nonsubtly, to carry forth his interests in both the public and the professional arena.)  {211} 

Obviously, Roux had adopted Darwinian language for his title. Just as obviously, he hoped to apply the Darwinian apparatus at a level below its conventional locus of organisms. Roux's book surely occupies a place in the history of hierarchy theory, if only because its verbal image of struggling parts led many evolutionists, notably Weismann himself, to consider multiple levels of selection. But, curiously, as several critics soon noted (Plate, 1905; Kellogg, 1907; and even, with some ambiguity, Weismann, 1904, as well), Roux's the­ory does not really treat descent at all. Weismann's germinal selection, as we shall see, is a true theory of suborganismal selection and inheritance; but Roux's battle of the parts includes no statement about heredity, and ranks in­stead as a theory of functional adjustment in development.

Roux argued that the construction of a harmonious and well-designed organism emerges from a struggle among parts competing for limited nutri­ment. Lung cells compete with liver cells, and bone cells battle with other bone cells for best locations in the flow of nutriment. To cite Roux's favorite example, made even more famous by D'Arcy Thompson's later analysis (1917, and see Chapter 11, pp. 11951196), the bony trabeculae in the head of a hu­man femur form a virtual diagram of forces imposed on the bone during loco­motion, and must therefore be optimally designed to counter stress. But no one can argue that details of the arrangement in any single bone represent an evolutionary adaptation, if only because the trabeculae of a broken, and im­properly mended, femur reform along the new stress lines of a limping walk.

Roux argued that stresses establish lines of preferred flow for nutriment. Bone cells that happen to lie in the stream prosper and proliferate; others in less advantageous positions wither and die — leading to a functional honey­comb of struts and empty spaces. Roux used this argument to explain the functional design of tissues and organs in general, but he focused upon such complex and exquisite examples of optimal form as the barbules on bird feathers, the hairs that cover the spiracles of many insects, the arrangement of muscle fibers in the walls of blood vessels, and the bony trabeculae discussed above.

This “battle of the parts” may account for the flexible construction of opti­mal form in each organism. Indeed, such a principle, appropriately modern­ized, remains essential for a developmental biology that cannot invoke a specially tailored gene for each villus on an intestinal surface. But Roux's pro­posal cannot operate as a theory of evolutionary change for two reasons. First, the struggling parts do not vary in heritable ways, and victory cannot lead to beneficial changes in future generations. Bone cells that prosper on the growing trabeculae cannot be designated as superior to, or even in any sense intrinsically different from, the losing cells that die for lack of nutriment in the spaces between. The winners owe their success only to the good fortune of a favorable location. Kellogg (1907, p. 207) wrote: “This competition de­pends chiefly on the hazard of position ... Not the best qualified but the best situated fibers have vanquished the others by robbing them of food and thus finally destroying them.” Second, Roux's Kampf der Theile includes no theory  {212}  about inheritance. No matter how exquisite or optimal the outcome for any one organism, the results of the struggle cannot be imparted to offspring. (The capacity for functional adaptation might, of course, be heritable and might evolve by ordinary natural selection, but Roux never discusses this quite different issue.)

Weismann reacted to Roux's theory in a complex and ambiguous manner. He always credited Roux as an antecedent of germinal selection (a reasonable attribution, if only because an explicit metaphor of struggling parts can direct another scientist's thinking to a truly selectionist theory, even if the original proposal operated in a different domain). Weismann, particularly in his early work, seems to credit Roux — incorrectly — as a true suborganismal selection­ist: “Functional adaptation is itself nothing else than the efflux of intrabiontic selective processes” (1896, p. 15). Roux's theory, he argues in several passages, rests upon a variational base, and is therefore Darwinian.

But, by 1904, Weismann had recognized that Roux's suborganismal strug­gle could not operate as a theory of evolutionary change: “There is an essen­tial difference between personal and histonal selection, inasmuch as the latter can give rise to adaptive structural modifications corresponding to the needs of the tissue at the moment, but not to permanent and cumulative changes in the individual elements of the tissue” (1904, volume 1, p. 248). “No one will be likely to suppose that the distorted position of the spongiosa of a badly healed fracture could reappear in the straight bone of a descendant” (ibid., p. 251).

Moreover, Weismann added, even the metaphorical linkages of Roux to Darwin cannot be logically sustained. Most of Roux's examples do not in­clude competition among members of the same cell population (as in bone cells within the developing femur), but between entirely different organs: liver cell with lung, or kidney or heart. This process cannot be viewed as a struggle for existence at all, but only as a sorting out of different “species” into their appropriate places: “The struggle for existence and for descendants, in this case, is between two kinds of cell which were different from the beginning, and of which one has the advantage at one spot, another at another” (1904, volume 1, p. 248). Weismann then drew a striking analogy* between different  {213}  tissues in an organism, and different species of birds in a broad geograph­ical region:

The case may be compared to that of a flock of nearly allied species of bird, of which one species thrives best in the plains, another among the hills, and a third among the mountain forests, all mingled together in a vast new territory to which they had migrated, and in which all three kinds of conditions were represented. A struggle would arise among the different species, in which in every case the particular species would be victorious which was best adapted to the local conditions ... This would be the result of a struggle between the three species, not between individ­uals within each species, and it could not therefore bring about an im­provement of a single species, but only the local prevalence of one or an­other (1904, volume 1, pp. 248-249).

Weismann's strong and valid critique of Roux leaves us with a puzzle: why did Darwin, who understood the nature of selection so much better than any­one else (see next section), become so intrigued with Roux's book, if Kampf der Theile does not really develop a selectionist, or even a truly evolutionary, theory at all? Several resolutions may be suggested. Most mundanely, Darwin was no German scholar, and he may not, as he himself suggested to Romanes, have properly understood the theory in his cursory reading. Secondly, Darwin was not a strict selectionist, and may simply have valued Roux's insights on functional adaptation, including the Lamarckian implications for a theory of heredity by extension. But, in a third and intellectually more intriguing hy­pothesis, perhaps Darwin valued Kampf der Theile for two genuine benefits or consonances that Roux's book granted to natural selection — one practical, the other metaphorical.

In a practical sense, Roux explicitly provided the resolution of a paradox that had plagued natural selection — the problem of too much adaptation (“organs of extreme perfection” in Darwin's designation in Chapter 6 of the Origin). Can we really believe that organismal selection constructs each barbule on every feather — even with the immensity of geological time and the hecatomb of deaths in each generation? Roux offered Darwin a sensible exit from such an untenable implication: selection builds the capacity for an auto­matic functional response that can directly shape each organism in minutely adaptive ways during growth: “Through the capacity of the struggle of parts, a much higher perfection, the purposefulness of the functioning part down to the last molecule, can arise, and occur much more rapidly, than if it had to originate, by the Darwin-Wallace principle, through selection of variation in the struggle for existence among individuals” (Roux, 1881, p. 239).

But Roux offered even more, by way of metaphor, to Darwin's cardinal vision — the paradox of higher stability arising through struggle among lower elements. Functional adaptation might not rank as an evolutionary theory,  {214}  but such a process does produce internal order within a body by struggle — just as natural selection engenders the external harmonies of adaptive design and ecological balance: “As the struggle of parts [Kampf der Theile] yields purposefulness within an organism ... so does the analogous struggle for ex­istence [Kampf urn's Dasein] among individuals yield purposefulness with re­spect to external conditions of existence” (Roux, 1881, p. 238). Roux also echoed Darwin's most general and most important philosophical principle:

To many, the direction of this book may well seem very strange — for it holds that, in an animal, in which everything is so exquisitely ordered, in which all the different parts interlock with such excellence, and work to­gether in such perfected coordination, that a struggle of parts occurs, so that in one place, where everything works together according to firm principles, a conflict among the individual parts exists. But how can an entity [ein Games] exist, whose parts are at variance? ... How shall the good and the stable arise from struggle and battle?... All good can only arise from struggle [alles Gute nur aus dem Kampfe entspringt] (1881, p. 64).

Darwin himself could not have penned a better epitome for his most radical claim.


Weismann proposed the theory of germinal selection as a logical solution to the problem of degeneration in a non-Lamarckian world. But germinal se­lection only makes sense under Weismann's concept of inheritance — yet an­other theory of structural hierarchy, and explicitly linked by Weismann to Haeckel's 6-fold sequence as a further breakdown and elaboration (for germ plasm) of categories within Haeckel's lowest unit of “plastids,” or cellular constitutents (1896, p. 42).

In Weismann's admittedly hypothetical system, the fundamental sub-microscopic particles of heredity are called “biophors.” Biophors aggregate to “determinants,” the key unit for the theory of germinal selection. The logic of panselectionism requires a high degree of easy dissociability among genetic “particles” responsible for “traits” that can be individually optimized to con­struct well-adapted organisms — for if “particles” become too tightly linked or coordinated, then each change entails too many consequences for other traits, and constraints begin to prevail over adaptation. “Determinants” play this necessary role in Weismann's panselectionist theory of heredity. Each de­terminant builds an organ or a particular part of the body — in other words, an “item” of the phenotype that selection can mold independently.

Determinants, like their constituent biophors, are invisible and hypotheti­cal. They aggregate into the first observable unit, the “id” — an earlier use of a term that Freud coopted for a much different role and purpose (just as pale­ontologists had coined and developed a meaning for “mutation” (Waagen, 1869) before the new science of genetics outcompeted us with a later and altogether  {215}  different definition). Each id contains a determinant for every trait, and can therefore build a body. Weismann identified ids with the disk-like microsomes, recently observed as linearly ordered on chromosomes. The chromosomes themselves, or “idants” in Weismann's system, carry ids in rows, and stand atop the hierarchy of hereditary units.

Germinal selection rests upon the notion that determinants within germ cells may be analogized to organisms within habitats. Just as organisms strug­gle for limited resources (and not all can survive), determinants battle for the restricted flow of nutriment available to any cell. The winners grow and pro­liferate; the losers wither or disappear entirely. The strength of determinants governs the phenotypic expression of their resulting trait. Thus, if the deter­minants of a particular trait decrease or wither by germinal selection within cells, the trait will suffer in expression by exhaustion of its molecular base.

Weismann viewed germinal selection as an analog to interspecific struggle, rather than to competition among members of a single population — that is, determinants for one organ battle for limited nutriment with determinants for other parts of the body. But, unlike Roux's Kampfder Theile, Weismann's germinal selection does operate as a theory of altered heredity, and therefore as a potential evolutionary mechanism — for determinants weakened by ger­minal selection not only build a diminished body part; they also pass fewer (and debilitated) offspring determinants to germ cells of the next generation.

The ingenuity of this argument lies in its capacity to take the most serious potential challenges to the Allmacht of selection — a group of overt phenom­ena that seem to lie outside the possible control of organismal selection, and therefore within the domain of Lamarckism or orthogenesis — and to reinter­pret them as consequences of selection acting at a lower level. For if organis­mal selection can produce directional trends in phenotypes during geological time, then germinal selection can forge trends in strengthening or weaken­ing determinants (and their phenotypic expressions) across generations. The gradual and unidirectional shriveling to oblivion of an organ not subject to personal selection certainly suggests Lamarckian inheritance and evolution­ary loss by disuse; but if we descend a level and peer within the germ cells, we may envision (though we cannot directly see) a constant competition and se­lection among determinants, with losers paying the usual price of gradual and inexorable elimination. The Allmacht of strict Darwinism may be sacrificed, as organismal selection loses its exclusivity; but selection itself remains pre­eminent by expansion:

Powerful determinants in the germ cell will absorb nutriment more rap­idly than weaker determinants. The latter, accordingly, will grow more slowly and will produce weaker descendants than the former . . . Since every determinant battles stoutly with its neighbors for food, that is, takes to itself as much as it can, consonantly with its power of assimila­tion and proportionately to the nutriment supply, therefore the unimpoverished neighbors of this minus determinant will deprive it of its nu­triment more rapidly than was the case with its more robust ancestors (1896, p. 24).  {216} 

Weismann offered the same pugnacious defense for germinal selection that he had long championed for the Allmacht of organismal selection; the argu­ment must hold, lest we be driven either to mysticism or to the patently false Lamarckian mechanism:

No one who is unwilling to accept germinal selection can be compelled to do so, as he might be to accept the Pythagorean propositions. It is not built up from beneath upon axioms, but is an attempt at an explanation of a fact established by observation — the disappearance of disused parts. But when once the inheritance of functional modifications has been dem­onstrated to be a fallacy ... he who rejects germinal selection must renounce all attempt at explanation. It is the same as in the case of per­sonal selection. No one can demonstrate mathematically that any varia­tion possesses selection value, but whoever rejects personal selection gives up hope of explaining adaptations, for these cannot be referred to purely internal forces of development (1903, volume 2, p. 121).

Weismann, at least for public consumption, insisted that germinal selection represented the advancing wave of triumphant Neo-Darwinism. (In the late 19th century, “Neo-Darwinism,” a term coined by Romanes, referred to the panselectionist school of Wallace and Weismann, not to the pluralism of Dar­win himself. The modern meaning, associated with the evolutionary synthesis of the 1930's and onward, is not genealogically linked to this earlier defini­tion.) Many critics responded by charging that this invisible process repre­sented little more than an ad hoc hypothesis invented to save selection from the otherwise unexplainable phenomenon of degeneration. Kellogg (1907) provided, I believe, the most balanced perspective. He labelled germinal selec­tion as “a new and radically un-Darwinian theory” (1907, p. 134) — recog­nizing that Darwin's own theory of “natural selection” specified organisms as the locus of selection. But he respected the theory, recognized its similarity with the selectionist logic of classical Darwinism, and regarded germinal se­lection as a credible attempt to explain, in expanded Darwinian terms, the apparently un-Darwinian property of directional variation. He wrote: “Obvi­ously Weismann in his theory of germinal selection has preserved the ac­tuality of the struggle and the selection, but with a 'rehabilitation' of natural selection in the real Darwinian meaning and only fair application of the phrase, the new theory has nothing to do. It is, much more, a distinct admis­sion of the inadequacy of natural selection to do what has long been claimed for it. It is the first serious attempt at a causomechanical explanation of a theory of orthogenesis, that is, variation along determined lines” (Kellogg, 1907, p. 199).

I particularly value Kellogg's interpretation because the logic of his argu­ment correctly represents, in my view, the relationship of modern hierarchical selection theory to classical Darwinism and to the Modern Synthesis as well. Hierarchy should be viewed as an expansion of Darwinism in its continuing reliance on selection as the primary mechanism of evolutionary change. But,  {217}  at the same time, hierarchical selection does not merely extend Darwin's exclusively organismal version in a simple, comfortable, or inexorable manner. Rather, in fracturing Darwin's reliance on organisms as evolutionary agents, and in rendering evolutionary variation and change by interaction among lev­els, the theory of hierarchical selection introduces enough conceptual novelty, and disperses enough inadequate orthodoxy, to rank as a new, and in some respects radical, formulation, rather than a fully comfortable expansion.

Kellogg also grasped the weaknesses of germinal selection (which, of course, would soon become irrelevant when the codification of Mendelism disproved Weismann's conjecture about the physical mechanism of heredity). He asked (1907, pp. 200-201) (1) why measured variation so often con­formed to a normal distribution if germinal selection could act so powerfully to promote directional variation; (2) why species generally displayed such geological stability if germinal selection provided such a strong mechanism for orthogenesis; and (3) why, if determinants wage such constant battle, each against all others, severe deprivation of food often produced a proportion­ately dwarfed organism, rather than a creature lacking phenotypic expression for particular losing determinants in such intensified struggle.

Weismann originally developed germinal selection to explain the disappearance of degenerate organs, once he recognized that panmixia could only yield a partial reduction. But he soon expanded the scope of his new theory into the domain of positive selection as well, hoping to resolve thereby most of the remaining dilemmas in classical Darwinism. The main promise of germinal selection lay in its capacity to explain a phenomenon that could scarcely be more inconvenient for Darwinism — directed variation. Darwin had emphasized the “random” or undirected character of variational raw material as a prerequisite for advocating natural selection as the cause of evo­lutionary change (see previous discussion of this crucial principle on pp. 144146). For directed variation, if sufficiently powerful, would demote natural selection to a negative force for eliminating the unfit (while the fit arise auto­matically by differential variation), and a minor accelerator of trends origi­nating for internal reasons (since random mortality can sustain an evolution­ary direction imparted by biased variation).

But germinal selection could now explain directed variation by differential survival of struggling components within germ cells. Weismann therefore made a sweep through evolutionary problems that might be resolved by his new and selectionist theory of directed variation. In his original formulation of 1896, Weismann identified three possible roles for germinal selection in positive adaptation.

1. Fleeming Jenkin (1867) had troubled Darwin with his analogy of poten­tial variation within a species to a rigid glass sphere. Selection could be effec­tive along any radius, but the movement of a modal form from center to surface exhausted all possible variation, and positive selection must there­fore damp itself to oblivion before achieving any substantial change. But, Weismann argued, germinal selection of successful determinants establishes a  {218}  highway for new variation, and any positive natural selection can automati­cally engender, by germinal selection, a wave of increasing variation ahead of the advancing mean.

2. Spencer (1893a and b) had invested his major defense of Lamarck in the phenomenon of co-adaptation. How could natural selection, working sepa­rately on each trait, produce an intricate coordination of numerous parts, all changing in the same direction? But Weismann now argued that any positive organismal selection will strengthen the determinants of all traits involved, thereby triggering a coordinated trend in germinal selection. Co-adaptation becomes less puzzling if all positively selected traits achieve such a strong boost from germinal selection: “As soon as utility itself is supposed to exer­cise a determinative influence on the direction of variation, we get an insight into the entire process and into much else besides that has hitherto been re­garded as a stumbling block to the theory of selection ... as, for example, the like directed variation of a large number of already existing similar parts, seen in the origin of feathers from the scales of reptiles” (1896, p. 39).

3. In his furthest extension — a remarkable claim given his previous faith in the Allmacht of purely organismal selection — Weismann now argued that any intricately precise adaptation probably requires a boost from germi­nal selection to reach a pinnacle of exquisite design. Writing of mimicry in Lepidoptera, he argued:

It would have been impossible for such a minute similarity in the design, and particularly in the shades of the coloration, ever to have arisen, if the process of adaptation rested entirely on personal selection ... In such cases there can be no question of accident, but the variations presented to personal selection must themselves have been produced by the princi­ple of the survival of the fit! And this is effected, as I am inclined to be­lieve, through such profound processes of selection in the interior of the germ plasm as I have endeavored to sketch ... under the title of germinal selection (1896, p. 32).

This list, while granting broad scope and power to the newly formulated domain of suborganismal struggle, also illustrates the strictly limited concep­tual role that Weismann envisaged for germinal selection when he first formu­lated the concept in 1895 and 1896. The action of germinal selection must al­ways be synergistic with conventional organismal selection. In degeneration, the “type” example if you will, germinal selection “finishes off” what natural selection began but could not complete. In all other cases of evolutionary construction (maintenance of variation, co-adaptation, and “extreme perfec­tion”), germinal selection works hand in hand with organismal selection, ei­ther to supply positive variation, or to accelerate change by supplying an im­petus in the same direction from another level.

Over and over again, Weismann emphasizes the purely synergistic action of germinal with organismal selection. Thus, for example, ordinary natural se­lection can initiate the decline of an organ rendered useless by environmental change. In  {219}  fishes that have migrated into dark caves, organismal selection  against eyes must lead to differential survival of animals with weakened de­terminants for these organs. Once debilitated, these determinants continue to lose battles of germinal selection, and phenotypic expression sinks below the threshold that organismal selection could regulate: “In short, only by this means are the determinants of the useless organ brought upon the inclined plane, down which they are destined slowly but incessantly to slide towards their complete extinction” (1896, p. 25).

The case for synergism becomes even clearer when selection acts for the increasing complexity of an organ — for the two levels need not engage in a relay, but may now work continuously together in the same direction. Organ­ismal selection favors a stronger part, thereby preserving organisms with more powerful determinants for the part, and unleashing a necessarily synergistic germinal selection: “As soon as personal selection favors the more pow­erful variations of the determinant ... at once the tendency must arise for them to vary still more strongly in the plus direction . . . From the relative vigor or dynamical status of the particles of the germ plasm, thus, will issue spontaneously an ascending line of variation, precisely as the facts of evolu­tion require” (1896, p. 27).

Weismann explicitly identified this inherent and automatic synergism as a major insight, and a logical completion of his argument for the Allmacht of selection. Noting “the harmony of the direction of variation with the require­ments of the conditions of life” (1896, p. 38), Weismann continues: “The de­gree of adaptiveness which a part possesses itself evokes the direction of vari­ation of that part. This proposition seems to me to round off the whole theory of selection and to give to it that degree of inner perfection and com­pleteness which is necessary to protect it against the many doubts which have gathered around it on all sides like so many lowering thunderclouds . . . The principal and fundamental objection that selection is unable to create the variations with which it works, is removed by the apprehension that a germi­nal selection exists.” And, more succinctly for the centennial of Darwin's birth, Weismann reduced this theme to a celebratory aphorism (1909, p. 39): “Germinal selection supplies the stones out of which personal selection builds her temples and palaces: adaptations” — thus amalgamating the synergistic link of levels, the Allmacht of selection, and the primacy of adaptation (as re­corded in a metaphorical linkage with the noblest of buildings, both spiritual and temporal).


Weismann had granted an important role to germinal selection in his initial formulation, but he had not yet developed a full theory of hierarchy, for germinal selection could only walk the same paths established by ordinary natural selection — at most accelerating or intensifying the journey. But when Weismann wrote the major book and summation of his later career, Vortrage iiber Descendenztheorie (1902; English translation, 1903), he devoted two full chapters to germinal selection and, without explicitly acknowledging the  {220}  change or extension, radically altered his conception into a full selectionist theory of hierarchy, the first such proposal in the history of evolutionary thought.

Taking an opposite tack from his original formulation, Weismann now proposed to define the scope of germinal selection by what such a process could accomplish without and beyond organismal selection: “We shall at­tempt to gain clearness as to what it can do, and how far the sphere of its in­fluence extends, and, in particular, whether it can effect lasting transforma­tions of species without the cooperation of personal selection, and what kind of variations we may ascribe to it alone” (1903, vol. 2, p. 126).

Weismann now recognized that he could use germinal selection to escape the straitjacket of Panglossian adaptation. He could finally admit the non-adaptive character of some phenotypic features without straying from the Allmacht of selection, or giving comfort to Lamarckism — for traits arising by germinal selection may be neutral or even harmful to survival in the Dar­winian world of competition among organisms. Weismann recognized two classes of nonadaptive features potentially ascribable to germinal selection.

Neutral features of small or no importance. Since germinal selection can promote changes invisible to organismal selection, Weismann wondered whether such minor variations as human racial differences — attributed by Darwin to sexual selection based on disparate standards of beauty arising for capricious reasons in various societies — might actually arise as effects of germinal selection: “It cannot be denied that there are char­acters which have no special biological significance [although] ... it is dif­ficult and often impossible to point these out with certainty. The shape of the human nose and the human ear, the color of the hair and the iris, may be such indifferent characters whose peculiarities are to be referred solely to germinal selection” (1903, vol. 2, p. 134).

Orthogenetic drives that may yield harmful features and even lead to extinction. In a remarkable departure from the al­most strident panselectionism of his earlier years, Weismann now approved certain claims for orthogenesis, and admitted the existence of harmful trends (based on directionality in variation) that organismal selection could not re­verse. He even accepted the classic examples of the anti-Darwinian schools as orthogenetic and harmful prima facie — the antlers of Irish Elks, and the mas­sive canines of saber-toothed cats (1903, vol. 2, p. 139). He embraced the best cases of his opponents because germinal selection — once he reversed his original view and grasped its power to work against organismal selection — could convert these enemy troops to the doctrine of Allmacht. For when ger­minal selection acts with sufficient power, then all the determinants favored by organismal selection may be eliminated entirely, leaving only the vigorous determinants of harmful orthogenetic features, and rendering conventional selection impotent for lack of raw material: “In this case the variation-direc­tion which had gained the mastery in all ids could no longer be sufficiently held in check by personal selection, because the variations in the contrary di­rection would be much too slight to attain to selective value” (1903, vol. 2, p. 139).  {221} 

Following the early 20th century vogue for eugenics, Weismann used this argument to promote a positive program of breeding to save the human race. Germinal selection must be responsible for any decline of the human stock engendered by relaxation of natural selection — for panmixia lacks sufficient strength to produce such an effect, while Lamarckian inheritance does not ex­ist. This orthogenetic deterioration by germinal selection can only be reversed by a reimposition of Darwinian competition in the organismal mode, with re­productive success to the victors. Arguing that military service might operate as a good filter for identifying bodies well suited for success in organismal se­lection, Weismann suggested (1903, vol. 2, p. 147): “It would indeed be well if only those who had gone through a term of military service were allowed to beget children” (thus adding another example to the compendium of social nonsense advanced by prominent evolutionists in the name of Darwinism — see Chapter 7, and Fisher, 1930).

But, far more important than merely extending the domain of germinal selection to explain potentially nonadaptive organismal traits, Weismann also enlarged the conceptual realm of levels in selection from a narrow mechanism for synergism to a full theory of hierarchy. Weismann had now worked his way through the logic of multi-level selection theory, and had recognized the two key ingredients of any full account.

Independence of levels and potential for conflict. The attribution of orthogenesis to germinal selection implies that suborganismal selection can act separately from conventional Darwinian selection, and even work in a contrary direction to decrease phenotypic adaptation. Thus, the process that Weismann had originally promoted to make natural selection even more effective had become, by honest probing into all corners and impli­cations of the argument, a separate cause that could work either with or against the canonical Darwinian mode. In fact, Weismann now argued that potential independence from the Darwinian level of organismal selection es­tablishes the primary significance for germinal selection in evolutionary the­ory (1903, vol. 2, p. 119): “In this fact lies the great importance of this play of forces within the germ-plasm, that it gives rise to variations quite indepen­dently of the relations of the organism to the external world. In many cases, of course, personal selection intervenes, but even then it cannot directly ef­fect [sic, and correctly; he means 'cause,' not just 'affect'] the rising or falling of the individual determinants — these are processes quite outside of its influ­ence.”

In his most revealing change, Weismann even reinterprets his type case of degeneration. He had previously tried to explain degeneration as a result of germinal selection completing a process that natural selection had started but could not finish. Now, without any change in evolutionary mechanics, he speaks of germinal selection working differently and independently. He even claims that degeneration offers the purest case for potential independence of levels — for germinal and organismal selection usually act together, thus ren­dering their individual contributions operationally inseparable. But we know that organismal selection has disappeared in the final stages of degeneration, and we can therefore observe the unsullied action of germinal selection alone!  {222}  “In one direction variation can be proved to go on without limit, and that is downwards, as is proved by the fact of the disappearance of disused organs, for here we have a variation-direction, which has been followed to its utmost limit, and which is completely independent of personal selection; it proceeds quite uninterfered with by personal selection, and is left entirely to itself” [Weismann's italics] (1903, vol. 2, p. 129). At England's major centennial cel­ebration of Darwin's birth, Weismann presented an even stronger statement (1909, p. 38): “Useless organs are the only ones which are not helped to as­cend again by personal selection, and therefore in their case alone can we form any idea of how the primary constitutents behave, when they are subject solely to intragerminal forces.”

Such independence of levels implies potential conflict, with stability achieved through balance, or by the victory of one level. Natural selection, as a powerful force operating directly on phenotypes, usually prevails in its own domain of visible form. If germinal selection weakens a useful organ, natural selection can intervene in an antagonistic mode (except in degenera­tion, where natural selection ceases to act, and germinal selection reigns un­checked). “If a struggle for food and space actually takes place, then every passive weakening must lead to a permanent condition of weakness and a lasting and irretrievable diminution in the size and strength of the primary constituent concerned, unless personal selection intervenes, and choosing out the strongest among these weakened primary constituents, raises them again to their former level. But this never happens when the organ has become use­less” (1903, vol. 2, p. 122).

Similarly, if germinal selection initiates an orthogenetic trend, natural selection can impose a halt by eliminating organisms with traits exaggerated beyond utility, thereby removing their positive determinants from the germinal population. (This basic antagonism, Weismann finally concluded in his stron­gest recognition of potential conflict between levels, may be virtually omni­present in nature, and therefore fundamental to evolution, because positive organismal selection almost always elicits an upward trend in determinants by Weismann's earlier argument for synergism. Most stable species may not be quiescent with respect to selection, but balanced by a policing of germinal selection with opportune removals based upon Darwinian organismal com­petition.) “In the majority of cases the self-regulation which is afforded by personal selection will be enough to force back an organ which is in the act of increasing out of due proportion to within its proper limits. The bearers of such excessively increased determinants succumb in the struggle for existence, and the determinants are thus removed from the genealogical lineage of the species” (1903, vol. 2, p. 130).

But germinal selection can also triumph, and such victories may not be infrequent in nature. All orthogenetic and nonadaptive traits may record the potency of suborganismal processes in conflicts between levels of selection: “All excessive or defective hereditary malformations may be referred to ger­minal selection alone, that is, to the long-continued progressive or regressive variation of particular determinant-groups in a majority of ids” (1903, vol. 2, p. 138).  {223} 

Expansion of the hierarchy both up and down from a Darwinian focus on organisms. Weismann devised germinal se­lection as an ad hoc hypothesis to resolve his longstanding embarrassment over the problem of degeneration. In 1896, he applied selection throughout Haeckel's hierarchy of “individuals,” extending from cell constitutents to clonal colonies (1896, p. 42), and recognizing three primary levels — Darwin's conventional struggle for existence among organisms, Roux's histonal selec­tion, and his own germinal selection.

But by 1903, in a statement that I regard as wonderfully prophetic of cur­rent concerns, Weismann had proceeded beyond his immediate theoretical needs to full generality. He had used germinal selection to break through the Allmacht of exclusivity for Darwin's level. But now he recognized the inexo­rable logic of a fully developed and extended theory of hierarchy — reaching right up to species selection at the top.

I have called these processes which are ceaselessly going on within the germ-plasm, Germinal Selection, because they are analogous to those processes of selection which we already know in connection with the larger vital units, cells, cell-groups and persons. If the germ-plasm be a system of determinants, then the same laws of struggle for existence in regard to food and multiplication must hold sway among its parts which hold sway between all systems of vital units — among the biophors which form the protoplasm of the cell-body, among the cells of tissue, among the tissues of an organ, among the organs themselves, as well as among the individuals of a species and between species which compete with one another (1903, vol. 2, p. 119).

If Weismann had presented this full elaboration of hierarchy only as a foot­note, a flash of insight in a book devoted to other goals, I could not claim him as an intellectual forebear of our modern excitement. Good ideas originate in fair abundance, and we must look to development and application for our main criteria of sustained scientific worth. If Weismann had even devoted an isolated chapter to hierarchy, I would note his insight with praise, but grant him limited success for failing to recognize the power of this theme as an or­ganizing framework for evolutionary mechanisms. But, in fact, Weismann did fully grasp the fundamental difference between classical Darwinism and the expanded theory of interacting levels of selection — and he regarded his expo­sition of hierarchy as both the central feature of his mature thinking, and the unifying concept of all evolution: “This extension of the principle of selection to all grades of vital units is the characteristic feature of my theories; it is to this idea that these lectures lead, and it is this — in my own opinion — which gives this book its importance. This idea will endure even if everything else in the book should prove transient” (1903, in preface, vol. 1, p. ix).

How ironic that a man who so explicitly promoted the centrality of hier­archy should be remembered today primarily for his earlier advocacy of Allmacht at the traditional level of organismal struggle alone! A study of Weismann's intellectual ontogeny should lead us to respect the logic and power of hierarchy, whatever our  {224}  eventual judgment about merit and utility. Weismann, by far the most thoughtful of Darwinians in the first generation after the founder himself, could not “cash out” the exclusivity of organismal selection when he pushed the theory to the edges of its necessary application. He then invented an auxiliary theory of suborganismal selection to rescue himself from an uncomfortable corner; but he eventually followed the relent­less logic of his own argument to a full theory of hierarchy.

The theory of hierarchical selection does not constitute either a small and merely incremental nuance, or a modern concoction and exaggeration of bored Darwinians trying to stir up some trouble. Hierarchy has accompanied the theory of selection from its very inception — if only because no truly tena­cious and thoughtful Darwinian could ever avoid its appeal and logic, while at least one of the wisest and the most committed adherents, August Weis­mann, came to regard hierarchy as the implied and necessary centerpiece of any evolutionary theory fully rooted in selectionist principles, and truly com­prehensive in explanatory range. We have largely forgotten Weismann's intel­lectual journey today. But we should recover his chain of argument — for his motives and insights retain full validity, even if later discoveries about the physical basis of heredity invalidated his particular form of suborganismal selection.

Hints of Hierarchy in Supraorganismal Selection: Darwin on the Principle of Divergence


Charles Darwin cannot be judged as a consistently felicitous writer, but he could turn a phrase with the very best of craftsmen. As noted before, many of his lines, particularly his wonderful metaphors, have become parts of our cul­ture — the image of the “entangled bank” at the very end of the Origin, or the “tree of life” that closes Chapter 4 on natural selection. His posthumously published autobiography contains many memorable and oft-quoted state­ments, including his description of intellectual eureka: “I can remember the very spot in the road, whilst in my carriage, when to my joy the solution oc­curred to me” (in F. Darwin, 1887, vol. 1, p. 84).

Poll our biological colleagues, and most will tell you that this horse-drawn epiphany describes Darwin's Malthusian insight of September 1838, and the resulting formulation of natural selection. But the passage refers to a much later event, and this error of attribution may be the most common in all Dar­winian exegesis. The statement recounts an insight — the “principle of diver­gence” in his own description — that Darwin ranked as equal in importance with natural selection itself, an idea whose formulation sometime in the early to mid 1850's (the true date of the carriage ride) allowed him to complete his theoretical structure and begin writing his magnum opus.

Darwin describes the phenomenon that a principle of divergence must resolve, and states his surprise at his own obtuseness before the fateful car­riage ride:  {225} 

But at that time [after the Malthusian insight of 1838 and his composi­tion of the sketches of 1842 and 1844] I overlooked one problem of great importance; and it is astonishing to me, except on the principle of Columbus and his egg,* how I could have overlooked it and its solution. The problem is the tendency in organic beings descended from the same stock to diverge in character as they become modified. That they have di­verged greatly is obvious from the manner in which species of all kinds can be classed under genera, genera under families, families under sub­orders, and so forth (in F. Darwin, 1887, vol. 1, p. 84; the carriage state­ment directly follows).

Darwin (loc. cit.) then epitomizes the solution that he named “the principle of divergence” and ranked with natural selection as a foundation of his theory: “The solution, as I believe, is that the modified offspring of all dominant and increasing forms tend to become adapted to many and highly diversified places in the economy of nature.”

Darwin's principle of divergence has puzzled many biologists: why did Darwin rank the concept so highly, and as a principle separate from natu­ral selection? May we not view divergence as a logical consequence or sim­ple spin-off from natural selection itself? Yet when one considers the issue in Darwin's terms, both his separation of divergence from natural selection, and his joy in resolution, make excellent sense. Natural selection, as formu­lated under the Malthusian insight of 1838, states a principle of anagenetic change within phyletic lines — an argument about adaptation to local cir­cumstances (biotic and abiotic). This principle says nothing, by itself, about diversification, or splitting of one lineage into two or several descendant  {226}  taxa. So much of what Darwin needed to explain — plenitude in ecology, branching models in phylogeny, the hierarchical structure of taxonomy, to name just a few items of obvious centrality — rested upon the fact of diversi­fication, not adaptation (see Mayr, 1992, on Darwin's several theories of evo­lution).

One might say — indeed many of us do say, thus leading us to downgrade and misinterpret Darwin's explanation of diversity — that “divergence of character” requires no separate principle beyond adaptation, natural selec­tion, and historical contingency. After all, the earthly stage of evolution pro­vides ecological and biogeographical prerequisites for diversification. Cli­mates alter; topography changes; populations become isolated, and some, adapting to modified environments, form new species. What more do we need? Insofar as Darwin considered the issue at all between 1838 and the early 1850's, his thinking followed this general line (Sulloway, 1979; Ospovat, 1981). But Darwin grew dissatisfied with a theory that featured a general principle to explain adaptation, but then relied upon historical acci­dents of changing environments to resolve diversity. He decided that a fully adequate theory of evolution required an equally strong principle of diversity, one that acted intrinsically and predictably. If adaptation and diversification specify the central phenomena of evolution, each must have its principle, and their union would then define his complete theory.

(In modern evolutionary parlance, we may relate the growing intensity of Darwin's search for a general, “law-of-nature” explanation of divergence to his changing views about allopatric and sympatric speciation. During the 1840's, when diversity did not greatly trouble him as a theoretical issue, Dar­win tended to view speciation as allopatric, and therefore as a consequence of historical accidents in geography and ecology. When a population becomes spatially isolated, he reasoned, natural selection can act independently upon it, and eventually accumulate enough divergence from the ancestral form to establish a new species. But Darwin's preferences then shifted to sympatric views of speciation — and he therefore developed a conviction that some gen­eral law, and not just historical accidents of isolation, must promote the mul­tiplication of species. A complete theory of natural selection required that this elusive “law” of speciation or divergence also be based on the predictable op­eration of organismic selection. In the light of our current preferences for allopatric speciation, Darwin's shift may seem ironic, but our opinions and certainties, as presently defined, must be deemed irrelevant to such historical analysis.)

In the context of this book and its principal theme of hierarchical selection, I stress the centrality of Darwin's changing views on divergence because I think that I have made a small discovery about the structure of his argument. I shall try to show that this most brilliant of all theorists, this rigorously hon­est thinker who worked so diligently to explain all evolution as a conse­quence of organismic struggle, tried mightily to render his second touchstone, his “principle of divergence,” by ordinary natural selection — and he failed. He could not succeed because the logic of his argument demands a major role  {227}  for sorting and selection at the species level — and Darwin, with characteristic honesty, faced his distress head on.

I am fascinated that the exegetical literature on Darwin's attitude to supra-organismal selection (Ghiselin, 1974; Ruse, 1980, for example) has focused entirely on putative cases of group selection (sterility in plant hybrids; neuter castes in Hymenoptera) and has quite properly concluded that Darwin, with a possible exception for invoking family or clan selection to explain human moral traits, doggedly and consistently carried through his program for the exclusivity of organismic selection (see Chapter 2, pp. 125137). In so doing, these scholars have missed the one area — the heart of Darwin's argument about diversity — where his logic falters because he needs (but hesitates to em­brace in his distress) the apparatus of species selection. I suspect that the in­ternal problems in this centerpiece of Darwin's thought have not been ad­dressed, or even recognized, because species selection itself did not become a subject of importance (or even an acknowledged subject at all) until recently, while debate about conventional group selection has long raged. Darwin has therefore been well combed for comments about interdemic selection, while his main engagement with supraorganismal selection on species went un­noticed.

In any case, whatever our attitude or ignorance today, Darwin clearly regarded his solution to the problem of divergence as his second great achieve­ment (after natural selection), and as the capstone to his theory. As Ernst Mayr notes (1985, pp. 759-760): “He referred to it always with great excite­ment, as if it had been a major departure from his previous thinking.” On June 8,1858, Darwin wrote to Hooker after completing his extended discus­sion of the principle of divergence for Chapter 6 of Natural Selection (the “big species book” that would never be completed because Wallace's paper, arriving within 10 days of this letter to Hooker, derailed his leisurely plans and led him to compose the “abstract” (in his own description) that we call The Origin of Species). “I am confined to the sofa with boils,” he begins, “so you must let me write in pencil.” He then goes on to describe “the 'Principle of Divergence,' which, with 'Natural Selection,' is the keystone of my book” (in F. Darwin and Seward, 1903, volume 1, p. 109).

A year before, in September 1857, Darwin wrote his first complete account of the principle of divergence in a famous letter to Asa Gray at Harvard University. Gray had explicitly asked Darwin for an epitome of his evolution­ary theory (previously revealed only to Darwin's closest confidants, Hooker and Lyell in particular): “It is just such sort of people as I that you have to sat­isfy and convince and I am a very good subject for you to operate on, as I have no prejudice nor prepossessions in favor of any theory at all” (quote in Kohn, 1981, p. 1107). Darwin responded positively with a lucid summary of his theory in six points:

• The power and effect of artificial selection.

• The even greater power of natural selection working on all characters at once and over vastly longer spans of time.  {228} 

• The operation of natural selection at the organismal level, powered by the Malthusian principle that all species produce far more offspring than can possibly survive.

• A description of how natural selection works in nature.

• A defense of gradualism as the solution to standard problems in accept­ing the factuality of evolution.

• An explication of the principle of divergence.

This account of the principle of divergence also became the first published version because Lyell and Hooker included this letter to Gray among the documents published in the Linnaean Society's journal for 1858 — the “delicate arrangement” that presented Darwin and Wallace jointly, stressing Darwin's priority but publishing Wallace's paper on the independent discovery of natu­ral selection in toto. Darwin's sixth point neatly summarizes his ideas on di­vergence:

Another principle, which may be called the principle of divergence, plays, I believe, an important part in the origin of species. The same spot will support more life if occupied by very diverse forms. We see this in the many generic forms in a square yard of turf, and in the plants or in­sects on any little uniform islet, belonging almost invariably to as many genera and families as species. We can understand the meaning of this fact amongst the higher animals, whose habits we understand. We know that it has been experimentally shown that a plot of land will yield a greater rate if sown with several species and genera of grasses than if sown with only two or three species. Now, every organic being, by prop­agating so rapidly, may be said to be striving its utmost to increase its numbers. So it will be with the offspring of any species after it has be­come diversified into varieties, or sub-species, or true species. And it fol­lows, I think, from the foregoing facts, that the varying offspring of each species will try (only a few will succeed) to seize on as many and as di­verse places in the economy of Nature as possible. Each new variety or species, when formed, will generally take the place of, and thus extermi­nate its less well-fitted parent. This I believe to be the origin of the classi­fication and affinities of organic beings at all times; for organic beings al­ways seem to branch and sub-branch like the limbs of a tree from a common trunk, the flourishing and diverging twigs, destroying the less vigorous — the dead and lost branches rudely representing extinct genera and families (from the 1858 published version, often reprinted, as, for example, in Barrett et al., 1987).

Darwin continually awarded his principle of divergence the central role specified in this letter to Gray. Nearly half of the key chapter in the Origin of Species (number 4 on “Natural Selection”) treats the principle of divergence, closing with the celebrated metaphor of the tree of life, sketched out at the end of point 6 to Gray. The only figure in the entire Origin of Species occurs in chapter 4. The intent of this famous diagram (reproduced here as Fig. 3-5 on p. 242) has almost always been misunderstood by later commentators.  {229}  Darwin did not draw this unique diagram simply to illustrate the generality of evolutionary branching, but primarily to explicate the principle of diver­gence.

After more than a century in limbo, Darwin's principle of divergence has been exhumed and subjected to careful scrutiny by historians of science. No subject in Darwinian studies has been more actively pursued during the past 25 years, and many excellent analyses have been published on the genesis and utility of the principle of divergence (Limoges, 1968; Sulloway, 1979; Browne, 1980; Schweber, 1980, 1985, 1988; Ospovat, 1981; Kohn, 1981, 1985). Therefore, the importance of this principle has finally been recognized. Ospo­vat, for example, writes (1981, pp. 170-171): “Darwin's 'principle of diver­gence' [is] the most important addition to his theory between 1838 and 1959 [sic, for 1859] and the one most intimately associated with the transforma­tion of his theory after 1844.” In all this literature, however, only Schweber has grasped Darwin's difficulties with divergence as an unresolved struggle between levels of explanation. Yet this theme, particularly Darwin's inability to “cash out” his usual argument about organismal struggle at the level of species birth and death, holds, I believe, the key to Darwin's treatment.

Darwin's argument about divergence begins with an unquestioned premise that strikes us as curious today (for we are immediately tempted to mount a challenge), but resonates with a central theme of Darwin's century — the clear and inherent “good” of maximizing the amount of life in any given region, and the consequent necessity for a cause to insure this natural goal. Maximi­zation, Darwin argues, arises by diversification: the more taxa in a given area (and the more different), the greater the total quantity of life. This theme can be traced to Darwin's earliest “transmutation notebooks” of the 1830's, the primary documents of his quest to formulate evolution: “The end [that is, goal] of formation of species and genera is probably to add to quantum of life possible with certain pre-existing laws — if only one kind of plant not so many” (C Notebook, p. 146 — in Barrett et al., 1987).

In the fullest discussion within Natural Selection (written in early 1858), Darwin firmly links maximization of life to diversification of taxa: “I con­sider it as of the utmost importance fully to recognize that the amount of life in any country, and still more that the number of modified descendants from a common parent, will in chief part depend on the amount of diversification which they have undergone, so as best to fill as many and as widely different places as possible in the great scheme of nature” (p. 234 of Stauffer edition, 1975).

Darwin proposes that the vague concept of “amount” or “maximization” of life be quantified chemically as total metabolic flow through a given area in a given time — and he illustrates the primary dependence of this quantity on diversification:

The fairest measure of the amount of life [is] probably the amount of chemical composition and decomposition within a given period. Imagine the case of an island, peopled with only three or four plants of the same order all well adapted to their conditions of life, and by three or four insects  {230}  of the same order; the surface of the island would no doubt be pretty well clothed with plants and there would be many individuals of these species and of the few well adapted insects; but assuredly there would be seasons of the year, peculiar and intermediate stations and depths of the soil, decaying organic matter etc., which would not be well searched for food, and the amount of life would be consequently less, than if our island had been stocked with hundreds of forms, belonging to the most diversified orders (ibid., p. 228).

Darwin then provides examples from agriculture and domestication. Sev­eral varieties of wheat, sown together on a plot, will yield more grain per acre than a monoculture. In one experiment, two species of grass yielded 470 plants per square foot, while a plot of 8 to 20 species produced a thousand plants per square foot (ibid., p. 229): “I presume that it will not be disputed that on a large farm, a greater weight of flesh, bones, and blood could be raised within a given time by keeping cattle, sheep, goats, horses, asses, pigs, rabbits and poultry, than if only cattle had been kept” (ibid., p. 229).

But why was Darwin so wedded to a principle of maximization that would strike most of us today as both metaphysical and indefensible (ecosystems, af­ter all, can work perfectly well with far fewer species and lower chemical “yield” per spot)? Schweber (1980), I think, has provided the correct answer by stressing Darwin's allegiance to one of the most popular philosophical ap­proaches of his day — the “Benthamite optimization calculus” promoted by Jeremy Bentham, and many other prominent thinkers in several disciplines, as the utilitarian principle in philosophy and political economy, the “greatest good for the greatest number.” William Paley, the intellectual hero of Dar­win's youth (see p. 116), spoke for a utilitarian consensus in writing (quoted in Schweber, 1980, p. 263): “The final view of all rational politics is, to pro­mote the greatest quantity of happiness in a given tract of country ... and the quantity of happiness can only be augmented by increasing the number of percipients or the pleasure of their perceptions.” In other words, make more objects and make them better. Nature achieves this desired maximization and progress by diversifying the number of species in each region of the globe.

Darwin explicates and defends the maximization of life with his favorite rhetorical device — analogy — and by invoking another fundamental tenet in the political economy of his era: the division of labor. As taxa specialize ever more precisely to definite and restricted roles in local ecologies, more species can be supported (leading to maximization of life as measured by chemical throughput). In a note of September 23, 1856, Darwin drew a direct parallel between diversification in nature and the economic principle of division of la­bor: “The advantage in each group becoming as different as possible, may be compared to the fact that by division of labor most people can be supported in each country.” For the public presentation in the Origin three years later, Darwin retained the centrality of division of labor, but chose a biological analogy drawn from the French zoologist Henri Milne Edwards (who had, himself, credited Adam Smith and the political economists, and who charac­terized his own view as an extension  {231}  of their principle):

The advantage of diversification in the inhabitants of the same region is, in fact, the same as that of the physiological division of labor in the or­gans of the same individual body — a subject so well elucidated by Milne Edwards. No physiologist doubts that a stomach by being adapted to di­gest vegetable matter alone, or flesh alone, draws most nutriment from these substances. So in the general economy of any land, the more widely and perfectly the animals and plants are diversified for different habits of life, so will a greater number of individuals be capable of there support­ing themselves (Darwin, 1859, pp. 115-116).

Consider the form of the classic argument in Darwin's two analogical sources. Adam Smith began the Wealth of Nations by discussing pinmaking to illustrate the advantages of division of labor. Smith states the basic argu­ment in the very first words of his classic book: “The greatest improvement in the productive powers of labor, and the greater part of the skill, dexterity, and judgment with which it is anywhere directed or applied, seem to have been the effects of the division of labor.” Pinmaking, Smith tells us, may be “a trifling manufacture,” but “18 distinct operations” are still needed to make the final product. If a single worker performed all these tasks, he “certainly could not make twenty” pins in a day, but allocation of separate tasks to 10 people (with some individuals continuing to perform 2 or 3 of the 18 opera­tions) allowed one small factory to make 48,000 pins per day, or 4800 per man. Now who benefits from this division of labor? In part, the workers who hone their skills and participate in the resulting prosperity. But primarily the larger polity — the factory through profits, or society itself in the availability of moderately priced goods. Similarly, in Milne Edwards' physiological divi­sion of labor, the prime beneficiary cannot be the organ (an omnivore's stom­ach works perfectly well qua stomach), but again the larger polity, in this case the organism.

Applying the same logic to Darwin's analog, the beneficiary of life's diversification through division of labor is not the individual, or even the species, but the larger polity — or life itself through the principle of maximization. Thus we can grasp the link between division of labor as a pervasive structural principle, and Darwin's goal in application — the summum bonum of maximi­zation of life, achieved through division of labor with the larger polity, or life itself, as the beneficiary.

The logic in this chain of reasoning also illustrates why the coupling of maximization with division of labor cannot validate Darwin's “principle of divergence.” These arguments may indicate why maximization should occur, but do not explain how such a plenteous state of nature arises. In other words, this chain of reasoning does not propose a cause for maximization. Above all else, Darwin clearly understood that his distinctive style of evolu­tionary argument demanded an explanation for any higher level phenome­non as a consequence of struggle among individual organisms for reproduc­tive success. Maximization and division of labor represent phenomenological statements about the constitution of life and ecology not claims about the efficient causes of diversity. Such statements provide, in other words, a basic  {232}  description of diversity, but emphatically not a “principle of divergence.” Darwin's insight in his carriage did not merely systematize the notions of maximization and division of labor; Darwin had known and used these con­cepts for years. The transforming insight — the argument worthy of being called a “principle of divergence” and becoming a keystone of his entire evo­lutionary theory — occurred when Darwin recognized how he could apply his distinctive style of argument, based on organismal selection, to the higher-level phenomenology of diversity. In other words, the “principle of diver­gence” embodies Darwin's argument for how and why ordinary natural selec­tion must, as a predictable consequence, yield divergence of character, leading to multiplication of successful taxa, extinction of others, ecologic plenitude, maximization of life, and the hierarchical structure of taxonomy.


This perspective on the jelling of Darwin's principle of divergence resolves a recent debate among historians about the timing and reason for Dar­win's formulation. Proposals for timing have ranged from the late 1840's to 1858, with cardinal inspirations from biogeography (Sulloway), systematics (Limoges, Ospovat), the “botanical arithmetic” of chapter 2 in the Origin, leading Darwin to defend the greater evolutionary potential for diversifi­cation in large genera (Browne), inspiration from the arguments of politi­cal economy (Schweber) and switch from allopatric to sympatric models of speciation (Kohn). All these influences surely played their parts, for the prin­ciple of divergence calls upon a wide and complex range of convictions, spanning many years and much turmoil in Darwin's mind. But Darwin's for­mulation and formalization of the “principle of divergence” records his con­viction, and his great pleasure, that he could encompass all these ideas as predictable consequences of natural selection working by struggle among organisms — that he could, in other words, bring all the higher-level phenom­enology of maximization, division of labor, and so forth, into his own distinc­tive explanatory framework.

Schweber dates the first full formulation to 1856 and writes (1988, p. 135): “That Darwin had the 'keystone' of the argument by January 1855 is proba­bly correct, but I would also suggest that the argument was still not com­plete in an important way — at least insofar as an explicit presentation is con­cerned. All the arguments up to that point referred to levels of descriptions above individuals: varieties, species, and higher taxa. Natural selection oper­ated on individuals, and the linkage by which diversity is accomplished had to be explicitly stated.” Using this insight, Schweber regards the fol­lowing note of September 23, 1856, as the first explicit formulation.* Just  {233}  as unfettered individual competition yields the best social order in Adam Smith's world, so too will natural selection among organisms lead to max­imization by division of labor: “The advantage in each group becoming as different as possible, may be compared to the fact that by division of labor most people can be supported in each country. — Not only do the individuals in each group strive against the others, but each group itself with all its members, some more numerous, some less, are struggling against all other groups, as indeed follows from each individual struggling” (Dar­win, September 23, 1856, cited in Schweber, 1988). (This consistent stress on the role of individuals as primary causal agents also characterizes the writings in political economy that so influenced Darwin. For example, I omitted by ellipsis an intermediary passage in the statement from Smith quoted above on p. 124. It reads: “Although we speak of communi­ties as sentient beings; although we ascribe to them happiness and misery, desires, interests, and passions; nothing really exists or feels but individu­als. The happiness of a people is made up of the happiness of single per­sons.”)

This proper characterization of Darwin's argument also overturned the most sensational charge ever based on the principle of divergence, and made with such attention in the public arena by Brackman (1980) — the claim that Darwin received Wallace's paper from Ternate earlier than the “official” date (June 18,1858), and then proceeded to steal the principle of divergence from him, thus formulating his complete theory by ripping off Wallace and cover­ing up the evidence. This charge, which can only be supported by ignorance of detail (see the analysis of Kohn, 1981), falls apart once we recognize Dar­win's full principle of divergence as an explanation of maximization by natu­ral selection through division of labor. Darwin clearly formulated this com­plete argument in 1856, and sent a lucid epitome to Asa Gray in 1857. Thus, a possible receipt of Wallace's paper earlier in June, or even in late May of 1858, cannot affect this chronology.

Brackman, of course, does not deny these facts. He must therefore claim that Darwin had been spooked by Wallace for years, that he pinched the ini­tial idea of diversification from Wallace's 1855 paper, and that he then moved faster (and stealthily) when the firmer statement arrived in 1858. But if we turn to Wallace's 1855 paper, we note that this article contains nothing relevant to a principle of divergence properly defined as a set of complex arguments for linking natural selection on organisms with the phenomenol­ogy of higher levels of biological organization. At most, Wallace's 1855 paper includes a passing description of the simple property of divergence itself — a fact well recognized by Darwin, who had been noting the centrality of this theme since the transmutation notebooks of the late 1830's (see quote on p. 229). (We shall also see, at the close of this section (p. 248), that even Wallace's 1858 paper contains only a cursory statement about divergence with no hint of the central feature of bridging levels.) Brackman has con­fused the noting of a fact with the development of an explanation. He has also failed to recognize Darwin's long awareness of the fact and its impor­tance.  {234} 


In Chapter 2, I noted the radical character and intellectual power of Darwin's primary argument as embedded in the Malthusian insight about natural se­lection. In an irony that overturned the entire tradition of natural theology, Darwin held that all the higher order “harmonies” of good design and eco­logical balance arose as side consequences of a process — struggle among or­ganisms for personal reproductive success — that would demand an opposite interpretation if we sought moral messages in nature. Now, in the mid 1850's, Darwin attempted the same philosophical coup to accomplish for diversity exactly what he had done for adaptation in the initial formulation of natural selection — that is, to render a higher level “good,” the maximization of life through division of labor, as a side consequence of organismic struggle. In January 1855 (in the note that Schweber views as the genesis of the principle of divergence), Darwin takes this fateful repeat step into the philosophical radicalism of rendering higher harmonies by individual struggle (quoted in Schweber, 1988): “On Theory of Descent, a divergence is implied and I think diversity of structure supporting more life is thus implied ... I have been led to this by looking at heath thickly clothed by heather and a fertile meadow both crowded, yet one cannot doubt more life supported in second than in the first; and hence (in part) more animals are supported. This is the final cause but mere result from struggle (I must think out last proposition).”*  {235} 

So Darwin recognized in early 1855 that maximization would have to be explained by natural selection (a “mere result from struggle”); he also stated that development of such an argument would be complex and difficult (“I must think out last proposition”). The “principle of divergence of character,” or more succinctly the “principle of divergence,” emerged as the result of this intellectual labor. How, then, did Darwin finally render maximization of life as a consequence of struggle, or ordinary natural selection?

Darwin's solution, embedded as Kohn notes (1985) in his increasing will­ingness to accept sympatric speciation, holds that natural selection will gener­ally favor the most extreme, the most different, the most divergent forms in a spectrum of variation emanating from any common parental stock. Thus, each vigorous and successful stock produces a cone of varying forms about its own modal design (see Fig. 3-5 on p. 242). If natural selection generally fa­vors extreme variants in such arrays — the core claim of the “principle of di­vergence of character” — then vigorous ancestors will generate two or more descendant taxa fanning out towards maximally different form and adapta­tion. Two sequelae now complete the argument by drawing both ecological plenitude and taxonomic structure from the principle of divergence: First, the process of divergence must continue (see Fig. 3-5), impelling each vigorous descendant to produce still more advantageous extremes — thereby entraining phyletic trends of constantly increasing specialization. (The full extension ele­vates subspecies to species, species to genera, etc. — as extreme variants prolif­erate and diversify. The taxonomic tree of life emerges as an ultimate result.) Second, descendants will, in general, be competitively superior to parents, and must therefore tend to exterminate them in competition — for the number of species cannot increase indefinitely, and some ecological mechanism for re­placement of ancestors must exist.

In the Origin, Darwin begins, in his characteristic fashion, by analogy to artificial selection. Breeders, he argues, tend to favor extreme variants when trying to improve a stock; nature must follow suit (1859, p. 112). For the breeder's conscious aim, Darwin substitutes the natural advantages of ex­treme variants: “The more diversified the descendants from any one species become in structure, constitution, and habits, by so much will they be better enabled to seize on many and widely diversified places in the polity of nature, and so be enabled to increase in numbers” (1859, p. 112). With a botanical example, Darwin then strongly argues that divergence occurs because natural selection tends to favor extreme variants:

We well know that each species and each variety of grass is annually sowing almost countless seeds; and thus, as it may be said, is striving its utmost to increase its numbers. Consequently, I cannot doubt that in the course of many thousands of generations, the most distinct varieties of any one species of grass would always have the best chance of succeeding and of increasing in numbers, and thus of supplanting the less distinct varieties; and varieties, when rendered very distinct from each other, take the rank of species (1859, pp. 113-114).  {236} 

Parental forms will then tend to succumb because natural selection favors their extreme and divergent descendants in competition: “As in each fully stocked country natural selection necessarily acts by the selected form having some advantage in the struggle for life over other forms, there will be a con­stant tendency in the improved descendants of any one species to supplant and exterminate in each stage of descent their predecessors and their original parent” (1859, p. 121).

All evolutionists know that the Origin of Species contains only a single fig­ure. This statement has been endlessly repeated in textbooks and lectures, but the true significance of this figure remains obscure, because we nearly always misinterpret the diagram (Fig. 3-5). We read this sole figure as Darwin's basic illustration of evolution as a branching process. But Darwin did not construct his diagram for such a general purpose. Rather, he devised this unique figure to provide a surgically precise description of the principle of divergence, ac­companied by several pages of explanatory text (pp. 116-126). Note how only two species of the original array (A—L) ultimately leave descendants — the left extreme A and the near right extreme I. Note how each diversifying species first generates an upward fan of variants about its modal form, and how only the peripheral populations of the fan survive to diversify further. Note that the total morphospace (horizontal axis) expands by divergence, even though only two of the original species leave descendants. Darwin writes (1859, p. 121): “In each genus, the species, which are already ex­tremely different in character, will generally tend to produce the greatest number of modified descendants; for these will have the best chance of filling new and widely different places in the polity of nature: hence in the diagram I have chosen the extreme species (A) and the nearly extreme species (I), as those which have largely varied, and have given rise to new varieties and spe­cies.” Darwin also states that the success of extremes records the action of natural selection in its usual mode of organismic struggle: “And here the im­portance of the principle of benefit being derived from divergence of charac­ter comes in; for this will generally lead to the most different or divergent variations (represented by the outer dotted lines) being preserved and accu­mulated by natural selection” (1859, p. 117).


I apologize to readers for this laborious (though, I trust, not uninterest­ing) exposition of Darwin on divergence, but I have now reached the crux both of this argument and, in one sense, of this entire book as well. I am ad­vocating both the necessity and importance of a hierarchical expansion of the theory of natural selection, defending this position by combining the stan­dard techniques of validation in science and scholarship: empirical example, the logic of argument, and historical illustration. Why, then, should so much space be accorded to Darwin's views on divergence of character — especially since I have just documented Darwin's attempt to render this second keystone  {237}  of his theory as a consequence of ordinary natural selection at the organismic level?

I have proceeded in this way for a simple reason: Darwin's argument doesn't work, and he came to recognize this failure in the face of his brave at­tempt. Darwin struggled mightily to render this second keystone of his full theory by natural selection alone, but he could not carry the logic to comple­tion. He failed because his full argument demands a major contribution from species level selection (or, at the very least, strong attention to explicit sorting at the species level). I don't know that Darwin ever grasped this need in a fully explicit way, committed as he was to the exclusivity of selection on organ­isms. But he recognized the crucial difficulty at several places in his exposi­tion; and, with his usual honesty, he made his distress palpable again and again.

I perceive his discomfort in the labored description of divergence given in the Origin — 15 pages for a few points repeated many times, first in one way, then in another, all in a book so compressed that Darwin wanted to include the word “abstract” in the title. (John Murray, his publisher, demurred on ob­vious practical grounds). I sense Darwin's malaise in the fact that for this con­cept alone (among all the complex ideas developed in the Origin) he supplied both a figure and a meticulous “caption” (as it were), running for nearly ten pages. I note his dissatisfaction in the frequent shifting of attribution within his text — from consequences of organismic struggle (his usual and distinctive argument) on the one hand, to advantages for higher level units (usually spe­cies) on the other. In my reading, these shifts cannot be interpreted as com­fortable transitions rooted in a confident reduction of higher level phenomen­ology to lower level causality (as he had achieved in explaining adaptation by natural selection), but must instead be regarded as genuine gropings and confusions. Finally, Darwin recorded his distress in explicit exclamations of doubt — from his “I must think out [this] last proposition” of the 1855 note, to his description of divergence as “this rather perplexing subject” in the Ori­gin (1859, p. 116).

If the founder of the non-hierarchical organismal view, this doggedly persistent, fiercely honest and brilliant thinker, tried so hard to make the canoni­cal argument of natural selection work for the central higher-level phenome­non of species diversity — and could not bring the logic of his argument to a satisfactory completion — then perhaps his failure tells us something about the necessity of hierarchical selection. Darwin, by his own formulation, faced two great issues — adaptation and diversity. He tried to render both by natural selection based on struggle among organisms — adaptation by the Malthusian insight of 1838, diversity by the principle of divergence formulated in the mid 1850's. His explanation worked well, or at least sprung no logical holes, for adaptation. But he could not carry the same argument through, despite exten­sive and valiant attempts, for diversity — the primary domain of species selec­tion, as all modern advocates hold (see, for example, Gould and Eldredge, 1988, replying to Maynard Smith's 1988 misconception).

Schweber (1980, 1985, 1988) noted Darwin's trouble with discordance between  {238}  levels, though he does not provide the technical arguments detailed be­low. In Schweber's view, Darwin was driven to formulate an argument that “does not cohere” (Schweber, personal communication) because his century's ignorance of hereditary mechanisms drove him to describe variation within species and varieties treated as units, while the causal structure of natural se­lection rested upon individual organisms. Arguments about organisms and species are not comfortably intertwined or mutually supporting within Dar­win's conceptual structure: rather, the two levels remain discordant and inad­equately (if not illogically) bridged. “This difference in the 'units' used is im­portant. It accounts for the fact that at times levels of description were interchanged and some confusion necessarily crept in” (Schweber, 1980, p. 240). “There was no link between adaptation and speciation, except what­ever could be supplied by a quasi-historical developmental idea of optimizing the amount of life” (ibid., pp. 287-288). “The problem of the different levels of descriptions was confined to how the properties of variations in individuals ... were responsible for the assumed variability characteristic of varieties and species. This problem Darwin never solved” (ibid., p. 288).

We can exemplify Schweber's perceptions about Darwin's incoherence of argument by dissecting the logic of Darwin's attempt to use ordinary natural selection as the basis of divergence. For three basic reasons, his attempt to in­voke selection among organisms as an explanation for patterns in speciation and extinction — the heart of the “principle of divergence,” and the primum desideratum for a complete theory of natural selection — fails because the level of species must be addressed both directly and causally, while Darwin's rationale for explanation from below includes gaps and fatal weaknesses.

The calculus of individual success

Darwin treats the principle of divergence in two extensive discussions — the long and even labored account of chapter 4 in the Origin of Species (1859, pp. 111-126), and the even more detailed exposition intended for the “big species book” that Wallace interrupted and Darwin never published because he rushed to compose the Origin instead. The manuscript for most of this larger project survives, including the full discussion of divergence intended for chapter 6 “On Natural Selection.” This text, published under R. C. Stauffer's editorship in 1975, treats the principle of divergence on pages 227-251.

When, in the 1970's, I first read the Origin with the notion of hierarchical selection in mind, I was fascinated by Darwin's struggle to bridge the levels, and his ultimate lack of success. Schweber speaks of “incoherence”; I would rather describe Darwin's “moves” of argument as an oscillation between one mode and the other. In some passages (including those cited above), Darwin speaks of ordinary natural selection and the advantages enjoyed by extreme variants. In others, he judges the success of a parental form not by the vigor or competitive prowess of offspring, but by the number of descendant species emanating from a rootstock.

These themes could, of course, be complementary. One perspective might  {239}  imply and grade into the other; the two levels could represent alternate solu­tions of the Necker cube (see Dawkins, 1982) — that is, views of the same con­figuration from different vantage points. The individual success of extreme organisms may simply imply, ipso facto and necessarily, the ultimate multipli­cation of species, and success measured by number of descendant taxa might therefore act as a surrogate for, or simple extension of, natural selection. (For this reason, I assume, Darwin includes the principle of divergence as the larg­est section of chapter 4, entitled natural selection.)

But this argument does not cohere, and I think that any careful reader must be struck by Darwin's discomfort as he mixes and juggles the argument for success of extreme organismal variants with the calculus of advantage mapped by number of descendant taxa. Darwin's argument falters because the use of a lower level (success of extreme variants, in this case) to explain a phenomenon at a higher level (multiplication of species) can only work if “perfect transfer” can be defended — that is, if the lower level entails the higher as a direct consequence without any intervention (even a synergistic boost, not to mention a contrary force) from causes at the higher level itself. Darwin understood this principle perfectly well. Indeed, he was probably the only man who, in this infancy of evolutionary science, had carefully and con­sistently thought the logic of selection through to this correct interpretation. Thus, Darwin tried to construct an argument for perfect transfer — but he failed. Darwin advanced these claims in Natural Selection, but ultimately dropped them from the Origin, because (I suspect) he recognized their weak­nesses. (I criticize the arguments for perfect transfer in points two and three below; in this subsection, I document Darwin's bold claim for a higher level calculus of individual success.)

Consider Darwin's first hypothetical example of the principle of divergence in the Origin (1859, p. 113). He speaks of a “carnivorous quadruped” that, by ordinary natural selection, has expanded in population to the limits of lo­cal environments. To do even better in the struggle for life, this form must now diversify into several descendant taxa. But how can the canonical argu­ment for natural selection be cashed out in terms of multiplicity of descen­dant species? The logic of individual struggle carries no implications about the splitting of populations (especially for an in situ sympatric splitting that implies general predictability rather than the simple good fortune of geo­graphic isolation).

Take the case of a carnivorous quadruped, of which the number that can be supported in any country has long ago arrived at its full average. If its natural powers of increase be allowed to act, it can succeed in increasing (the country not undergoing any change in its conditions) only by its varying descendants seizing on places at present occupied by other ani­mals: some of them, for instance, being enabled to feed on new kinds of prey, either dead or alive; some inhabiting new stations, climbing trees, frequenting water, and some perhaps becoming less carnivorous. The  {240}  more diversified in habits and structure the descendants of our carnivo­rous animals became, the more places they would be enabled to occupy (Darwin, 1859, p. 113).

Darwin continually invokes the calculus of individual success by number of descendant taxa (see also 1859, p. 116, and Stauffer, ed., 1975, p. 228): “As a general rule, the more diversified in structure the descendants from any one species can be rendered, the more places they will be enabled to seize on, and the more their modified progeny will be increased” (Darwin, 1859, p. 119).

Often, he mixes both criteria — the adaptive success of extreme variants in struggle, and the calculus of descendant taxa — in a single statement.

Here in one way comes in the importance of our so-called principle of divergence: as in the long run, more descendants from a common parent will survive, the more widely they become diversified in habits, constitution and structure so as to fill as many places as possible in the polity of nature [organismic level], the extreme varieties and the extreme species will have a better chance of surviving or escaping extinction, than the in­termediate and less modified varieties or species [taxon level]. But if in a large genus we destroy all the intermediate species, the remaining forms will constitute sub-genera or distinct genera, according to the almost ar­bitrary value put on these terms (in Stauffer, ed., 1975, p. 238).

But are these two statements really equivalent? Does the lower level claim for organismic advantage imply the higher level phenomenon of species prolifer­ation without reference to any higher level causes?

In his most striking passage, indicating that he did grasp the need for higher level sorting based upon such group properties as the range of varia­tion, Darwin attributes the success of introduced placentals in Australia not, as we might anticipate from ordinary natural selection, to the adaptive biomechanical superiority of placental design (honed in the refiner's fire of more severe competition in Eurasia and America), but to a greater range of placen­tal variation across taxa, produced by their later stage in the historical pro­cess of divergence.

A set of animals, with their organization but little diversified, could hardly compete with a set more perfectly diversified in structure. It may be doubted, for instance, whether the Australian marsupials, which are divided into groups differing but little from each other, and feebly repre­senting, as Mr. Waterhouse and others have remarked, our carnivorous, ruminant, and rodent mammals, could successfully compete with these well-pronounced orders. In the Australian mammals, we see the process of diversification in an early and incomplete stage of development (Dar­win, 1859, p. 116).

The causes of trends

Trends represent the primary phenomenon of evolution at higher levels and longer time scales. Trends therefore pose the key challenge, the ultimate making  {241}  or breaking point, for extrapolationist theories that seek the causes of macroevolution in microevolutionary processes centered upon organismic selection. Darwin understood and accepted this challenge; his principle of divergence marks his attempt to depict trends as extrapolated results of natural selection. (The principle of divergence attempts to explain morphological trends by specialization and progressive departure from ancestral form, and also to account for numerical trends by multiplication of some taxa at the ex­pense of others within a clade.)

Advocates of species selection hold that trends must be described as the differential birth and death of species (not the simple anagenetic extrapolation of change within a population), and that the causes for such differentials must be sought, at least in part, in irreducible species-level fitness (see Chap­ter 8). The standard extrapolationist rejoinder invokes two arguments: (1) Differential death and survival rather than differential birth (of species) usu­ally fuels trends. The death and persistence of groups can be reduced more easily to organismic competition, while differential production of species more often demands irreducible causes, for an organism cannot speciate by itself, while the death of a population may represent no more than the accu­mulated demise of all organisms. (2) The cause of differential survival or death must be reducible to ordinary natural selection.

Darwin did not offer his principle of divergence as a rejoinder to any explicitly developed theory of species selection, for no such formulation existed when he wrote. But he understood the logical requirements of his theory so well that he provided the necessary rationale without the spur of a formally stated alternative. He also, and uniquely, reinforced his argument with an il­lustration of the need for differential survival of certain kinds of variants within random arrays. In Natural Selection, Darwin presents his case as a second figure (reproduced here as Fig. 3-6) that he did not include in the abridged Origin of Species in 1859. (Virtually no one knew about the exis­tence of this figure or argument until Stauffer published the manuscript of Natural Selection in 1975.)

The basic figure of both Natural Selection and the Origin illustrates Dar­win's claim that only a few vigorous species will produce the variants leading to the “recruitment” of new species. (These vigorous species are the extreme forms favored by ordinary natural selection — A and M in Natural Selection, A and I in the Origin, see Figs. 3-5 and 3-6.) The variants of these vigorous species radiate in an even fan, or random array, about the modal form of their ancestor. A trend then arises because ordinary natural selection favors ex­tremes within the fan. Darwin recognizes that a trend to specialization and diversity cannot be generated only by the greater vigor of extreme species in the initial array; he must also defend a second proposition about differential survival among offspring of these favored extremes.

The second figure of Natural Selection now comes into play (Fig. 3-6). Darwin shows us what happens under a regime of random survival within the fan of variants generated by the favored vigorous species at ecological extremes, as opposed to a regime of selection positively directed towards extreme members


3-5. The famous and only diagram published by Darwin in the Origin of Species, 1859. Darwin did not construct this much misunderstood diagram as a simple and general model of branching phylogeny, but quite specificaiiy as an illustration of his Principle of Divergence. Note that each species tends to produce a fan of variants and that the most successful forms tend to emerge from extreme positions of each fan. See text for details.


3-6. The expanded version of Darwin's figure, drawn in preparation for the long version, Natural Selection, that was never published because Darwin hurried to complete the shorter Origin of Species after receiving Wallace's manuscript. This version was not published until 1975 (see Stauffer, 1975). Here, Darwin shows us how the expectation of enhanced success for extreme variants within each fan, as predicted by his Principle of Divergence, will lead to trends (upper part of diagram); whereas random survival of variants (lower part of diagram) yields no trend and no ecological expansion. See text for details.

of each fan. The trend to diversification halts in the random regime. The final products need not become any more distinct than the initial parents (f10, h10, and l10 lie right under ancestor A, while m10 does not differ from parental M). On the other hand, differential survival of extremes within the fans will produce trends (also seen in Fig. 3-6). Darwin writes of the second diagram  {244}  (the lower half of Fig. 3-6), where variation to the left and right of A repre­sents greater or lesser adaptation to drought in plants: “Everything is the same as in diagram I... except that it is left to mere chance in each stage of descent, whether the more or less moisture loving varieties are preserved; and the result is, as graphically shown, that a10 and l10 [sic, he has no a10 in the drawing, but represents the leftward extreme as f10] differ in this respect; and so in other respects, hardly more than did the first varieties (a1l1) which were produced” (in Stauffer, ed., 1975, p. 244).

The argument for a trend that can be reduced to natural selection therefore hinges upon reasons for differential survival of extremes within the fans of varying species; for the trend cannot emerge simply from the greater evolu­tionary vigor of the ancestral extremes themselves. (Interestingly, Darwin never considers the alternative, more congenial to species selection, of greater production of variants at the extremes, with random survival within fans.) Darwin now makes his crucial move for ordinary natural selection, using his principle of divergence: extremes enjoy differential survival within the fans of variants, because natural selection favors adaptation to peripheral, over ad­aptation to central, “stations” in any region. (We must remember that all members of the fan are well adapted to their own local bits of the environ­ment). Now, at the crux of his development, Darwin tries to defend his posi­tion on the differential value of extreme stations, and his argument falls apart — to be rescued only with a forced and self-contradictory ad hoc hypothesis (explicitly stated in Natural Selection, but wisely omitted from the Origin).

Darwin provides two potential reasons for differential success of organ­isms adapted to extreme environmental stations. The first remains perfectly acceptable, and would pass muster today as a standard ecological argument featured in all textbooks — reduced competition in less “crowded” extreme environments: “From our principle of divergence, the extreme varieties of any of the species, and more especially of those species which are now ex­treme in some characters, will have the best chance, after a vast lapse of time, of surviving; for they will tend to occupy new places in the economy of our imaginary country” (in Stauffer, ed., 1975, p. 239).

If Darwin had stopped here, his argument would have remained consistent, if dangerously weak. But his relentless probing would not permit such a course — for he knew that a key problem remained unsolved:* extreme vari­ants may be favored in their own extreme environments, but why should they  {245}  prevail over other variants in more central environments? After all, the more central ancestor continues to survive after the extreme variant buds off. The ancestor should therefore be favored in its own central environment, and the descendant in its new peripheral station. Why, then, should the descendant ever replace the ancestor — so long as central environments persist along with marginal places? In all his writing on divergence, Darwin recognized that trends to specialization could not occur unless extreme descendants tended to wipe out more central ancestors in competition: trends, in other words, re­quired a pattern of differential extinction as well, for the number of species in a region cannot increase indefinitely.

And here, after so much effort and careful development, Darwin bogged down. For this most resolutely higher level phenomenon of the supposed differential success of extreme vs. central species, Darwin could not provide a tenable argument based upon natural selection. With evident discomfort, Darwin resorted to an ad hoc assumption: he argued that while extreme vari­ants adapt to their marginal stations, they also retain all the adaptations of their parents for the original central habitats. Thus, the descendant extremes remain as good as their parents in the ancestral environment, while adding a capacity for survival in marginal habitats.

But how can such a proposition be defended? Why should a species that has left one environment, and explicitly adapted to another, still retain all its prowess in an environment no longer inhabited (and from which it has ac­tively diverged)? Not only does this proposition make no sense prima facie; such a claim also contradicts the canonical argument (often embraced by Darwin and his contemporaries) that specialization leads to “locking in” and decreased flexibility. In short, Darwin knows that he has run into a severe log­ical problem in trying to justify a central implication of his general argument: the differential survival of extreme taxa with a consequently preferential ex­tinction of central species. How can such a pattern be explained — for central and marginal species should not, after all, be in overt competition, and cen­tral environments cannot be regarded as generally more evanescent than ex­tremes? Darwin therefore invokes his ad hoc argument (described just above) for an expanded range of adaptation in extreme species, in order to place the organisms from these extremes into competition with their parents, thus gen­erating a hypothetical explanation for differential parental death in terms of natural selection.

Darwin begins by stating his ad hoc assumption: “As m1 tends to inherit all the advantages of its parent M [see right side of Fig. 3-6], with the additional advantage of enduring somewhat more drought, it will have an advantage over it, and will probably first be a thriving local variety, which will spread and become extremely common and ultimately, supplant its own parent” (in Stauffer, ed., 1975, p. 239).

But Darwin immediately senses a problem and recognizes that descendants might not retain the parental range, and that ancestors might survive the on­slaught of their phyletic children by living in a different station, thereby avoiding competition. “If m1-10 had been produced, capable of enduring more  {246}  drought, but not at the same time enduring an equal amount of moisture with the parent M, both parent and modified offspring might co-exist: the parent (with perhaps a more restricted range) in the dryer stations, and m1-10 in the very driest stations” (ibid., p. 240).

Darwin then rejects this reasonable but debilitating scenario with his ad hoc assumption, though he senses the weakness of his proposal and salvages his argument almost apologetically, especially at the end:

In the imaginary case of the varieties m1-10 which are supposed to inherit all the characters of M, with the addition of enduring more drought; these varieties would inhabit stations, where M could not exist, but in the less dry stations m1-10 would have very little power of supplanting their parent M; nevertheless during unusually dry seasons m1-10 would have a great advantage over M and would spread; but in damper sea­sons M would not have a corresponding advantage over m1-10 for these latter varieties are supposed to inherit all the characters of their parent. So there would be a tendency in m1-10 to supplant M, but at an exces­sively slow rate. It would be easy to show that the same thing might oc­cur in the case of many other new characters thus acquired; but the sub­ject is far too doubtful and speculative to be worth pursuing (ibid., pp. 241-242).

Species selection based on propensity for extinction

Relentlessly probing as usual, Darwin now identifies another weak point in his argument. If divergence follows this predictable and necessary pattern, given the propensity of natural selection to favor extreme variants in all direc­tions, then what prevents this inexorable process from reaching absurdity in a final state of such precise and extended diversification that each species con­tains but a single individual? (This issue became important for Darwin when he moved from his earlier allopatric view of speciation to embrace a largely sympatric model with an intrinsic and predictable “motor” for the generation of species by selection of extreme variants): “But if the time has not yet ar­rived, may it not at some epoch come, when there will be almost as many spe­cific forms as individuals? I think we can clearly see that this would never be the case” (in Stauffer, ed., 1975, p. 247).

Darwin proposes three reasons for nature's avoidance of such an absurd outcome, the first conventional, but the second and third invoking species se­lection on population properties of size and variability. For the first reason, Darwin cites ecological notions that have since become standard — limiting similarity and a restricted number of “addresses” in the economy of nature. Diversity does beget more diversity, and the physical environment sets no strict a priori upper bound. But limits imposed by “inorganic conditions” will eventually cause selection to rein in the intrinsic process of ever finer diversi­fication:

Firstly, there would be no apparent benefit in a greater amount of modification than would adapt organic beings to different places in the polity  {247}  of nature; for although the structure of each organism stands in the most direct and important relation to many other organic beings, and as these latter increase in number and diversity of organization, the conditions of the one will tend to become more and more complex, and its descen­dants might well profit by a further division of labor; yet all organ­isms are fundamentally related to the inorganic conditions of the world, which do not tend to become infinitely more varied (ibid., p. 247).

But Darwin also recognizes that ecological limits may not be sufficient to restrain diversification, and he advances two other mechanisms, based on species selection against traits of small populations. First, Darwin argues that long before diversity reaches a physical limit (as quoted just above), species selection against finely divided taxa with consequently small populations will balance ordinary natural selection for further diversification: “If there exist in any country, a vast number of species (although a greater amount of life could be supported) the average number of individuals of each species must be somewhat less than if there were not so many species; and any species, repre­sented by but few individuals, during the fluctuation in number to which all species must be subject from fluctuations in seasons, number of enemies, etc., would be extremely liable to total extinction” (ibid., pp. 247-248).

Finally, Darwin cites another reason for species selection against popu­lations of unusually small size. Such populations are not only more prone to extinction; they are also less subject to further speciation because such a re­stricted number of individuals per population implies an insufficiency of op­portunity for the origin of rare favorable variants: “Lastly we have seen . . . that the amount of variations, and consequently of variation in a right or ben­eficial direction for natural selection to seize on and preserve, will bear some relation within any given period, to the number of individuals living and lia­ble to variation during such period: consequently when the descendants from any one species have become modified into very many species, without all be­coming numerous in individuals, . . . there will be a check amongst the less common species to their further modification” (ibid., p. 248).

In a lovely closing metaphor, Darwin provides a fine description of the cen­tral hierarchical concept of balance produced by negative feedback between levels. Number of species will equilibrate at a stable level of diversification when positive selection at the organismal level (Darwin's argument for the advantages of extreme variants) becomes balanced by negative selection at the species level (disadvantages of small population size): “the lesser number of the individuals,” Darwin writes, “serving as a regulator or fly-wheel to the increasing rate of further modification, or the production of new specific forms” (ibid., p. 248).

In summary, I have documented both Darwin's discomfort with his forced attempt to explain the primary species-level phenomenon of diversity by nat­ural selection of extreme organismal variants, and his inability to complete the argument without an explicit invocation of species selection against taxa with small populations. His attempt to render his “principle of divergence”  {248}  by organismal selection alone founders on three bases, the first two negative in logical barriers forced by the premise of organismal exclusivity, and the third positive in a potential “rescue” sought by acknowledging a necessary role for species selection: (1) He promotes a calculus of ultimate organismal success in terms of number of descendant taxa, but cannot extend his argu­ment for diversification by selection of extreme variants to achieve the re­quired perfect transfer to the species level. (2) In order to explain trends, he backs himself into a contradictory and ad hoc explanation for the elimination of ancestral forms by natural selection in competition with descendants. (3) Faced with the logical dilemma of runaway diversification under natural se­lection alone, he advocates negative species selection based upon small popu­lation sizes to bring the process of divergence into equilibrium.


The power of a new framework often becomes most apparent in its capa­city to solve small and persistent puzzles. I therefore end this section with a solution to an old conundrum, and with another refutation for Brakeman’s (1980) claim that Darwin pinched the principle of divergence from Wallace, and then lied to cover up the theft. What did Wallace say about divergence? Did he really develop the principle in useful ways that Darwin had not an­ticipated, and might have coveted as his own? When we turn to Wallace's celebrated Ternate paper (sent to Darwin in 1858), we find only a cursory statement about divergence. Wallace only discusses anagenetic trends of de­scendants away from ancestors. He does not even consider the production of multiple taxa from single sources: “But this new, improved, and populous race by itself, in course of time, gives rise to new varieties, exhibiting several diverging modifications of form, any of which, tending to increase the facili­ties for preserving existence, must, by the same general law, in their turn become predominant. Here, then, we have progression and continued diver­gence deduced from the general laws which regulate the existence of animals in a state of Nature, and from the undisputed fact that varieties do frequently occur” (Wallace, 1858, in Barrett et al., 1987).

We must conclude that Wallace regarded a principle of divergence as “no big deal.” He grasped the idea in outline and apparently found no problem therein. His short statement could not possibly have taught Darwin anything useful, for Darwin had already carried the argument far beyond this basic comment. Why then did Wallace fail to share Darwin's puzzlement, excitement and sense of complexity about the principle of divergence? I can imagine two explanations. Either Wallace simply didn't think the issue through to all the difficulties and implications that Darwin recognized. (After all, malarial fits on Ternate are less conducive to deep thought than years of protracted strolling around the sandwalk at Down.) Or he did think the issue through and, finding nothing problematical, therefore devoted little attention to the subject. If the second alternative is correct, then my framework for  {249}  considering the principle of divergence as a problem in levels of selection can resolve Wallace's puzzling lack of appreciation.

As Kottler (1985) has shown, Wallace and Darwin were not identical peas from the pod of natural selection. They battled long and hard on several cru­cial issues, mostly involving Wallace's panselectionism vs. Darwin's more sub­tle view of adaptation (Gould, 1980d). One key area of disagreement cen­tered upon the target of natural selection. Darwin labored to work out a consistent theory that virtually restricted selection to struggles among organ­isms (see Chapter 2 for his interesting reasons). Wallace, as Kottler shows, never grasped the centrality, or even the importance, of the issue of levels and agency for a theory of natural selection. He moved from level to level as the situation seemed to imply, choosing whatever target of selection would best support his panselectionist leanings. (For example, he ascribed hybrid steril­ity to species selection in order to preserve his conviction that features of such importance must originate as active adaptations; whereas Darwin, committed to a consistent theory of organismal selection, regarded hybrid sterility as an incidental side consequence of accumulated differences arising by ordinary selection in two initially isolated lineages — see p. 131.) Thus, if Wallace ever pondered the principle of divergence to the point of recognizing an issue in levels of selection, he would not have responded, as Darwin did, with such a sustained, almost impassioned, quest for resolution. Wallace would not have identified any problem at all, for he never grasped the thorny issue of a need to specify levels in the first place. A simple statement about divergence would have sufficed — as Wallace indeed provided. Darwin, in his over reliance upon organismal selection, may never have reached the finish line in explicating the principle of divergence; but Wallace scarcely got off the blocks.


No one would argue that persistence in history makes anything right or even worthy — lest cruelty, murder and mayhem win our imprimatur by a misplaced criterion of longevity. Still, in the world of ideas, long pedigrees through disparate systems, and recurrence in the face of attempted avoid­ance; usually signify something about the power of an idea, or its necessary place in the logic of a larger enterprise.

Causal — and not merely descriptive — accounts of hierarchy have infused evolutionary biology in this way from the beginning of our subject. Lamarck initiated this conceptual rubric with a version in the invalid mode of causal differences based on opposition between levels. Darwin knew what he didn't like about this style, and his theory — preserved unchanged to our own ortho­dox commitments today — sinks a strong foundation in an active rejection of hierarchy in this inherently combative mode. Weismann and Darwin him­self — the two greatest evolutionists and deepest thinkers with an explicit commitment to the single-level theory of natural selection — tried to extend the logic of this idea to encompass every important issue in evolution. Both  {250}  became stymied, and eventually surrendered, with pleasure in Weismann's case at least, to the need for hierarchy in devising any complete system for the logic of evolutionary explanation in selectionist terms — Weismann on subcellular selection for explaining trends, Darwin on species selection for encompassing diversity. Both men accepted causal hierarchy in the modern and valid sense of similar forces working in distinct ways at different levels.

Very few evolutionists know anything about this history, and may there­fore doubt the importance of the subject. But downgrading on this criterion would represent a great mistake rooted in the conservative premise that any­thing vital must be easily visible in all contexts. Anyone tempted to accept such a basis for dismissal should consider the conventional tales of rulers and conquerors — virtually the only subject matter of so many secondary school and undergraduate history courses — and recognize what modern scholars have taught us by probing the hitherto invisible pathways of daily life among ordinary people. Compare the overt and conventional history of diplomacy with the often more potent, but academically invisible, history of technology.

Evolutionary thought began with hierarchy, wrongly conceived. Our canonical theory of natural selection arose as an attempted rebuttal. The most brilliant practitioners of that theory could not bring the argument to comple­tion; so both Weismann and Darwin brought hierarchy back, in a valid style this time, to render the theory of selection both coherent and comprehensive. I previously offered a choice of proverbs: you may view hierarchy as a bad penny or a pearl of great price. But hierarchy, like the poor, has always been with us (and, perhaps, shall inherit the earth as well!). This situation can only recall James Boswell's famous statement about one of Dr. Johnson's col­leagues who lamented that he had tried to be a philosopher, but had failed be­cause cheerfulness always broke through. Too many of us have tried to be good Darwinian evolutionists, and have felt discouraged because hierarchy always breaks in. I suggest that we rejoice — with good cheer — and welcome an under appreciated and truly indispensable old friend.



Internalism and Laws of Form:
Pre-Darwinian Alternatives
to Functionalism

Prologue: Darwin's Fateful Decision

Thinking in dichotomies may be the most venerable (and ineluctable) of all human mental habits. In his Lives and Opinions of Eminent Philosophers (circa A.D. 200), Diogenes Laertius wrote: “Protagoras asserted that there were two sides to every question, exactly opposite to each other.”

Darwin follows this tradition of dichotomy in a passage that he earmarked for special impact as the concluding paragraph of his crucial Chapter 6, “Difficulties on Theory.” I regard this passage as among the most important and portentous in the entire Origin, for these words embody Darwin's ulti­mate decision to construct a functionalist theory based on adaptation as pri­mary, and to relegate the effects of constraint (a subject that also commanded his considerable interest — see Section IV of this chapter) to a periphery of low relative frequency and subsidiary importance. Yet this passage, which should be emblazoned into the consciousness of all evolutionary biologists, has rarely been acknowledged or quoted. Darwin begins (1859, p. 206), ex­pressing his alternatives in upper case (and using the categories of the great debate between Cuvier and Geoffroy — see Section III of this chapter): “It is generally acknowledged that all organic beings have been formed on two great laws — Unity of Type, and the Conditions of Existence.”

Conditions of Existence, of course, express the principle of adaptation — final cause or teleology to pre-evolutionists. Organisms are well designed for their immediate modes of life — and intricate adaptation implies an agent of design, either an intelligent creator who made organisms by fiat as an expres­sion of his wisdom and benevolence, or a natural principle of evolution that yields such adjustment between organism and environment as a primary re­sult of its operation. (Both Darwinian natural selection and Lamarckian re­sponse to perceived needs, for example, build adaptation as the most general consequence of their basic mode of action.)

Darwin then continues by defining the other side of the classical dichot­omy: Unity of Type (1859, p. 206): “By unity of type is meant that fundamental  {252}  agreement in structure, which we see in organic beings of the same class, and which is quite independent of their habits of life.” In another critically placed passage, introducing the subject of “Morphology” in Chapter 13, Darwin waxes almost poetic about unity of type (p. 434): “This is the most interesting department of natural history, and may be said to be its very soul. What can be more curious than that the hand of a man, formed for grasping, that of a mole for digging, the leg of the horse, the paddle of the porpoise, and the wing of the bat, should all be constructed on the same pattern, and should include the same bones in the same relative proportions.”

These two principles have always dwelled together in exquisite tension. Any complete account of morphology must call upon both phenomena, for most organisms are well adapted to their immediate environments, but also built on anatomical ground plans that transcend any particular circumstance. Yet the two principles seem opposed in a curious sense — for why should structures adapted for particular ends root their basic structure in homologies that do not now express any common function (as in Darwin's example of mammalian forelimbs)?

The designation of one principle or the other as the causal foundation of biology virtually defines the position of any scientist towards the organic world and its causes of order (see, especially, Russell's superb 1916 book on this dichotomy). Shall we regard the plan of high-level taxonomic order as primary, with local adaptation viewed as a set of minor wrinkles (often con­fusing) upon an abstract majesty? Or do local adaptations build the entire system from the bottom up? This dichotomy set the major debate of pre-Darwinian biology: does God reveal himself in nature primarily by the harmony of taxonomic structure, or by the intricacies of particular adaptations (see Section II, this chapter)? This dichotomy continues to define a major issue in modern evolutionary debates: does functional adaptation or structural con­straint maintain priority in setting evolutionary pathways and directions (see Chapters 1011)?

This issue of primacy between the two principles has held the central stage of natural history for so long that national traditions have developed, with continental preferences usually emphasizing unity of type (despite important exceptions like Georges Cuvier), and mainstream anglophonic science gener­ally favoring adaptation (with exceptions for a few important pluralists like Richard Owen, or dissenters like William Bateson or D'Arcy Thompson). We often blunder in our historical understanding by assuming that evolution must be an ultimate watershed, marking a complete break between a bad be­fore and an enlightened after. In fact, much continuity pass right through Darwin's rupture of history, with evolution only providing a different expla­nation for unaltered principles and phenomena. The good ship Dichotomy — Unity of Type vs. Conditions of Existence — entered the Darwinian current by converting its terms from a debate about God's primary mode of self-expression in nature to an argument about constraint and adaptation in evolution.

We cannot understand Darwin without grasping this fundamental continuity  {253}  in national styles. As a young man, Darwin adored Paley's Natural Theol­ogy (see p. 116); later, in a courageous act of intellectual parricide, he con­structed a theory that subverted Paley's mode of explanation. But Darwin never abandoned Paley's conviction that adaptation must be designated as the primary phenomenon of natural history. Darwin remained true to an English tradition stretching at least as far back as Robert Boyle and John Ray in the late 17th century of Newton's founding generation for modern science, run­ning through Paley, the Bridgewater Treatises, Wallace, and Poulton in Dar­win's own time, on to R. A. Fisher and finally to E. B. Ford, A. J. Cain, and R. Dawkins in the later 20th century.

When we properly place Darwin in this lineage, a genealogy unfractured by evolutionary theory, we can make sense of his fateful decision for resolving Unity of Type vs. Conditions of Existence at the end of Chapter 6 — a choice faithful to Paley and the English tradition in reaffirming the primacy of adap­tation. Darwin writes, in words that define the causal basis of his theory (and continuing from the previous quotation on p. 25):

On my theory, unity of type is explained by unity of descent. The expression of conditions of existence, so often insisted on by the illustrious Cuvier, is fully embraced by the principle of natural selection. For natu­ral selection acts by either now adapting the varying parts of each being to its organic and inorganic conditions of life; or by having adapted them during long-past periods of time: the adaptations being aided in some cases by use and disuse, being slightly affected by the direct action of the external conditions of life, and being in all cases subjected to the several laws of growth. Hence, in fact, the law of the Conditions of Existence is the higher law; as it includes, through the inheritance of former adapta­tions, that of Unity of Type (1859, p. 206).

Darwin's brilliant intellectual move clearly expresses the revolutionary im­pact of evolutionary explanations against the previous range of creationist paradigms. Creationist biology saw Unity of Type and Conditions of Exis­tence, homology and adaptation, as opposite, but equally contemporary (or timeless), poles in a dichotomy of originating forces. Darwin literally added a new dimension to the debate — the axis of history. (And no intellectual expan­sion can be more profound than the introduction of a new dimension, or­thogonal to previous modes of explanation.)

Thus, in this passage, Darwin makes a stunningly simple suggestion to break the impasse between Unity of Type and Conditions of Existence. (And yet, to be able to see anything at all in this clear and simple light, one must first grasp the revolutionary implications of evolution itself — the truly dif­ficult intellectual transition out of Paley's world!) To be sure, the homologies of Unity of Type do not embody, and seem actively to oppose, current func­tions. Must Unity of Type therefore represent a principle of order dichotomously contrary to adaptation? In a world without history, where all features of organisms express their initially created state, the answer must be “yes.” But the addition of history, by a theory of genealogical connection, permits  {254}  (and even privileges) another interpretation. Suppose that Unity of Type re­cords no mysterious groundplan of created design, but only the actual, re­tained form of a common ancestor at the base of a bush of descent? Then homology can be simply explained as passive retention in the genealogy of diversified descendants — not an archetype of intelligent design, but only the signature of history.

In this context of evolutionary reform, we may then inquire about the causes of these common ancestral structures in distant pasts. And Darwin now makes his fundamental choice by affirming fealty to the English lineage of adaptationist thought. He argues that ancestral structures, forming the great homologies of Unity of Type, initially arose, by natural selection, as adaptations to “organic and inorganic conditions of life” in ancestral envi­ronments. Thus, the dichotomous poles of Unity of Type and Conditions of Existence achieve a single and unified explanation under natural selection — as immediate adaptations to present environments (Conditions of Existence), or as adaptations to ancient environments, transmitted by inheritance to diversified descendants (Unity of Type). The old dichotomy, in fact, expresses no clash of opposites at all, but only marks the temporally sequential repre­sentations of one dominant principle in evolution — adaptation by natural selection. Thus, since adaptation embodies the principle of Conditions of Ex­istence, and since adaptation builds both ends of the old dichotomy, Condi­tions of Existence becomes the victorious pole of the old contrast, in Darwin's own words the “higher law; as it includes, through the inheritance of former adaptations, that of Unity of Type.”

Yet Darwin, far too sophisticated a thinker to embrace extreme positions, could not claim that natural selection and adaptation — though responsible for both poles of the old dichotomy — reigned exclusively in nature. Darwin knew that primary judgments in natural history must be rendered in terms of relative frequencies. Indeed, he had written as his last line in the Introduction to the Origin of Species, first edition: “I am convinced that Natural Selection has been the main but not exclusive means of modification” (1859, p. 6). He also reacted as strongly as his genial temperament ever permitted against those who charged him with false claims of exclusivity. In such cases, he usu­ally cited this line from the Origin in vindication — as in his famous, almost rueful statement (Origin, 6th ed., 1872b, p. 395): “As my conclusions have lately been much misrepresented, and it has been stated that I attribute the modification of species exclusively to natural selection, I may be permitted to remark that in the first edition of this work, and subsequently, I placed in a most conspicuous position — namely, at the close of the Introduction the fol­lowing words: 'I am convinced that natural selection has been the main but not the exclusive means of modification.' This has been of no avail. Great is the power of steady misrepresentation.”

Thus, while extending natural selection to cover both poles of the old dichotomy between unity of type and conditions of existence, Darwin also listed the main supplements to selection among causes of evolutionary change: use and disuse, direct action of external conditions, and laws of  {255}  growth. We reject the first two today, and Darwin also grants them little space by his qualifiers: “in some cases” and “being slightly affected.” But Darwin put more store by the third — laws of growth — as indicated by his only positive qualifier: “being in all cases subjected to the several laws of growth.” And we would offer the same judgment today, since laws of growth, under the more fashionable designation of “developmental constraints,” have become a “hot topic” in evolutionary biology once again (see Chapter 10). And now we come to the Darwinian trope of argument, the ploy that makes this chapter (and, to a large extent, this entire book) necessary.

Darwin wrote his crucial closing paragraph of Chapter 6 to argue that Unity of Type should be subsumed under Conditions of Existence — for Unity of Type, he asserted, only expresses past episodes of ordinary adaptation and natural selection, subsequently inherited by numerous modern descen­dants. Unity of Type has always defined the main arena for naturalists who view adaptation as secondary, and some principle of morphological order (for many versions exist) as primary. Darwin removed the rationale for a sep­arate principle of Unity of Type by noting that ancient adaptations would, if inherited throughout a subsequent lineage, become sources of deep homology. Yet he could not deny — and had no desire to subvert — the idea of mor­phological principles working separately from natural selection, and build­ing exceptions to adaptation. In this sense, Darwin supported the concept of constraint, but only if this principle could be carefully circumscribed within a category distinctly subsidiary to natural selection in relative frequency and biological importance. Darwin fully understood the crucial role of relative frequency in evolutionary arguments, and he rested his case for natural selec­tion squarely upon such a judgment of quantitative importance. In so do­ing, he pursued the following strategy: take the old dichotomy, and show that both poles arise as products of natural selection. Then, having removed constraint as the primary cause of one pole (where a high relative fre­quency could not have been denied), allow constraint to reenter as a subsid­iary force to natural selection (with a consequent guarantee of low relative frequency). Natural selection then becomes the primary force of evolution. Recall the full title of Darwin's book: On the Origin of Species by Means of Natural Selection, or the Preservation of Favored Races in the Struggle for Life.

The classical form of a relative frequency argument upholds a favored position and then degrades an alternative by two strategies, both used by Darwin in making constraint subservient to natural selection.

Nooks and crannies. Argue that your principle works nearly all the time, while the alternative occupies just a few subordinate holes of absence. By attributing both poles of the classic dichotomy (unity of type and condi­tions of existence) to natural selection as a primary cause, thereby robbing constraint of its potentially largest domain, Darwin granted dominance to adaptation.

Sequelae. Argue that your principle works as a prior and primary cause (in both temporal order and effect), and that the alternative only produces  {256}  secondary modifications upon this fundamental action. In the closing para­graph of Chapter 6, all forces other than selection become sequelae to its pri­mary action.*

Thus, Darwin invoked relative frequency to uphold his evolutionary world view: a theory of trial-and-error externalism, with natural selection as the only major creative force for change, and with internal variation restricted to the role of generating raw material for selection's perusal, and not of supply­ing important or consistent direction. Why, then, do many evolutionary biol­ogists continue to demur? Darwin's basic argument in closing Chapter 6 can only be judged both brilliant and undoubtedly correct. Most homologies of Unity of Type are, indeed, adaptations inherited from a distant past, not fod­der for constraint theorists who wish to demote the relative frequency and importance of natural selection. (Homologies of Unity of Type do act as phyletic constraints upon present possibilities — elephants will never fly — but such current limitations exist as consequences of initial adaptations, and therefore cannot stand against natural selection in any toting of relative fre­quencies.)

Modern constraint theorists, myself included, balk at Darwin's resolution because his argument demotes a large chunk of biology to a chink in a corner. The old Unity of Type theorists, lacking the alternative of “just history,” did falsely assume that deep homology must stand against adaptation. But much validity still attends their cardinal insight that principles of design, laws of growth, rules of architecture, nature of materials — generalities transcend­ing the particulars of specific genealogical pathways — work as important in­terior channels of constraint in the positive sense of that undervalued word: for constraints not only prevent evolutionary motion by failing to supply variation; they also act positively to set preferred channels of change. Inter­nal forces do not only present isotropic raw material to the fully creative externality of natural selection. Constraint does not exist in subservience to adaptation under the nooks-and-crannies and sequelae arguments of relative frequency. Constraint may never again (and rightly so) be able to claim primacy,  {257}  as the old Unity of Type theorists held, but partnership with adaptation remains a reasonable and minimal demand.

We may epitomize Darwin's brilliant reconstruction of causation in natural history, the new dimension that he added, and the one that we need to rein­sert, in diagrammatic form. Pre-evolutionary theorists, entirely lacking the concept of historical change, attributed created form to a dichotomous dis­tinction of causes: immediate and functional vs. deeper and architectural (Fig. 4-1). Darwin literally added the dimension of history, but removed a previous axis of explanation by redefining constraints of Unity of Type as conse­quences of past adaptation (ancient Conditions of Existence) — Figure 4-2. Yet Darwin understood that he had not abolished the concept of constraint in undermining the primary example — homologies of Unity of Type — by real-location to the opposite camp. He therefore, in the same passage, established a different domain for constraint, as a category subservient to adaptation by the two standard arguments of relative frequency: the spatial claim of lim­ited room (nooks and crannies) and the temporal claim of secondary status (sequelae to adaptation) — see Figure 4-3.

4-1. The standard pre-evolutionary and dichotomous conception of the causes of form as working either by adaptation to immediate conditions of existence, or by manifestation of laws of form that reflect unity of type.

4-2. Darwin literally adds a third dimension of history for the explanation of form. But he greatly devalues the domain previously ascribed to unity of type, admitting constraints of laws of form only by redefining such similarities as homologies based on the inheritance of past adaptations, and therefore adaptational in their origin and primarily due to the other (and now predominant) do­main of conditions of existence.


But many 19th century biologists, and many evolutionists in our day again, feel that Darwin demoted constraint too far, and that the two domains — con­straint and adaptation — must again share potential partnership, as expressed by the important relative frequency of each component. We would therefore restore the strength of the dimension that Darwin first eliminated (when he reinterpreted Unity of Type as a consequence of adaptation) and then reintroduced in weakened form (when he allowed laws of growth to fill nooks and crannies in a domain ruled by natural selection) — see Figure 4-4.

This full model of Figure 4-4 shows three dimensions of form and their interactions: adaptation, constraint, and history. A current trait of an organism may arise as an immediate adaptation to surrounding environments, as a con­straint not particular to the contingent history of its lineage (architectural or structural principles, correlations to current adaptations), or by inheritance of an ancestral form (often called historical or phylogenetic constraint, but quite different in principle from nonhistorical styles of constraint). This dis­tinction suggests a recursion, because contributions from the axis of “history”

4-3. Darwin does allow minor influence for constraint apart from mere inheri­tance of past adaptations. See text for details.

4-4. Constraint reestablished as equal in importance to adaptation as an immedi­ate cause of form. See text for details.


represent traits that, at their origin in an ancestor, arose as either adap­tations or constraints. Nonetheless, the immediate form of an organism can still be meaningfully parsed into three major contributions of current adapta­tion, current constraint, and historical inheritance — Figure 4-5. This insight has generated the various “triangular” models of evolutionary causation that have gained vogue in recent years (see Fig. 4-6).

These issues and parsings have pervaded natural history since Plato and Aristotle argued about abstract form vs. teleology. Darwin made a seminal

4-5. Because constraint and adaptation act either from the past or in the present, we may envisage three primary determinants of present form: present constraint, present adaptation, and inheritance due to past history of either constraint or adaptation.

4-6. One of a group of “triangular” models constructed to express the major in­fluences upon the genesis of form. The three vertices of this triangle refer to the three influences depicted in Figure 4-5.


contribution by adding a dimension for history, and by formulating a theory that granted controlling relative frequency to adaptation. But he did not in­vent the issues, or the scheme of classification. The triumph of Darwinian functionalism did, however, erase much historical memory for the old alter­native of constraint. The next two chapters sink their rationale in a simple premise: Our current need to reinvigorate constraint as a vital topic in evolu­tionary explanation (see Stearns, 1986; Maynard Smith et al., 1985; Gould, 1989a, 1992b) — based upon advancing knowledge of genetic architecture, development and macroevolution — requires that we rediscover this legacy of structuralist thought,* and recognize that the entire history of evolutionary theory has been pervaded by an issue that simply would not disappear, if only because the dialectic of inside and outside, structure and function, design and adaptation, must be resolved at some fascinating interplay and synthesis, not as a victory for either pole in a debate without true sides.

Two Ways to Glorify God in Nature

We cannot comprehend the past from the vantage point of a newly constructed present reality. Once the 19th century had discovered evolution as the primary cause of relationships among organisms, the historical axis not only sprang into being as a pole of explanation, but quickly assumed a pri­mary status (Figs. 4-2 to 4-6). More than a century later, we can hardly imag­ine biology without this theme. What kind of questions could be posed be­fore history became an option for resolution? What kinds of explanations could be rendered when a biologist couldn't ask (or even conceptualize): “How has this feature changed from an ancestral state; what do its differing forms in various species tell us about phyletic relationships; what are the causal bases for both the origin and later alterations of this feature?”

Immediate appearance in a fully formed state provides the only alternative to history — whether such “creation” is achieved by the direct hand of a di­vine agent, or by spontaneous organization from elements according to some unknown law or principle of nature. If basic taxa originated as we find them now, then the range of theoretical explanation remains wide. Species might be purposely ill designed to suit the black humor of a diabolical creator; or they might be cobbled together with no rhyme or reason by forces of universal randomness. The list of possibilities continues ad infinitum.

But, in fact, Western cultural traditions greatly limited the range of acceptable  {261}  alternatives. Very few creationists could imagine that species might be purposely ill formed, or constructed in a disorganized fashion. With these at­tributes — purpose and order — as part of a cultural heritage, the basic expla­nations for organic form could be reduced to two major alternatives, express­ing the primacy of one or the other overarching principle for a rational and benevolent world. These principles have been called structuralism and functionalism, order and teleology, laws of form and adaptation, Unity of Type and Conditions of Existence. These poles set the dichotomy that Darwin ex­panded by introducing history (see last section), but never really fractured be­cause the new axis of time could also be divided into structural vs. functional explanations for ancestral forms. This dichotomy continues to set an impor­tant agenda for evolutionary theory at the opening of a new millennium, especially since the overly adaptationist Modern Synthesis (representing a temporary triumph of the functionalist pole — see Chapter 7) has yielded to a pluralism of structuralist alternatives as partners rather than subsidiary forces (Chapters 10 and 11).

In this light, I find it fascinating that the oldest tradition in modern natural history — the natural theology of so many pre-Darwinian biologists* — also existed in two primary versions, expressing the two poles of the same dichot­omy. Since Darwin built his evolutionary theory in continuity with the pole favored by a long English heritage — the adaptationism of William Paley — this subject cannot be dismissed as an arcane issue from a forgotten past, but re­mains a vital presence in our daily concerns (by our own fundamental evolu­tionary criterion of genealogy and phyletic heritage!). For we still struggle with adaptation and constraint just as Paley and Agassiz contrasted the com­parable positions in natural theology: “the creator foresaw the needs of each species and created just those organs that were necessary to carry them out” vs. “God had in the beginning established laws, and nature was left to unfold in accordance with them” (characterizations of Appel, 1987, p. 7). Do not Fisher vs. Wright, or Cain and Maynard Smith vs. Goodwin and Kauffman carry on the same debate, evolutionarily transmogrified of course?

Natural theology held, as a central premise, that the works of nature not only demonstrated God's presence, but could also reveal his character as well. We could learn about him, not only persuade ourselves that he exists. Paley's full title (1802) reads: Natural Theology: or, Evidences of the Existence and Attributes of the Deity, Collected from the Appearances of Nature. From this shared premise, two traditions proceeded, both “preadapted” to a later evo­lutionary transformation.  {262} 

In this section I shall contrast the two great texts of these alternative traditions — Paley's Natural Theology (1802) with Agassiz's Essay on Classification (1857). The two works dovetail with remarkable symmetry in their op­position: Paley the British adaptationist vs. Agassiz the continental formalist. One might almost believe that the two works were explicitly written to flesh out (and fully clothe) the central dichotomy of form, with each awarded ex­actly half the totality. In a curious sense, this lack of contact almost allows the two texts to speak to each other — as if they formed a sand painting with one (Paley for temporal priority) filling in half the area up to an elaborate and jag­ged boundary, and the other then pouring sand of a different color right up to the previous boundary, leaving no space between at the contact. I am puzzled that these two texts have not been explicitly contrasted before.


Just a few years before Paley wrote his Natural Theology in 1802, Coleridge's Ancient Mariner (1798) proclaimed his hard-won message to a wedding feast, and to the world:

He prayeth best, who loveth best

All things both great and small;

For the dear God who loveth us,

He made and loveth all.

Paley probably appreciated the sentiments and surely longed to extend the argument. He entertained no doubt that all things proclaimed God's existence. But he believed that we must be able to learn more if we hope to use natural theology as a strategy of exegesis. That is, we must also be able to infer im­portant aspects of God's nature and character from the works of creation.

The search to infer God's attributes from general features of natural objects led Paley to open his book with one of the most famous images in all English literature — a strong competitor with Adam Smith's “invisible hand” (a line also found in Paley, 1803, p. 344) and Darwin's tangled bank or tree of life. The good Reverend, crossing a heath on shank's mare, bumps his foot against a stone, feels the pain, but learns nothing about the origin of rocks because the object is too simple and too disordered to reveal a source of production. But if he should then kick a watch, he would surely know that the timepiece had been fashioned by a purposeful agent:

When we come to inspect the watch, we perceive (what we could not discover in the stone) that its several parts are framed and put together for a purpose, e.g. that they are formed and adjusted as to produce motion, and that motion so regulated as to point out the hour of the day... The inference, we think, is inevitable; that the watch must have had a maker; that there must have existed, at some time and at some place or other, an artificer or artificers who formed it for the purpose which we find it actually  {263}  to answer; who comprehended its construction, and designed its use (Paley, 1803, p. 203 — I am using my personal copy of the widely read 1803 edition for all quotes).

Two features of the watch compel this conclusion. First, and less impor­tant, its complexity — for chance could not make anything so intricate: “What does chance ever do for us? In the human body, for instance, chance, i.e. the operation of causes without design, may produce a wen, or a wart, a mole, a pimple, but never an eye” (1803, pp. 67-68). Second, and far more impor­tant, the watch's design, its adaptation to a clearly perceived end.* A high de­gree of order might arise from laws of nature with no reference to final cause, but complexity for a clear purpose implies a designer. “There cannot be de­sign without a designer; contrivance without a contriver; arrangement, with­out anything capable of arranging” (p. 12). Thus does Paley attack his hypo­thetical opponent and partial straw man throughout his work? * “Nor would any man in his senses think the existence of the watch, with its various ma­chinery, accounted for, by being told that it was one out of several possible combinations of material forms; that whatever he had found in the place where he found the watch, must have contained some internal configuration or other; and that this configuration might be the structure now exhibited” (p. 6).

The watch implies, by its utility, a mind capable of forethought, design and construction: “In the watch which we are examining, are seen contrivance, design; an end, a purpose; means for the end, adaptation to the purpose. And the question, which irresistibly presses upon our thoughts, is, whence this contrivance and design. The thing required is the intending mind, the adapt­ing hand, the intelligence by which that hand was directed” (p. 16).

But organisms surely display more complexity and more purposeful design than any watch. Just as Darwin would exalt natural selection as vastly more powerful than artificial human selection in breeding or agriculture, so does  {264}  Paley identifies God's work as incomparably superior to any human art. If the existence of the watch implies a skilled craftsman, how can we even conceive the more awesome skill of he who made all living things: “For every indica­tion of contrivance, every manifestation of design, which existed in the watch, exists in the work of nature; with the difference, on the side of nature, of being greater and more, and that in a degree which exceeds all computa­tion” (p. 19).

In succinct epitome of the entire argument, Paley writes (p. 473): “The marks of design are too strong to be got over. Design must have a designer. That designer must have been a person. That person is God.”

Since we often misuse the past for ridicule, Paley has emerged as every­body's favorite whipping boy from the bad old days of creationism. As a lively writer, he is, to be sure, eminently quotable. And he does sometimes stray into the kind of Panglossian perfectionism (or, rather, far-fetched ratio­nalization for beneficence within apparent evil) that Voltaire savaged with such glee in Candide.

Paley, for example, does engage in “just-so” storytelling to support adaptationist explanation, though he presumably read this account of Babyrussa in a fallacious traveler’s report, and can only be charged with insufficient skep­ticism, not fabrication (Fig. 4-7):

I shall add one more example for the sake of its novelty. It is always an agreeable discovery, when, having remarked in an animal an extraordi­nary structure, we come at length to find out an unexpected use for it. The following narrative furnishes an instance of this kind. The baby-rouessa, or Indian hog, a species of wild boar found in the East Indies, has two bent teeth, more than half a yard long, growing upwards, and (which is the singularity) from the upper jaw. These instruments are not wanted for defense, that service being provided for by two tusks issuing from the under jaw, and resembling those of the common boar. Nor does the animal use them for defense. They might seem therefore to be both

4-7. Paley does not include a drawing of the skull of Babyrussa, but this figure comes from an equally interesting source — P. H. Gosse's Omphalos of 1857, his treatise arguing for the sudden and recent creation of the Earth, including all its fossils, which therefore only display an appearance of great age.


a superfluity and an incumbrance. But observe the event. The animal sleeps standing; and, in order to support its head, hooks its upper tusks on the branches of trees (pp. 270-271).

More in the Panglossian mode, pain (an adaptation, Paley argues, for signaling distress to the mind so that we may care for our bodies) also shows God's benevolence on the theme of the old moron joke — we feel so good when the suffering stops! (On the subject of good in apparent noxiousness, compare John Ray (1735, p. 309) on why God made lice: “I cannot but look upon the strange instinct of this noisome and troublesome creature a louse, of searching out foul and nasty clothes to harbor and breed in, as an effect of divine providence, designed to deter men and women from sluttishness and sordidness, and to provoke them to cleanliness and neatness. God Himself hateth uncleanliness, and turns away from it.” Or, as Robert Burns would later generalize the lesson in “To a Louse”: “Oh wad some power the giftie gie us / To see oursels as ithers see us!”) “A man resting from a fit of the stone or gout, is, for the time, in possession of feelings which undisturbed health cannot impart. They may be dearly bought, but still they are to be set against the price. And, indeed, it depends upon the duration and urgency of the pain, whether they be dearly bought by suffering a moderate interruption of bodily ease for a couple of hours out of the four and twenty” (pp. 523-533).

To complete the picture of joyous nature made by a loving God, signs of non-utility in sheer behavioral exuberance, particularly in the play of young creatures, testify to the sheer pleasure of being alive on such a wondrous planet:

Swarms of newborn flies are trying their pinions in the air. Their sportive motions, their wanton mazes, their gratuitous activity, their continual change of place without use or purpose, testify their joy, and the exulta­tion which they feel in their lately discovered faculties . . . Other species are running about with an alacrity in their motions which carries with it every mark of pleasure. Large patches of ground are sometimes half cov­ered with these brisk and sprightly natures. If we look to what the waters produce, shoals of the fry of fish frequent the marshes of rivers, of lakes, and of the sea itself. These are so happy, that they know not what to do with themselves (pp. 490-491).

(Paley's prose may be purple, but his purpose is sanguine. He argues, in stat­ing his primary case, that organic adaptation proves the personhood of God. But we want to know more. God could, after all, be a consummate crafts­man, but a crabby character. Paley's arguments on pain and natural happiness indicate that God is not only skillful, but also benevolent as well.)

These statements, taken out of context (as usually done), promote an un­fair caricature of a subtle argument. Paley cannot be dismissed as an intellec­tual slouch. His Evidences of Christianity (1794) remained a required text for entrance to Cambridge University until the 20th century, and Darwin would never have chosen a cardboard dogmatist for a hero or, later, for an opponent  {266}  worthy of overturning as the essential thrust of a revolutionary theory (see pp. 116125). Paley's totality presents a subtle, coherently reasoned brief for an adaptationist natural theology.

First of all, Paley cannot be caricatured as a Panglossian perfectionist. He states explicitly that we cannot use perfection as a criterion for identifying good design, or even as the necessary mark of divinity in craftsmanship: “It is not necessary that a machine be perfect, in order to show with what design it was made: still less necessary, where the only question is, whether it were made with any design at all” (p. 5).

Paley also provides, if only occasionally, positive arguments for imperfec­tion, as in feathers of an ostrich's wing. “The filaments hang loose and sepa­rate from one another, forming only a kind of down; which constitution of the feathers, however it may fit them for the flowing honors of a lady's head­dress, may be reckoned an imperfection in the bird, inasmuch as wings, com­posed of these feathers, although they may greatly assist in running, do not serve for flight” (p. 236). And he acknowledges that the creator's preference for utility lies revealed in the overwhelming relative frequency, not the ubiq­uity, of adaptation (but adding the conventional rider, still commonly ad­vanced today, that, if we look hard enough, we will discover uses for traits now judged “nonadaptive”). “Instances . . . where the part appears to be to­tally useless, I believe to be extremely rare: compared with the number of those, of which the use is evident, they are beneath any assignable propor­tions; and perhaps, have never been submitted to a trial and examination suf­ficiently accurate, long enough continued, or often enough repeated” (Paley, 1803, p. 64).

In fact, Paley uses adaptationism primarily as a theoretical argument about depth of causality, not as an excuse to rhapsodize about happy nature. Oppo­nents who wish to see “physical law” as the source of form might cite sexual generation and embryology as leading examples. But these processes only provide the immediate physical continuity of efficient causation: “The truth is, generation is not a principle but a process” (p. 453). We need a deeper rea­son, a true principle, for the evident adaptation of form to function — in short, a final cause. Even if watches gave birth to new watches, Paley argues, we would not identify ontogeny as the ultimate source of timekeeping. Neither can embryology be the cause of optical excellence in the human eye, if only because “things generated possess a clear relation to things not generated” (p. 455) — the eye to external light and to the objects we need to see in this case. (We now recognize this otherwise persuasive argument as wrong only because life, unbeknownst to Paley, possesses history and mutability.)

But the main case for taking Paley seriously lies in his formulation and refutation of opposing visions. Anyone can spin out a rationale for an idée fixe, but a well-crafted system requires both full analysis and principled denial of alternatives. Natural Theology merits our respect, and becomes a key docu­ment for this chapter on the history of functionalism vs. formalism, because Paley recognized the structuralist alternative and provided a coherent defense.  {267}  His arguments span two chapters (15 on relations and 16 on compen­sations), treating the phenomenon always viewed as crucial and primary by advocates of structural constraint — stable correlations among parts of the body.

Since Paley's main argument holds that intricate contrivance implies a contriver, two main rebuttals might be offered in principle: (1) the adaptations exist, but they originated by a natural process of evolution, not by creative acts of a deity; (2) organisms were created, but adaptation does not permeate or even dominate their form.

Since Paley never imagined the alternative of natural change by selection or weeding out, he confines his refutation of adaptive evolution to the “Lamarckian” principle of use and disuse. (I doubt that Paley, writing in 1802, knew Lamarck's work directly, since his French colleague had just begun to publish evolutionary views. But use and disuse, as an item of folk wisdom, frequently entered the arguments of evolutionists.) Paley begins em­pirically by pointing out that centuries of disuse do not cause organs to disap­pear, though modesty leads him to cloak a classic case entirely in untranslated Latin: “The mammae of the male have not vanished by inusitation; nee curtorum, per multa saecula, Judaeorum propagini deest praeputium” [nor has the foreskin of Jews become any shorter in offspring through many centu­ries of circumcision] (p. 446).

Paley then asks, more theoretically, how any natural evolution of useful structures could be attributed to a stimulus structurally unrelated to biologi­cal form, and often inorganic. The eye is a contrivance for perceiving light, but light cannot make an eye. “Yet the element of light and the organ of vi­sion, however related in their office and use, have no connection whatever in their original. The action of rays of light upon the surfaces of animals has no tendency to breed eyes in their heads. The sun might shine forever upon living bodies without the smallest approach towards producing the sense of sight” (p. 317).

When two structures have been similarly fashioned for a common purpose by a strengthening of one and a weakening of the other (the subject of Paley's “compensations” in chapter 16), natural adjustment by evolution might be defended (as when an elephant elongates its trunk to compensate a shortness of neck). But Paley denies this “best case” by the standard argument that in­termediary stages could not be well designed: “If it be suggested, that this proboscis may have been produced in a long course of generations, by the constant endeavor of the elephant to thrust out his nose, (which is the general hypothesis by which it has lately been attempted to account for the forms of animated nature), I would ask, how was the animal to subsist in the mean­time, during the process, until this elongation of snout were completed? What was to become of the individual, whilst the species was perfecting?” (p. 299).

If the first alternative (adaptation, but by evolution) can be thus refuted, how can the second possibility (creation, but with adaptation secondary or absent) be dismissed as well? Paley now meets the formalist alternative face-to-face —  {268}  and rejects this last challenge with three arguments that, taken to­gether, develop his strongest case for adaptationism (the first two remain in prominent use today):

1. Formalists do not deny the evident utility of most organic structures. The focus of their argument, rather, rests upon a claim for temporal and causal primacy (homology based upon historical order for evolutionists, or similarity based upon repeated themes in manufacture for creationists). Adaptationists hold that structures must evolve or be fashioned for utility: functional needs come first, and form follows. Formalists argue, on the other hand, that morphology may arise for reasons other than use, with later “up­take” of function as subsidiary: that is, form comes first, and organisms may then discover usages. In a remarkable passage, showing his grasp of this fun­damental alternative (now being reasserted as the basis for revival of interest in constraint among modern evolutionists), Paley admits that the formalist argument must be acknowledged as “intelligible”: “To the marks of contriv­ance discoverable in animal bodies, and to the argument deduced from them, in proof of design, and of a designing Creator, this turn is sometimes at­tempted to be given, viz. that the parts were not intended for the use, but that the use arose out of the parts. This distinction is intelligible. A cabinet-maker rubs his mahogany with fish-skin; yet it would be too much to assert that the skin of the dogfish was made rough and granulated on purpose for the polish­ing of wood, and the use of cabinet makers” (p. 72).

Paley's refutation invokes the classic response: the formalist argument will work for simple structures like fish-skin, but not for complex organs, com­posed of multiple parts, all apparently adjusted for current function. “Is it possible to believe that the eye was formed without any regard to vision; that it was the animal itself which found out, that, though formed with no such in­tention, it would serve to see with; and that the use of the eye, as an organ of sight, resulted from this discovery, and the animal's application of it?” (p. 73).

2. The first argument epitomizes a conceptual mainstay of formalism, but the empirical foundation of structuralist morphology has always depended more strongly upon correlation among parts of an organism, buttressed by the inference that structural relations, rather than utility, establish the link­age. Again Paley provides the classic functionalist refutation, still promi­nently in use. The correlations, he argues, do not arise by formal necessity, or “laws of growth,” but as coordinated adaptations, each separately useful and required for good design. Swans have long necks and webbed feet for reasons of common function, not “necessary connection”: “The long neck, without the web foot, would have been an incumbrance to the bird; yet there is no necessary connection between a long neck and a webfoot. In fact they do not usually go together. How happens it, therefore, that they meet, only when a particular design demands the aid of both?” (p. 293).

Paley then discusses a favorite example of British adaptationists since John Ray, and a pest in British gardens from time immemorial: the mole. “From soils of all kinds the little pioneer comes forth bright and clean. Inhabiting  {269}  dirt, it is, of all animals, the neatest” (p. 294). Paley defends adaptation with an explicit rejection of the strongest argument for constraint (what Darwin would later call “correlation of growth”). Recalling his opening metaphor, Paley writes: “Observe then, in this structure, that which we call relation. There is no natural connection between a small sunk eye and a shovel pal­mated foot. Palmated feet might have been joined with goggle eyes; or small eyes might have been joined with feet of any other form. What was it there­fore which brought them together in the mole? That which brought together the barrel, the chain, and the fusee, in a watch: design; and design, in both cases, inferred, from the relation which the parts bear to one another in the prosecution of a common purpose” (p. 296).

3. But what can an adaptationist say about the overarching homologies of broad taxonomic structure? Are these widespread properties not formal con­straints, logically prior to any subsequent utility forged by specific tinkering with such common elements? (We certainly acknowledge such priority today, but we also recognize Darwin's incisive argument that these “phyletic” con­straints may have arisen as ancestral adaptations — see last section. Paley en­joyed no conceptual access to this legitimate adaptationist exit from the di­lemma.)

In a clever twist of argument, Paley turns homology to the cause of adapta­tion in two steps:

(1) God devised general plans with foreknowledge of their requisite modification for specific purposes in individual species. For if these grand homologies had been generated automatically by abstract laws of nature, with no reference to final causality, how could such widespread structures be so subtly subject to such varied adaptation in the service of so many particu­lar modes of life? “Whenever we find a general plan pursued, yet with such variations in it, as are, in each case, required by the particular exigency of the subject to which it is applied, we possess, in such a plan and such adaptation, the strongest evidence, that can be afforded, of intelligence and design ... If the general plan proceeded from any fixed necessity in the nature of things, how could it accommodate itself to the various wants and uses which it had to serve, under different circumstances, and on different occasions?” (Paley, 1803, p. 227).

(2) Yet Paley recognized the potential circularity in this claim, if taken by itself. To be sure, once such homologies have been established, they may be examined for susceptibility to adaptive modification. But why did God proceed in this manner at all? Why didn't he just make each species from scratch, opti­mally suited for its own peculiar mode of life? Why bother with common plans at all, when creatures sharing the plans work so differently? Here, at the crux of his difficulty, Paley invokes a venerable solution that has always (both then and now) struck critics as at least slightly sophistic (in the sense that any potential refutation could be so “accommodated,” thus making the theory irrefutable, untestable, and therefore useless): God shows his greatness by limiting his own power with ordered principles (secondary causes based on natural laws) and structural designs (grand homologies):  {270} 

God, therefore, has been pleased to prescribe limits to his own power, and to work his ends within those limits. The general laws of matter have perhaps the nature of these limits . . . These are general laws; and when a particular purpose is to be effected, it is not by making a new law, nor by the suspension of the old ones, nor by making them wind and bend and yield to the occasion (for nature with great steadiness adheres to, and supports them), but it is, ... by the interposition of an apparatus corresponding with these laws, and suited to the exigency which results from them, that the purpose is at length attained. As we have said, there­fore, God prescribes limits to his power, that he may let in the exercise, and thereby exhibit demonstrations of his wisdom (p. 43).

After all, adaptationism only requires that organic designs be complex and work well, not that they embody perfection: “Contrivance, by its very defini­tion and nature, is the refuge of imperfection. To have recourse to expedients implies difficulty, impediment, restraint, defect of power” (pp. 41-42).

Paley's closing paean, following this last statement, exalts adaptation as logically necessary, quite apart from any factual validation. Contrivance not only sets the dominant pattern of empirical nature. Such good design also represents the only way that God could proclaim his existence in principle! To quote the passage of page 119 once again:

It is only by the display of contrivance, that the existence, the agency, the wisdom of the Deity, could be testified to his rational creatures. This is the scale by which we ascend to all the knowledge of our Creator, which we possess, so far as it depends upon the phenomena, or the works of na­ture. Take away this, and you take away from us every subject of obser­vation, and ground of reasoning . . . Whatever is done, God could have done, without the intervention of instruments or means: but it is in the construction of instruments, in the choice and adaptation of means, that a Creative Intelligence is seen. It is this, which constitutes the order and beauty of the universe (p. 42).

Paley's argument coheres, yet sounds a peculiarly limited range of notes — the reason for my “sand painting” metaphor of page 262. Paley does mention the grand homologies that underlie all taxonomy — but only in a paragraph or two, and only to offer an adaptationist riposte. He does formulate the structuralist argument based on correlation — but only in passing reference, and only for refutation. We might be tempted to offer the Philistine's retort — “oh well, Paley was just a philosopher; what did he know about real biol­ogy?” But modern disciplinary boundaries did not exist in 1800, and great bi­ologists, including Darwin, valued Paley above all other books in natural his­tory. Moreover, as I shall show in the next section, a fine working biologist like Agassiz could present the other side with equally uncompromising exclu­sivity.

We must therefore grasp Paley's restricted compass as a consciously-chosen vision of life's substance and meaning. As such, we may utilize, for our own  {271}  instruction, a position so unsullied by nature's real complexities. We know that life cannot work at such a conceptual extreme, but any consistent and well-argued defense of such an edge remains fascinating — at least in illustrat­ing a set of mental habits that still motivates scientists. Just as we learn to grasp nature through controlled and simplified cases (the experimental method), so may we also comprehend mind by its defense of coherence at the philosophical endpoint of a continuum.


Louis Agassiz, as the first permanent immigrant among great European bio­logical theorists, became the symbol and actuality of maturation and prestige for American natural history in the mid 19th century. Romantic mythology proclaims that he ventured forth as an intrepid pioneer in a quest for pristine knowledge and uncharted species. In fact, Agassiz's primary reasons for reset­tlement were far more mundane — escape from trouble and hope for a new be­ginning. He had suffered the two classic reversals of personal misfortune after years of intellectual triumph: bankruptcy (when his lithographic press, ini­tially established to print the plates for his Poissons fossiles, failed) and famil­ial strife (when his wife moved out after he had turned their home into a fac­tory and boarding house for workers at his press). In any case, whatever the complex motives, Agassiz's decision to settle permanently at Harvard estab­lished a happy incongruity within an expanding and accepting culture — a great francophone theorist, with traditional continental attitudes, living in Yankee Boston.

Agassiz (1807-1873) came to America with grand plans to invest his boundless energy in systematic work on undescribed native faunas, following his own maxim: “study nature, not books.” But, as a consummate academic politician and promoter, he became sidetracked over the years (an old story, as deep as human nature itself), and published little technical work during his last two decades. The frustration in this familiar tale of good intentions lies best exposed in Agassiz's grandest project and its failure.

Early in the 1850's, he announced plans for a lavish 10-volume work to be called The Natural History of the United States. He gathered more than 2,000 paid subscriptions in advance, and began collection (for an initial monograph on turtles) with his old and characteristic zeal. But he soon bogged down — permanently. Only four volumes ever appeared (with the de­scriptive and taxonomic work largely done by others), and he talked less and less about his grand design as the years ticked away. Nonetheless, while still imbued with initial enthusiasm, he wrote, as a book-length introduction to volume 1, his finest theoretical work, the Essay on Classification. Published in 1857, and revised in 1859 (ironically just 3 months before publication of Darwin's Origin, the book that would undermine the central premise of Agassiz's work), the Essay on Classification stands as a unique and incongru­ous document — a statement of natural theology in the highest tradition of  {272}  continental formalism, published in the most English of American cities. Agassiz never mentions Paley by name, but his volume presents an almost perfect counterpoise to Paley's Natural Theology from the other pole of the great dichotomy in approaches to form — particularly, in this case, to the question of how an omnipotent God would manifest his glory in nature.

Modern supporters of systematics, in a world increasingly dominated by trendier forms of biological research, often feel beleaguered, and therefore impelled to provide a wider rationale for pursuing classification, an enterprise unfairly burdened with such epithets as “stamp collecting” by a miscompre­hending public. Today, the rationale for systematics tends to be given — quite legitimately of course — in terms of our current crises in environmental deteri­oration and declining biodiversity. Yet if any systematist ever yearned for a maximally grand rationale for his chosen profession, he could not find, or even imagine, a more audacious document than Agassiz's Essay on Clas­sification. (Unfortunately, changing philosophies and increasing knowledge have rendered Agassiz's argument obsolete, but we may still sense, and should still admire, the style and grandeur of his claim.)

In baldest terms, and from a Platonic perspective (with organisms con­strued as temporary, material incarnations, representing the permanent and transcendent mental structures of an overarching creative force), Agassiz ar­gues that taxonomy should be regarded, in principle, as the highest of the sci­ences. For species embody ideas in God's mind; and actual organisms then became transient configurations that represent, or incarnate, these ideas. Re­lationships among species, as expressed in classification, therefore reveal the structure of God's thought, for if each species denotes a divine idea, then their interconnections in taxonomy display the order of God's mentality.

Agassiz poses the key question: “Are these divisions artificial or natural? Are they the devices of the human mind to classify and arrange our knowl­edge in such a manner as to bring it more readily within our grasp and facili­tate further investigations, or have they been instituted by the Divine Intelli­gence as the categories of his mode of thinking?” (1857, pp. 7-8). He then provides his firm answer: “To me it appears indisputable, that this order and arrangement of our studies are based upon the natural primitive relations of animal life, — those systems [of classification] . . . being in truth but transla­tions, into human language, of the thoughts of the Creator.”

With this vision, Agassiz cuts through an old argument about the differen­tial “reality” of categories in a Linnaean hierarchy: Are species real and higher levels artificial? Are all categories real or do they only express the prac­tical needs of human convenience? If, as Agassiz argues, the entire taxonomic system, when properly “discovered,” records the structure of God's thoughts, then all categories must be objective segments of this divine totality. Only or­ganisms have material existence, but taxonomic categories embody higher re­ality as direct expressions of the divine mind:

Is not this in itself evidence enough that genera, families, orders, classes, and types have the same foundation in nature as species, and that individuals  {273}  living at the same time have alone a material existence, they be­ing the bearers, not only of all these different categories of structure upon which the natural system of animals is founded, but also of all the relations which animals sustain to the surrounding world, — thus show­ing that species do not exist in nature in a different way from the higher groups, as is so generally believed? (1857, p. 7).

Agassiz shares Paley's primary goal, the fundamental “research program” of “natural theology” — to infer, from the organic works of nature, not only God's existence, but as much as possible about his intellect and goodness. Yet, despite this common aim, Paley and Agassiz could not have advocated more disparate constructions of divine presence in nature.

Every good debater, following the principle of dichotomy, knows that arguments fare best by contrast with alternatives. Moreover, the more caricatured and cardboard the alternative, the better for your side (so long as you don't depict your opponent as so much of a straw man that he becomes unbeliev­able). Agassiz presents his vision of classification by contrast with a “materi­alist” alternative of his own construction. He defines a materialist as a natu­ralist who attributes the forms and properties of organisms to the shaping power of constant physical laws (secondary, efficient causes), and not to di­rect decisions of divine will. A materialist may escape the charge of godless-ness by arguing for divine establishment of natural laws at the beginning of time. But if God then absconds forevermore, and lets nature work in such an automatic and heartless mode, what practical difference could we discern be­tween outright materialism and such a divine clock winder? “I allude here,” Agassiz writes (p. 9) in defining his opponents, “only to the doctrines of ma­terialists.” The issue reduces to a simple dichotomy (given the inconceiv­ability of other alternatives, including randomness): are taxa fashioned by laws of nature (and therefore in harmony with physical order), or by God as incarnations of His categories of thought? Agassiz states the contrast, and an­nounces his own allegiance: “Others believe that there exist laws in nature which were established by the Deity in the beginning, to the action of which the origin of organized beings may be ascribed; while according to others, they owe their existence to the immediate intervention of an Intelligent Cre­ator. It is the object of the following paragraphs to show that there are neither agents nor laws in nature known to physicists under the influence and by the action of which these beings could have originated” (1857, p. 13).

In a grand verbal flourish, Agassiz then upholds taxonomy as the highest science, while branding the materialist alternative both dreary and soul de­stroying (as well as wrong). Taxonomic order records divine mentality:

I confess that this question as to the nature and foundation of our scien­tific classifications appears to me to have the deepest importance, an im­portance far greater indeed than is usually attached to it. If it can be proved that man has not invented, but only traced the systematic ar­rangement in nature, that these relations and proportions which exist throughout the animal and vegetable world have an intellectual, and  {274}  ideal connection in the mind of the Creator, that this plan of creation, which so commends itself to our highest wisdom, has not grown out of the necessary action of physical laws, but was the free conception of the Almighty Intellect, matured in his thought, before it was manifested in tangible external forms, — if, in short, we can prove premeditation prior to the act of creation, we have done, once and forever, with the desolate theory which refers us to the laws of matter as accounting for all the wonders of the universe, and leaves us with no guard but the monoto­nous, unvarying action of physical forces, binding all things to their inev­itable destiny (1857, p. 9).

By setting up his argument in this manner, Agassiz immerses himself di­rectly into the formalist-functionalist debate — with his own version of natu­ral theology as a strictly, almost excessively, formalist proposal: taxonomic order at all levels, not the behavior and function of individual creatures, re­cords God's nature and intent. But by characterizing (or caricaturing) his op­position as a claim for the direct production of form by physical forces, he places the chief category of putative evidence against his vision — correlation between morphology and physical conditions of life — into the functionalist camp. (One might object in principle that such a functionalist conclusion need not follow from Agassiz's version of “materialism.” After all, morphol­ogy might be fashioned by laws of nature, but without functional excellence. Still, Agassiz's chosen definition should not be dismissed as self-serving be­cause theorists who have espoused direct production of form by physical laws — D'Arcy Thompson (1917, 1942) in particular (see pp. 11791208) — have indeed used mechanical optimality as the criterion for their claim).

Thus, Agassiz commits himself to a “two-fisted” argument within the formalist-functionalist dichotomy: to demonstrate that taxonomic structure is a product of divine thought, he must show that classification records an an­atomical order independent of external conditions of life (the positive argu­ment for formalism), and also that a fit of form to immediate function cannot represent the generating principle of organic order (the negative argument against functionalism).

Agassiz, of course, does not deny that organisms tend to be well adapted; no formalist has ever made so strong a claim against the Paleyan alternative. He argues, rather — as formalists have done throughout history, no less so to­day than in Agassiz's time — that adaptation only expresses a secondary tin­kering and minor adjustment of prior and fundamental Bauplan built by formalist principles. In its strongest version, Agassiz's brand of formalism la­bels adaptation as a delusion because good fit only confuses our search for a deeper order by imposing a superficial overlay of specific and immediate ad­aptation upon a Bauplan, thereby obscuring the more important underlying structure.

Agassiz's chief positive argument rests upon his unswerving allegiance to Cuvier's establishment of four anatomical ground plans as the foci of animal  {275}  design: Radiata, Mollusca, Articulata, and Vertebrata.* Agassiz was particu­larly impressed that von Baer, the century's greatest embryologist, had, inde­pendently of Cuvier, recognized the same system by developmental standards. If morphology and embryology coincided so well, and if the greatest students of both subjects had reached agreement by such different criteria, then the fundamental principle of natural order must lie revealed:

If we remember how completely independent the investigations of K. E. von Baer were from those of Cuvier, how different the point of view was from which they treated their subject, the one considering chiefly the mode of development of animals, while the other looked mainly to their structure; if we further consider how closely the general results at which they have arrived agree throughout, it is impossible not to be deeply im­pressed with confidence in the opinion they both advocate, that the animal kingdom exhibits four primary divisions, the representatives of which are organized upon four different plans of structure, and grow up according to four different modes of development (1857, p. 231).

But how shall taxonomists characterize the basis for this primary division into four? As God's mind lies so far beyond our poor faculties, we cannot identify his intent (though we can certainly record his decisions); but we may surely specify the criteria that he did not use. Much of Agassiz's Essay fea­tures a litany of claims in this negative mode: as only two alternatives exist, any argument against production of form by physical laws (a mode of origin that would induce a functional correlation of morphology and environment on the broadest scale) must provide support for the organization of relation­ships as categories of divine will and thought. After an introductory chapter, for example, the first two sections of Agassiz's Essay present a contrast with a common intent. How can physical laws simply produce the “best” solution for each particular circumstance if (1) identical environments house creatures of all four great body plans, and if (2) each of the body plans manages to in­habit all major environments? Agassiz summarizes: “The simultaneous existence  {276}  of the most diversified types under identical circumstances exhibits thought, the ability to adapt a great variety of structures to the most uniform conditions. The repetition of similar types, under the most diversified circum­stances, shows an immaterial connection between them; it exhibits thought, proving directly how completely the Creative Mind is independent of the in­fluence of a material world” (p. 132).

Agassiz claimed even stronger support from the geological record. Environmental change exhibits no directional pattern through time, but life's history features progressive change (via successive creations, not by evolution) within each of the four immutable types. How could unaltered physical laws and nondirectional physical change fashion a progressive history of life?

Who could, in the presence of such facts, assume any causal connection between two series of phenomena, the one of which is ever obeying the same laws, while the other presents at every successive period new relations, an ever changing gradation of new combinations, leading to a final climax with the appearance of Man? Who does not see, on the contrary, that this identity of the products of physical agents in all ages, totally dis­proves any influence on their part in the production of these ever-changing beings, which constitute the organic world, and which exhibit, as a whole, such striking evidence of connected thoughts! (p. 101).

I do not claim that the refutation of Paleyan natural theology motivated this line of argument. As his major aim, Agassiz tried to debunk his carica­tured version of “materialism” by showing that organisms cannot be directly constructed by physical laws. Agassiz advances his argument primarily by in­voking numerous variations on the same theme: organisms do not “match” the physical world in the way that ice forms as the predictable and appropri­ate state for water at certain temperatures and pressures; thus, we see “how completely the Creative Mind is independent of the influence of a material world” (p. 132 as quoted above).

Agassiz begins his explicit attack on functionalism by acknowledging Paley's style of natural theology as the more common argument for God's ex­istence and benevolence (Agassiz cites the Bridgewater Treatises, the primary Paleyan documents of his generation), but then holding that adaptation can­not represent God's primary mark upon natural history for two reasons: (1) Good correlation of function to environment would not illustrate God's care in any case, for such a relation may only record the production of form by physical causes. (2) Adaptationism fails as a generality because too many constraints, imposed by unity of type, limit any organic approach to optimality:

The argument for the existence of an Intelligent Creator is generally drawn from the adaptation of means to ends, upon which the Bridge-water Treatises, for example, have been based. But . . . beyond certain limits, it is not even true. We find organs without functions, as, for in­stance, the teeth of the whale, which never cut through the gum, the  {277}  breast in all males of the class of Mammalia; these and similar organs are preserved in obedience to a certain uniformity of fundamental structure, true to the original formula of that division of animal life, even when not essential to its mode of existence. The organ remains, not for the performance of a function, but with reference to a plan (pp. 9-10).

Adaptation exists, of course, but only as a superficial and secondary over­lay upon unity of type — the deeper and true reflection of God's majestic or­der: “When naturalists have investigated the influence of physical causes upon living beings, they constantly overlooked the fact that the features which are thus modified are only of secondary importance in the life of ani­mals and plants, and that neither the plan of their structure, nor the various complications of that structure, are ever affected by such influences” (p. 17).

Most importantly, this deeper unity of type not only represents a natural principle in dichotomous opposition to adaptation, but also proves that cre­ative thought, not mere mapping upon physical conditions, establishes or­ganic order.

In all these animals and plants, there is one side of their organization which has an immediate reference to the elements in which they live, and another which has no such connection, and yet it is precisely this part of the structure of animals and plants, which has no direct bearing upon the conditions in which they are placed in nature, which constitutes their essential, their typical character. This proves beyond the possibility of an objection, that the elements in which animals and plants live . . . cannot in any way be considered as the cause of their existence (p. 33).

Having cleared away the notion that something so trivial as adaptation might represent God's signature in nature, Agassiz can now complete his ulti­mate defense of taxonomy as the custodian of God's presence in nature, as manifested in the broad relationships sanctioned by unity of type. Consider how much we may know of God's nature — a veritable volley of adjectives — once we locate his correct signature at the appropriate pole of nature's great dichotomy:

The products of what are commonly called physical agents are every­where the same, (that is, upon the whole surface of the globe) and have always been the same (that is, during all geological periods); while orga­nized beings are everywhere different and have differed in all ages. Be­tween two such theories of phenomena there can be no causal or genetic connection. The combination of space and time of all these thought­ful conceptions exhibits not only thought, it shows also premeditation, power, wisdom, greatness, prescience, omniscience, providence. In one word, all these facts in their natural connection proclaim aloud the One God, whom man may know, adore, and love; and Natural History must, in good time, become the analysis of the thoughts of the Creator of the Universe, as manifested in the animal and vegetable kingdoms (p. 135).  {278} 

Moreover, in understanding taxonomy as an incarnation of divine thought, we also sense our own importance in the cosmos. For if our taxonomy can mirror God's order so well, then our minds must also resemble His in princi­ple, however infinitely poorer in capacity:

Do we not find in this adaptability of the human intellect to the fact of creation, by which we become instinctively, and, as I have said, unconsciously, the translators of the thoughts of God, the most conclusive proof of our affinity with the Divine Mind? And is not this intellectual and spiritual connection with the Almighty worthy of our deep consider­ation? If there is any truth in the belief that man is made in the image of God, it is surely not amiss for the philosopher to endeavor, by the study of his own mental operation, to approximate the workings of the Divine Reason, learning, from the nature of his own mind, better to understand the Infinite Intellect from which it is derived. Such a suggestion may, at first sight, appear irreverent. But, which is the truly humble? He who, penetrating into the secrets of creation, arranges them under a formula which he proudly calls his scientific system? Or he who, in the same pur­suit, recognizes his glorious affinity with the Creator, and, in deepest gratitude for so sublime a birth right, strives to be the faithful interpreter of that Divine Intellect with whom he is permitted, nay, with whom he is intended, according to the laws of his being, to enter into communion? (p. 8).

With so much at stake, from the basis of natural order to confidence in our mental affinity with God himself, the primacy of broad taxonomic formal­ism over local adaptationism (however exquisite) becomes an issue of highest moment and passion. Darwin's added dimension of history would derail Agassiz's grand design just three months after the Essay received its definitive printing, but we should remember Agassiz's effort, and grasp his argument, as perhaps the noblest brief ever presented for the centrality of systematics among the sciences.


While acknowledging some historical interest in the contrast, modern evolu­tionists might question, on two grounds of supposed irrelevance to current is­sues, the time I have taken to contrast Paley with Agassiz: (1) Paley and Agassiz struggled to find the proper signature of God in nature, and such an effort no longer counts as part of science; (2) Darwin added a third, historical dimension, thereby fracturing the old dichotomy of form and function, and rendering its terms obsolete.

I would argue, in response, that Darwin's addition, though surely the most important and revolutionary event in the history of biology scarcely rendered the old dichotomy irrelevant (see pp. 251260 for a fuller development of this point). As Figure 4-3 shows, any morphology attributed to Darwin's histori­cal dimension must still, by recursion, be judged by the dichotomy at its time  {279}  of origin — that is, we must still know whether an ancestral form arose by adaptation or constraint (or by what mixture of the two poles). Thus, we may say that Darwin's new dimension expanded the scope of the dichotomy by compelling its application to two domains — past and present — when we ana­lyze the basis of any trait in a living organism.

Evolution does not establish an ultimate divide for all transitions in the his­tory of biology. Several themes pass right through this great revision, only al­tering their terms and explanations. Formalism vs. functionalism may be the most prominent and persistent of issues too grand even for evolution to undo (or fully resolve). Paley and Agassiz once fought this battle in grand style; Dawkins and Goodwin cannot cast so broad a conceptual net, or muster the same stylistic panache today, but they pursue the same conflict. Paley vs. Agassiz remains relevant to modern evolutionists by the primary criterion of genealogical continuity.

If Paley and Agassiz represent the yin and yang of totality for the analysis of form, then Darwin, though a pluralist who understood both poles, did ulti­mately cast his lot with the Paleyan yin, in filial piety with a British tradition that has spanned centuries, and still continues today. This imbalance, and the struggle for redress that now commands so much discussion in contemporary evolutionary biology, defines one of the three major issues that led me to write this book. The formalist alternative, as embodied in the subject now generally called “constraint,” provides a counterweight to stabilize the sec­ond leg in Darwin's essential tripod of support — the primacy of adaptation in asserting the creativity of natural selection at overwhelming relative fre­quency among the causes of evolutionary change.

The past holds sufficient interest and capacity for illumination all by itself, and no justification in terms of present enlightenment need ever be given. Still, as a practicing scientist, I do favor the use of history as a current guide — while I struggle not to wrench the meaning and motivation of arguments from the primary matrix of their own time. I don't know how else to proceed when tides of history overwhelm a worthy subject for little reason beyond the vagaries of fashion and contingency. Scientists too often become convinced that inexorable logic or irrefutable data have closed a subject forever. Even worse, given our propensity for historical ignorance, we often collectively forget that an alternative ever existed at all. In such cases, I know no better tactic for reopening an important subject than the record of history — the proof that brilliant scientists (so worthy of our admiration that we cannot be­little their concerns) devoted their concentrated attention to an issue that never achieved true settlement, but only veered towards transient “resolu­tion” by sociological complexities of shifting preferences, rather than logic of proof or exigencies of data. I believe that structuralist and formalist ap­proaches to anatomy fell out of favor for such invalid reasons of fashion, and that the full range of this primary dichotomy must now be reestablished. And I unabashedly call upon the great formalists of history to state their case; while I ask modern evolutionists to make the proper translation to mod­ern terms.  {280} 

To reassert the importance of both poles in this dichotomy, I again cite my primary candidate for the unenviable title of most worthy “invisible man” — an important and influential thinker and educator in his day, but now entirely forgotten. I tried to resurrect the Reverend James McCosh, president of Princeton University, in Chapter 2 (pp. 116-118), and I now again want to call upon his fine book, published in 1869 in collaboration with George Dickie: Typical Forms and Special Ends in Creation. The Greek inscription on the title page — typos kai telos (type and purpose) — epitomizes the argu­ment. The two poles of the dichotomy inhere in all natural objects, and full explanation demands attention to both:

In taking an enlarged view of the constitution of the material universe, so far as it falls under our notice, it may be discovered that attention, at once extensive and minute, is paid to two great principles or methods of procedure. The one is the Principle of Order, or a general plan, pattern, or type, to which every given object is made to conform with more or less precision. The other is the Principle of Special Adaptation, or partic­ular end, by which each object, while constructed after a general model, is, at the same time, accommodated to the situation which it has to oc­cupy, and a purpose which it is intended to serve. These two principles . . . meet in the structure of every plant and every animal (McCosh and Dickie, 1869, p. 1).

McCosh also recognized the contingent and socially embedded nature of national preferences. He notes that English tradition — from Robert Boyle and John Ray through Paley to the Bridgewater Treatises — has favored the adaptationist theme. Thus, he argues, recent discoveries in formalist mor­phology have been viewed as threatening by some biologists (McCosh cites the French and German schools of ideal or transcendental morphology, espe­cially in their English translation through the work of Richard Owen, whom I treat later in this chapter): “The arguments and illustrations adduced by Brit­ish writers for the last age or two in behalf of divine existence, have been taken almost exclusively from the indications in nature of special adaptation of parts. Hence, when traces were discovered in the last age of a general pat­tern, which had no reference to the comfort of the animal or the functions of the particular plant, the discovery was represented by some as overturning the whole doctrine of final cause; not a few viewed the new doctrine with sus­picion or alarm” (McCosh and Dickie, 1869, pp. 6-7).

But McCosh regards this perceived threat as false, and urges that formalist insights be welcomed — for full explanation demands attention to both poles. McCosh expresses the two key ideas in religious terms as natural illustrations of “lofty wisdom” (formalism) and “providential care” (functionalism). We call the same themes constraint and adaptation, but the image of exquisite balance remains every bit as valid today:

We do not know whether to admire most the all-pervading order which runs through the whole of nature, through all the parts of the plant and  {281}  animal, and through the hundreds of thousands of different species of plants and animals, or the skillful accommodation of every part, and of every organ, in every species, to the purpose which it is meant to serve. The one leads us to discover the lofty wisdom which planned all things from the beginning, and the enlarged beneficence reaching over all with­out respect of persons; whereas the other impresses us more with the providential care and special beneficence which, in attending to the whole, has not overlooked any part, but has made provision for every in­dividual member of the myriads of animated beings (p. 439).

Though mercilessly savaged for intellectual mediocrity by W. S. Gilbert and other satirists and activists, the British peerage did turn out an occasional scholar or two. The Duke of Argyll might have won his title fair and square if Gilbert's ultimate recommendation had ever been instituted. (The Fairy Queen in Iolanthe, royally pissed off at a group of nobles, threatens: “peers shall teem in Christendom, and a Duke's exalted station be attainable by competitive examination!”) In a presidential address to the British Associa­tion, the good Duke, as a prominent critic of evolution and author of several books still worth reading today, argued that relations between both necessary poles of the dichotomy still persisted as a key issue in Darwin's new biology: “What is the meaning of that great law of adherence to type and pattern, standing behind, as it were, and in reserve, of that other law by which organic structures are specially adapted to special modes of life? What is the relation between these two laws; and can any light be cast upon it derived from the history of extinct forms; or from the conditions to which we find that existing forms are subjected?” (quoted in McCosh and Dickie, 1869, p. 68).

Since then, countless events, from meanderings of history to permanencies of empirical discovery, have rocked this subject back and forth. But equi­librium at a center of dynamic tension, not of complacent rest, may foster our best biological understanding, and the Duke's question could not be more current, more a propos.

Unity of Plan as the Strongest Version of Formalism:
The Pre-Darwinian Debate


A prevalent myth of our time proclaims that broad and interdisciplinary vi­sions, though held in disrepute today, were once valued in a more ecumenical age that celebrated the “Renaissance man.” But the motto that “a cobbler should stick to his last”* dates from the 4th century bc, and people who wander  {282}  outside their primary field have always attracted suspicion or ridicule. In 1831, near the end of a long life, a poet who had ventured into science de­plored his failure to obtain a fair hearing, but defended his forays as inter­nally necessary for a broad and searching intellect:

The public was taken aback, for inasmuch as it wishes to be served well and uniformly, it demands that every man remain in his own field. This demand is well grounded, for a man who wishes to achieve excellence, which is infinite in its scope, ought not to venture on the very paths that God and nature do. For this reason it is expected that a person who has distinguished himself in one field, whose manner and style are generally recognized and esteemed, will not leave his field, much less venture into one entirely unrelated. Should an individual attempt this, no gratitude is shown him; indeed, even when he does his task well, he is given no spe­cial praise. But a man of lively intellect feels that he exists not for the public's sake, but for his own. He does not care to tire himself out and wear himself down by doing the same thing over and over again. More­over, every energetic man of talent has something universal in him, caus­ing him to cast about here and there and to select his field of activity ac­cording to his own desire (1831 essay, in Mueller and Engard, 1952, p. 16.9).

We might ignore this statement, if its author stood among the many hope­fuls whom history fails to memorialize in either their chosen or their adopted professions. But the writer cited above, J. W. von Goethe, wrote a thing or two of enduring merit! Moreover, and in retrospect, his ventures into science far transcended the brief forays of an amateur dabbler.

In any case, Goethe did not suffer complete neglect from scientists during his lifetime. In 1831, the great anatomist Etienne Geoffroy Saint-Hilaire praised Goethe's science as the work of “a poet trying to sing the grandeur of the universe in another form” (“un poete s'essayant de chanter sous une autre forme les grandeurs de l'univers” — 1831, p. 189). Geoffroy continued (1831, p. 193): “If Goethe had not already amassed enough titles to be proclaimed the greatest genius of his century, he would have added, to his crown of great poet and profound moralist, the fame of a wise naturalist — due to him for the profundity of his views, and for the philosophical force of his opinions on the subject of botanical analogies.”*

But Geoffroy's praise (see Fig. 4-8) cannot be reckoned as entirely disinterested, for Goethe had just favored his side in the greatest brouhaha of early 19th century zoology — the celebrated 1830 debate with Cuvier before the  {283}  Academie des sciences. Geoffroy needed all the help he could get (and he would later recruit other literary figures, including the novelists Balzac and George Sand, to his cause as well). Cuvier, after all, was no ordinary oppo­nent, and the subject of their argument — the age-old dispute of formalism and functionalism — could not have been more central to natural history.

Geoffroy, with good cause, viewed Goethe as the doyen and spiritual leader of formalist morphology. Not only had Goethe coined the word “morphol­ogy,” but he had, long before, defended for plants the central proposition that Geoffroy championed for animals as the starting point for his anatomical views — the reduction of form to a single generating archetype (the leaf for Goethe, the vertebra for Geoffroy). While the young Geoffroy worried about establishing a career and surviving a revolution, Goethe was traveling in It­aly and developing the theory of his 1790 work, Versuch die Metamorphose der Pflanzen zu erkldren (Fig. 4-9). (This work, little more than a pamphlet, consists of 123 numbered and almost aphoristic paragraphs. I shall quote by number from the standard translation of Mueller and Engard (1952). But I have read and own a copy of the original, which I highly recommend to any­one who appreciates the fusion of great writing and fascinating science.)

Goethe had been strongly interested in morphology throughout his life, and his preferences had always tended towards formalism, particularly to­wards the strongest version of the argument (and subject of this section) — the vision of a single, generating archetypal form, setting both the bounds and the possibilities of realized morphology. His two most famous forays into ani­mal anatomy both rested upon a formalist foundation: (1) his early support for the vertebral theory of the skull, a conviction that he traced to 1791 when he examined “a battered sheep skull from the sand of the dune like Jewish

4-8. A letter from Etienne Geoffroy Saint-Hilaire, written in January 1831 to a Mr. Payaud, with Goethe featured in the penultimate paragraph. (Geoffroy pre­sumably enclosed some writings of Goethe along with this letter to Payaud.) The text of the paragraph reads as follows: “Goethe est ci-joint. Sa bonhomie qui n'exclue pas la force et la justesse d'esprit, vous frappera.” (Goethe is enclosed. His good nature, which does not preclude strength and fairness of spirit, will strike you.) (Author's collection.)


4-9. Title page of Goethe's 1790 pamphlet on the growth and archetypal expressions of plants. (Author's collection.)

cemetery in Venice” (1823 essay, in Mueller and Engard, 1952, p. 237); and (2) his discovery of the human premaxillary bone in 1784, based on its pres­ence in other mammals and his convictions about unity of type. (Goethe called this bone the intermaxilare; others referred to it as “Goethe's bone.” In an essay written in 1832, the year of his death, Goethe recalled this discov­ery as “the first battle and the first triumph of my youth” (Goethe, 1832, p. 573).)

But Goethe chose botany for his most extensive study in formalism, and probably his finest contribution to science. In this important work, Goethe applied to plants the same vision that Geoffroy and Owen would later ad­vance in trying to reduce the great complexity and diversity of animal (or at least vertebrate) form to the single generating pattern of an archetypal verte­bra (see Geoffroy, 1831, for a homage to Goethe). For Goethe, the leaf rep­resented an archetypal form for all plant parts growing from the central  {285}  stem* — from cotyledons, to stem leaves, to sepals, petals, pistils and stamens, and fruit.

The common epitome of Goethe's system — all is leaf — should not be taken literally as the actual reduction of all serial diversity to the actual form of a stem leaf. Such a reading would contravene the Platonic character of arche­types in formalist theory. The “leaf” represents an abstract generating princi­ple, from which stem leaves depart least in actual expression. Goethe writes: “We ought to have a general term with which to designate this diversely metamorphosed organ and with which to compare all manifestations of its form . ... We might equally well say that a stamen is a contracted petal, as that a petal is a stamen in a state of expansion; or that a sepal is a contracted stem leaf approaching a certain stage of refinement, as that a stem leaf is a se­pal expanded by the influx of cruder saps” (1790, No. 120).

Goethe expressed the epitome of his system in measured tone within his es­say (1790, No. 119): “The organs of the vegetating and flowering plant, though seemingly dissimilar, all originate from a single organ, namely, the leaf.” In private, he became more effusive: “[I have traced] the manifold spe­cific phenomena in the magnificent garden of the universe back to one simple general principle” (1831 essay, in Mueller and Engard, 1952, p. 168). To friends, as to the philosopher J. G. Herder, he could become positively effu­sive (dare I say florid): “The archetypal plant as I see it will be the most won­derful creation in the whole world, and nature herself will envy me for it. With this model and the key to it, one will be able to invent plants..., which, even if they do not actually exist, nevertheless might exist, and which are not merely picturesque or poetic visions and illusions, but have inner truth and logic. The same law will permit itself to be applied to everything that is liv­ing” (letter of 1787, quoted in Mueller and Engard, 1952, p. 14).

Goethe dissects and compares, trying to find the leaf-like basis of appar­ently diversified and disparate structures. The anastomosed sepals, forming the calyx at the base of a flower, must be leaves that fail to separate when a cutoff of nutriment stops expansion of the stem: “If the flowering were re­tarded by the infiltration of superfluous nutriment, the leaves would be sepa­rated and would assume their original shape. Thus, in the calyx, nature forms  {286}  no new organ but merely combines and modifies organs already known to us” (1790, no. 38).

When parts become too distinct to show connection and reduction to the leaf archetype in one species, Goethe uses the comparative approach to find transitional forms in other taxa. The seedpod and sexual organs are mani­festly unleaflike in many plants, but Goethe establishes transitional series to species with, for example, leaflike seedpods, or fertile stem leaves (as in ferns). Consider his exposition of the comparative method for “difficult” seedpods: “Nature obscures the similarity to the leaf most when she makes the seed containers soft and juicy or firm or woody; however, the similarity will not escape our attention if we contrive to follow it in all its transitional stages” (1790, No. 79). Or for the even more divergent cotyledons that even­tually grow into tolerably leaflike form:

12. They are often misshapen, crammed, as it were, with crude matter, and as much expanded in thickness as in breadth; their vessels are unrec­ognizable and scarcely distinguishable from the mass as a whole. They bear almost no resemblance to a leaf, and we might be misled into re­garding them as special organs.

13. Yet in many plants the cotyledons approach leaf form: they flatten out; exposed to light and air, they assume a deeper shade of green; their vessels become distinct and begin to resemble veins.

14. Finally they appear before us as true leaves: their vessels are capa­ble of the finest development; their similarity to the subsequent leaves will not permit us to consider them separate organs; and we recognize them instead as the first leaves of the stem (1790, Nos. 12-14).

If Goethe's system really advocated, as often misportrayed, a simple and exclusive concept of the archetypal leaf, his theory could stake no claim for interesting completeness — for this central principle cannot explain system­atic variation in form up the stem, and therefore could not operate as a full explanation for both similarities and characteristic differences in the parts of plants. But, in his most fascinating intellectual move, Goethe proposes a com­plete account by grafting two additional principles onto the underlying no­tion of archetype: the progressive refinement of sap, and cycles of expansion and contraction. We may regard these principles as ad hoc or incorrect today, but the power of their conjunction with the archetypal idea can still be appre­ciated with much profit.

These two additional principles embody both necessary sides of the pri­mary Western metaphor for intelligibility in any growing, or historically ad­vancing, system — arrows of direction and cycles of repeatability (I called these conjoined principles “time's arrow” and “time's cycle” in my book on the discovery of geological time — Gould, 1987b). We must, in any temporal process, be able to identify both sources of story and order: vectors of change (lest time have no history, defined as distinctness of moments), and underly­ing constant or cyclical principles (lest the temporal sequence proceed only as one uniqueness after another, leaving nothing general to identify at all). Goethe,  {287}  faced with observations of both directionality and repeatability up the stem, recognized the need for both poles of this dichotomy.

Refinement of sap as a directional principle. Up and down, heaven and hell, brain and psyche vs. bowels and excrement, tuberculosis as a noble disease of airy lungs vs. cancer as the unspeakable malady of nether parts (see Susan Sonntag's Illness as Metaphor for a brilliant analysis of these conventional images). Almost irresistibly, we apply this major metaphorical apparatus of Western culture to plants as well — with gnarly roots and tubers as lowly objects of the ground, and fragrant, noble flowers as topmost parts, straining towards heaven. Goethe, by no means immune to such thinking in an age of Naturphilosophie, viewed the growth of a plant as progressing to­wards refinement from cotyledon to flower. He explained this directionality by postulating that, moving up the stem, each successive leaf modification progressively filters an initially crude sap. Inflorescence cannot occur until these impurities have been removed. The cotyledons begin both with minimal organization and refinement, and with maximal crudity of sap: “We have found that the cotyledons, which are produced in the enclosed seed coat and are filled to the brim, as it were, with a very crude sap, are scarcely organized and developed at all, or at best roughly so” (1790, No. 24).

The plant then grows towards a floral apotheosis, but too much nutriment delays the process of filtering sap — as material rushes in and more stem leaves must be produced for drainage. A decline in nutriment finally allows filtering to attain the upper hand, and the sap becomes sufficiently pure for inflor­escence: “As long as cruder saps remain in the plant, all possible plant organs are compelled to become instruments for draining them off. If excessive nutri­ment forces its way in, the draining operation must be repeated again and again, rendering inflorescence almost impossible. If the plant is deprived of nourishment, this operation of nature is facilitated” (1790, No. 30). Finally, the plant achieves its topmost goal: “While the cruder fluids are in this man­ner continually drained off and replaced by pure ones, the plant, step by step, achieves the status prescribed by nature, We see the leaves finally reach their fullest expansion and elaboration, and soon thereafter we become aware of a new aspect, apprising us that the epoch we have been studying has drawn to a close and that a second is approaching — the epoch of the flower” (1790, No. 28).

Cycles of expansion and contraction. If the directional force worked alone, then a plant's morphology would only express this smooth continuum of progressive refinement up the stem. Since, manifestly, plants do not display such a pattern, some other force must be operating.* Goethe describes  {288}  this second force as cyclical, in opposition to the directional principle of refining sap. He envisages three full cycles of contraction and expansion during ontogeny. The interplay of these progressive and cyclical forces pro­duces the full pattern of a general refinement up the stem, but impacted by discontinuities and transitions that express no directional pattern (“contrac­tion” of stem-leaves to sepals by bunching together in a circlet, for example). The cotyledons begin in a retracted state. The main leaves, and their substan­tial spacing on the stem, represent the first expansion. The bunching of leaves to form the sepals at the base of the flower marks the second contraction, and the subsequent elaboration of petals the second expansion. The reduction of archetypal leaf size to form pistils and stamens marks the third contraction, and the formation of fruit the last and most exuberant expansion. The con­tracted seed within the fruit then starts the cycle again in the next generation. Put these three formative principles together — the archetypal leaf, progressive refinement of sap up the stem, and three expansion-contraction cycles of veg­etation, blossoming, and bearing fruit — and the vast botanical diversity of our planet falls under the chief vision of formalism: production of realized va­riety from interaction of a few abstract, general, and internally based (not ex­ternally imposed and adaptationally driven) morphological laws: “Whether the plant vegetates, blossoms, or bears fruit, it nevertheless is always the same organs with varying functions and with frequent changes in form, that fulfill the dictates of nature. The same organ which expanded on the stem as a leaf and assumed a highly diverse form, will contract in the calyx, expand again in the petal, contract in the reproductive organs, and expand for the last time as fruit” (1790, No. 115).

This formalist commitment implies an aversion to primary explanation by adaptation, function or final cause. In accord with all the great formalists, Goethe often expressed his dislike of explanations based upon the externality of fit between form and function (though he delighted in the evident fact of such fit, as formalists also generally do, for such an admission poses no threat to the chief formalist argument for primacy of morphological order — see Chapter 11 on exaptation).

Goethe's statements on final cause often attack the larger idea of manufac­ture for explicitly human ends — not the chief complaint of formalist mor­phology, but worth recording, if only for the power of Goethe's prose:

For several centuries down to the present, we have been retarded in our philosophic views of natural phenomena by the idea that living organ­isms are created and shaped to certain ends by a teleological life force. ... Why should he not call a plant a weed, when from his point of view it really ought not to exist: He will much more readily attribute the exis­tence of thistles hampering his work in the field to the curse of an en­raged benevolent spirit, or to the malice of a sinister one, than simply re­gard them as children of the universal nature, cherished as much by her as the wheat he carefully cultivates and values so highly (from essay of 1790, in Mueller and Engard, 1952).  {289} 

But Goethe also attacked adaptationist primacy in the more focused realm of explaining morphology: “It is not a question of whether the concept of final causes is convenient, or even indispensable, to some people, or whether it may not have good and useful results when applied to the moral realm; rather, it is a question of whether it is an aid or a deterrent to physiologists in their study of organized bodies. I make bold to assert that it does deter them, therefore avoided it myself and considered it my duty to warn others against it” (2nd essay on plant metamorphosis, written in 1790, in Mueller and Engard, 1952, p. 80).

Citing a perennial complaint, then and now, against adaptationist explanations — that such efforts tell good stories in the speculationist mode, but do not explain morphology — Goethe compares final causes with Linnaeus' fan­ciful descriptions of sexual anatomy in plants: “For example, Linne calls flower petals 'curtains of the nuptial bed,' a parable that would do honor to a poet. But after all! the discovery of the true physiological nature of such parts is completely blocked in this way, just as it is by the convenient and false es­pousal of the theory of final causes” (2nd essay on plant metamorphosis, written in 1790, in Mueller and Engard, 1952, pp. 79-80).

Proper morphological explanations, Goethe asserts, must be sought on internal and formalist principles; external fit, though of great importance, can only be regarded as secondary: “In my opinion, the chief concept underlying all observation of life — one from which we must not deviate — is that a crea­ture is self-sufficient, that its parts are inevitably interrelated, and that noth­ing mechanical, as it were, is built up or produced from without, although it is true that the parts affect their environment and are in turn affected by it” (2nd essay on plant metamorphosis, written in 1790, in Mueller and Engard, 1952, p. 80).

In a remarkable passage, that could serve as a credo for modern formalism as well, Goethe asserts his central claim for internalist primacy, while also specifying the vital, but secondary, role of adaptation. Internal formation acts as a primary source that “must find external conditions.” Adaptation may then shape a range of diversity from an underlying form, but the arche­typal pattern cannot be explained by these secondary modifications, and the adaptations themselves can only express a superficial restructuring of inher­ent order:

Man, in considering all things with reference to himself, is obliged to assume that external forms are determined from within, and this assump­tion is all the easier for him in that no single living thing is conceivable without complete organization. Internally, this complete organization is clearly defined; thus it must find external conditions that are just as clear and definite, for its external existence is possible only under certain con­ditions and in certain situations.... An animal possesses external useful­ness precisely because it has been shaped from without as well as from within, and — more important and quite natural — because the external element can more readily adapt the external form to its own purposes  {290}  than it can reshape the internal form. We can best see this in a species of seal whose exterior has taken on a great deal of the fish character while its skeleton still represents the perfect quadruped (2nd essay on plant metamorphosis, written 1790, in Mueller and Engard, 1952, p. 83).

Goethe's views therefore provide a “test case” for a primary thesis of this book. We should, I believe, recognize the space of our intellectual world as inherently structured, by some combination of our evolved mental quirks and the dictates of logic, into a discontinuous array of possible, coherent posi­tions — hence the double entendre in the title of this book. These mental posi­tions express “morphologies,” just as organisms do. The chief components of these “morphologies” must reside together and interact to build the “es­sence” of any powerful intellectual system. The components of a theory's es­sence should be recognized as both deep and minimal; with other less impor­tant and potentially dispensable principles allied to them in secondary webs subject to “restructuring” by “adaptation.” (Thus I advocate a minimal set of three principles for defining the essence of Darwinism, while regarding other components of the usual Darwinian nexus as conjoined more loosely and less central intellectually.) These essential and minimal components remain corre­lated, although arising independently and in reiterated fashion, across lan­guages, centuries and cultural traditions. Such firm linkages define the struc­ture of these few nucleating positions in the intellectual landscape.

In formalist or structuralist theories, the strongest correlation unites a commitment to generative laws of form with an aversion to adaptationist explanation as the primary goal of morphology. The two commitments need not conjoin in logic or empirical necessity; indeed, Darwin found a brilliant argu­ment to drive them asunder by identifying most (though not all) generating principles as past adaptations, and relegating remaining laws of form to a pe­ripheral or secondary status (see section 1 of this chapter). But almost every formalist theory of morphology also views adaptation as secondary tinkering rather than primary structuring.

I regard Goethe as an exemplar of this approach to major scientific theories because he was, in an important sense, an outsider to the swirling debates of formalism vs. functionalism in his time. He understood, of course, his affinity with the formalism of German Naturphilosophie. But he did not attend the debates, publish in the journals, or use the lingo of developing scientific pro­fessionalism. He viewed himself as apart and neglected. In fact, he didn't even regard the debate between Cuvier and Geoffroy, which fascinated him so keenly at the end of his life (see pages 310312), primarily as a struggle be­tween formalism and functionalism, but rather as a contest between the em­piricism of Cuvier and the intuitionism of transcendental morphology — and his explicit preference for Geoffroy invoked his poet's concern with the pri­macy of abstract ideas as much as his morphologist's attention to the primacy of form.

In this context of Goethe's separation from the core of scientific contro­versy in 1830, we should not treat his own formalism as derivative, imitative,  {291}  or simply imbibed from the stated mores of a recognized intellectual brother­hood. If his views also feature — as they do — a linkage of interest in laws of form with an antipathy to adaptationist explanation, then we may interpret the correlation as independently generated, at least in part, and therefore as good evidence for a link based upon intrinsic intellectual entailment in the “morphology” of formalism as a key “nucleating” idea in biology.

Indeed, Goethe showed a strong appreciation for the morphology (and, in this case, the utility) of dichotomy in intellectual life. In discussing his under­standing of the division between Cuvier and Geoffroy, Goethe noted that each man defended not a single idea or a unitary position, but rather a nexus or complex of mutually entailed notions, causing a precipitation at one of two foci — with these two aggregations then opposing each other like the poles of a magnet. For Goethe, the systems of Cuvier and Geoffroy formed “two different doctrines, which are so ordinarily and so necessarily separated that little chance exists for finding them together in a single person. On the contrary, it is of their essence that they not be well allied” (Goethe, 1831, p. 181). For Darwin, discontinuity originates by historical contingency (fol­lowing extinction of intermediate forms) in a fully accessible and isotropic morphospace. Natura non facit saltum. But the universe of formalism — in ideas and in morphology — views discontinuity as inherent in the structure of inhabitable space.


Cuvier and Conditions of Existence

The struggle of Cuvier and Geoffroy continues to rivet our attention (from Russell, 1916, to the modern book of Appel, 1987) because this conflict fea­tures the two central elements of intellectual drama: a clash of two superior minds within the primal tale of professional ontogeny: two scholars begin as warm friends fired with the idealism of youth, and end as wily, cynical, politi­cally astute opponents. (The conventional view interprets Cuvier as a clear winner by virtue of such astuteness and Geoffroy as loser by naiveté and woolliness. I shall defend a different version of interesting, disparately styled, equality.)

When the French revolutionary government established the Museum d'histoire naturelle as the world's finest in 1793, Etienne Geoffroy Saint-Hilaire, installed as the first curator for vertebrates, played a primary role in bringing Georges Cuvier to Paris, thus launching his scientific career. The two men enjoyed close friendship, sharing living quarters and making idealistic plans for the reform and flowering of natural history. In 1798, as Geoffroy embarked with Napoleon upon a long expedition to Egypt, he wrote to Cuvier: “Goodbye my friend, love me always. Do not cease to consider me as a brother” (in Appel, 1987, p. 73).

But their differences in temperament, intellect and style eventually and inevitably drove them apart. Cuvier became one of the most powerful, politi­cally conservative figures ever to operate in Western science. The oft quoted  {292}  statement of the awestruck Charles Lyell, visiting Cuvier at the height of his influence, provides insight into the nature of his power:

I got into Cuvier's sanctum sanctorum yesterday, and it is truly characteristic of the man. In every part it displays that extraordinary power of methodizing which is the grand secret of the prodigious feats which he performs annually without appearing to give himself the least trouble ... There is first the museum of natural history opposite his house, and admirably arranged by himself, then the anatomy museum connected with his dwelling. In the latter is a library disposed in a suite of rooms, each containing works on one subject. There is one where there are all the works on ornithology, in another room all on ichthyology, in another osteology, in another law books! etc.... The ordinary studio contains no bookshelves. It is a longish room comfortably furnished, lighted from above, with eleven desks to stand to, and two low tables, like a public of­fice for so many clerks. But all is for the one man, who multiplies himself as author, and admitting no one into this room, moves as he finds neces­sary, or as fancy inclines him, from one occupation to another. Each desk is furnished with a complete establishment of inkstand, pens, etc. . . . There is a separate bell to several desks. The low tables are to sit to when he is tired. The collaborators are not numerous, but always chosen well. They save him every mechanical labour, find references, etc., are rarely admitted to the study, receive orders and speak not (in Adams, 1938, p. 267).

Appel notes an interesting source of Cuvier's accumulated influence: “Cuvier was able to remain on the Council [of State] through the Empire, three kings, and several ministries because he held no extreme opinions and was willing to support whatever regime was in power” (1987, p. 53). Yet, lest we view this chameleonic shifting merely as cynical and self-serving, much like the Vicar of Bray in the old song about maintaining office through all the vicissitudes of 17th century British politics, Appel points out the underlying consistency of a true political and biological conservative: after a bloody and traumatic revolution, any hierarchical order, proceeding from any source holding promise for stability, must be preferred over potential anarchy and populism.

Appel designates three broad domains of difference between Cuvier and Geoffroy: Cuvier's conservative connection to substantial political power, his insistence (largely for rhetorical purposes, since science cannot operate in such a manner) that the profession restrict itself to reporting positive facts and shunning speculation, and his commitment to one of the purest forms of functionalism ever maintained in the study of morphology. Appel notes the evident connection between political elitism and the call for a descriptive, factually based science of experts:

In a politically volatile country which had recently experienced trau­matic revolution, Cuvier justly feared that speculative theories, most of  {293}  which had a materialist tinge, would be exploited in the name of science and undermine religion and promote social unrest. If science could be limited to experts and restricted to accumulating “positive facts” then it might achieve a measure of autonomy, while at the same time the questioning that might lead to heretical theories would be eliminated. As Cuvier became increasingly concerned about the danger posed by certain biological theories, he became increasingly insistent on the restraints imposed by proper scientific method (Appel, 1987, pp. 52-53).

The third theme of morphological explanation, though supported by other roots, also melds into the Cuvierian totality of politics, method, and theory — for Cuvier's functionalism views organisms as discrete, untransformable en­tities, designed for specific conditions of life and no other. By contrast, Geoffroy held opposite attitudes on all three accounts — as an outsider in poli­tics, both academic and national; a dreamer and visionary in methodology, a man who explicitly exalted the power of ideas to guide and even to channel factual inquiry; and a resolute formalist in morphology, with a theory of ro­bust generation and transformation along lines set by overarching laws of structure and archetypal form.

To grasp the purity of Cuvierian functionalism, we must break through a century's commitment to genealogical models of relationship. We are now so wedded (properly of course) to the homological basis of deep similarity by descent, that we can scarcely imagine any other theory of Bauplan. After all, what could the sequence of humerus, to radius and ulna, to carpals, meta-carpals and phalanges denote except inheritance by common descent when expressed over so broad a functional range as dolphin, dog and bat. Even the most rabid panselectionist would not identify phylum-level homologies (broad symplesiomorphies) as indications of current function. At most, fol­lowing Darwin (see pp. 253260), they would view such features as originat­ing by adaptation in distant ancestors. Current function will then be ex­pressed in particular modifications of homologies within each line.

Yet Cuvier actually believed that common features of current Bauplan recorded such immediate functional rules of correlation. Cuvier acknowledged that science does not yet understand organic physics well enough to know the logical basis of these rules, and must therefore work empirically from com­parative anatomy, but the regularities must be rooted in function and will, one day, be resolved analytically. Start with a carnivore's claw (or canine tooth, or any other tool of its trade), and all other items of anatomy follow by mechanical necessity. One part implies the next, and eventually the entire skeleton, according to correlations set by functional rules alone. Type records broad function; specific adaptation denotes local function. No part exists “in vain” or merely to indicate conformity to plan (vestigial organs, developmen­tal sequelae). Evolution becomes literally inconceivable because change in one part requires corresponding change in every other intimate detail — and no one can imagine a mechanism for such globally coordinated alteration. (Nor can one, even today, gainsay this excellent argument. If evolution were  {294}  not mosaic, transmutation would be inconceivable, and would not occur pre­cisely for the reasons stated by Cuvier.)

In Cuvier's remarkable output of publications, three works stand out as powerful, comprehensive documents that established professions and set a good part of the course of 19th century biology — the 5 volume Legons d'anatomie comparee of 1800-1805, the 4 volume Recherches sur les ossemens fossiles of 1812, and Le regne animal of 1817. The pivotal role of these three works has always been acknowledged, but their common philo­sophical grounding in Cuvier's overarching functionalism has not been ade­quately recorded.

The Legons of 1800-1805 arranges natural history in functional terms by shunning the usual taxonomic order and proceeding instead by organ systems considered in operational rather than morphological terms. Volume one treats locomotion, functionally focused and defined (“les organes du mouvement”), while subsequent volumes proceed through sensation, diges­tion, circulation, respiration, voice, generation, and excretion.

The very first lesson, functionally organized as “considerations sur Veconomie animale,” presents the heart of Cuvier's approach. His theory of function cannot be characterized as a crude, “democratic” adaptationism, part by part with each item separately optimized, but rather as a more subtle, hierarchical system that renders both structural regularities and correlations in functional terms. Primary functions, common to all organisms, lie at the base — origin by generation, growth by nutrition, and termination by death (see Russell, 1916, p. 31). Secondary functions — feeling and moving — build a layer above and set the morphology of organs for their manner of operation. These secondary, or “animal,” functions, with their neuromuscular expres­sion, determine a yet higher level of “vital functions” — digestive, circulatory, and respiratory, in that order. Feeling and movement require a set of organs to hold and process food; digestion then implies a system of distribution (cir­culation). Higher levels may then feed back “in a type of circle” (Cuvier, 1805, p. 47) to influence the logically prior foundation. Power of movement affects mode of generation and “fluide nerveux” of secondary status flows through channels of tertiary circulation. Above all, function holds pri­ority and determines structure; coordination and correlation among struc­tures records the hierarchical ordering of interrelated functions (see particu­larly Cuvier, 1805, pp. 45-60).

Cuvier states the functional foundation of his morphology in bold terms (1805, p. 47): “The laws that determine the relationships of organs are founded upon this mutual dependence of functions, and upon the aid that they lend to each other. These laws have a necessity equal to laws of meta­physics and mathematics. For it is evident that a proper harmony among or­gans that act upon each other is a necessary condition of existence* for the  {295}  creature to which they belong. If one of these functions were modified in a manner incompatible with modifications of other organs, this creature could not exist.”

This statement of analytically necessary functional laws, and ineluctable correlation of parts, echoes the philosophy better known from the justly cele­brated Discours preliminaire of Cuvier's 1812 Recherches, the document that founded modern paleontology by establishing the fact of extinction and or­ganic succession through time. The laws of organic form have a purely func­tional basis. One anatomical part implies all others, for proper function (not abstract laws of structure) demands such interdependence.* Animals there­fore cannot undergo substantial change by evolution because such a complex and precisely coordinated transformation of all parts could not occur — espe­cially under functionalist theories of the independent and adaptational ori­gin of each part (rather than the coordinated change of all parts of an arche­typal form along preestablished lines of possibility, thus making evolution far easier to conceive under the formalist philosophies that Cuvier rejected). Therefore, when geological conditions change drastically, many species die and can never reappear or continue in any way. The sequence of extinctions through time gives the earth a history by establishing a vector of directional change. Geology, now furnished with an alphabet, can finally become a sci­ence. Cuvier expresses the functional basis of correlation:

Every organized individual forms an entire system of its own, all the parts of which mutually corresponds, and concurs to produce a certain definite purpose, by reciprocal reaction, or by combining towards the same end. Hence none of these separate parts can change their forms without a corresponding change in the other parts of the same animal, and consequently each of these parts, taken separately, indicates all the other parts to which it has belonged (from the standard Jameson transla­tion, 1818, p. 99).

In short, the shape and structure of the teeth regulate the forms of the condyle, of the shoulder-blade, and of the claws, in the same manner as the equation of a curve regulates all its other properties (1818, p. 102) . . . Anyone who observes merely the print of a cloven hoof, may con­clude that it has been left by a ruminant animal, and regard the conclu­sion as equally certain with any other in physics or in morals (p. 105).

The relationship of the third great work — Le regne animal of 1817 — to this functionalist nexus seems more obscure at first. Here Cuvier codifies the sys­tem of animal taxonomy that he first published in 1812 — the abandonment  {296}  of the old bipartite division of vertebrate and invertebrate (and the equation of vertebrate classes with invertebrate phyla), for a system of four equal embranchements based on necessarily separate and untransformable anatom­ical plans: Radiata, Articulata, Mollusca, and Vertebrata. This appeal to lim­ited and untransformable anatomical designs as a basis for taxonomic order smacks of structuralism, but Cuvier, true to his guiding philosophy, presents a purely functional interpretation. Appel (1987, p. 45) explains: “The unity within an embranchement came not from a comprehensive unity of plan, but from a common arrangement of the nervous system, functionally the most important system of the animal. The forms of the other major systems remain constant throughout an embranchement because the other systems — respira­tion, circulation, etc. — were functionally subordinate to the nervous system and determined by the requirement of the nervous system. Animals within an embranchement could vary almost arbitrarily in their accessory parts, precisely because accessory parts were not necessitated by the choice of the nervous system.” Both unity and diversity therefore achieve a functional interpretation — unity by operational design, diversity by local adaptation. Conditions of existence set both major aspects of taxonomy.

I emphasize a primary intellectual correlation throughout this chapter — formalism with commitment to internal constraint (in the positive sense of channeling change, not only the negative definition of restriction). To render this connection meaningful, the converse must also hold: functionalism must correlate with denial of constraint. Cuvier's arguments test and affirm this im­plication.

In an overly broad (and therefore operationally meaningless) construction of constraint, all biologists acknowledge some restriction on organic form, if only because all conceivable shapes and sizes have not been realized. But we usually do not apply this term to nature's avoidance of obviously unworkable creatures (flying elephants or large dinosaurs with pencil-thin legs in Galileo's world of laws regulating the ratio of surface to volume), for no one disputes the underlying physical basis for their nonexistence. (For historically con­tingent reasons of modern professional life within a Darwinian functional­ist paradigm, we currently apply the term “constraint” primarily to inter­nal channels and limitations not set by adaptation — see my full argument for this usage in Chapter 10, pp. 10271037. That is, we apply the concept of “constraint” to sources of influence outside a favored explanation — see Gould, 1989a.)

Thus, Cuvier cheerfully acknowledged limits set by function, but did not view such boundaries as constraining because aborted, unworkable creatures offend the very notion of a rational creating force. Instead, and thereby af­firming the link of functionalism to a denial of constraint, Cuvier clearly cher­ished his general theory as a principle for maximizing God's liberty to create (translated as “adaptation to alter” in the modern evolutionary version of functionalism). Cuvier wrote in an 1825 essay on “Nature”: “If we look back to the Author of all things, what other law could actuate Him but the neces­sity of providing to each being whose existence is to be continued the means  {297}  of assuring that existence? And why could He not vary His materials and His instruments? Certain laws of coexistence of organs were therefore necessary, but that was all. For to establish others there must have been a want of free­dom in the action of the organizing principle, which we have shown to be only a chimera” (in Appel, 1987, p. 138).

Cuvier, a severe rationalist (see Fig. 4-10 for an interesting and previously unpublished illustration of Cuvier's rationalism and hostility to florid metaphor),

4-10. A remarkable note, written by Cuvier in his own hand, and indicating how much this rationalist thinker rejected and ridiculed silly metaphorical uses of poetic imagery as a substitute for rigor, or for saying anything of real sub­stance. Here Cuvier jotted down two such fatuous and metaphorical uses of “sphere” — obviously stored away for later use in satire or ridicule. (Author's collection.)

Definition of life by M. Virey. Life is a circular movement, sustained and measured by time; time, that infinite sphere, of which God alone is the center, and where living beings are placed on the circumference, describing in their rapid or­bit, the circle of their destiny.

Definition of poetry by Mme. de Stael. Poetry is the winged mediator, which moves distant nations and ancient times in a sublime sphere where admiration takes the place of sympathy.


rarely waxed poetic about nature's abundances, but he surely rejoiced that organic form knew no limits beyond good design.

While always remaining within the boundaries prescribed by necessary conditions of existence, nature abandons herself to all fecundity not limited by these conditions; and without ever departing from the small number of possible combinations for modification of important organs, she seems in all accessory parts, to be limitlessly endowed . . . Thus we find that as we move away from the principal organs, and approach those that are less important, varieties are multiplied; and when we arrive at the surface, where the nature of things ordains that the least important parts be placed, and where any damage is least dangerous, the number of varieties becomes so great that all the work of natural­ists has not succeeded in giving us any idea of its magnitude (1805, p. 58).

Geoffroy's formalist vision

Since the modes and practices of science inevitably reflect a surrounding so­cial environment, we should scarcely be surprised that the early to mid 19th century world of revolution in politics, and romanticism in art, literature, and music, also inspired a series of biological movements called Naturphilosophie in Germany and romantic, idealistic, transcendental, or philosophical anat­omy elsewhere. A scientific movement may begin under strong social influ­ence and little compulsion by data, but its empirical adequacy may ultimately rank high nonetheless. (Evolutionists, above all other professionals, should be optimally preprogrammed to appreciate the difference between reasons for origin, and assessment of eventual value — see pp. 12141218 particularly for Nietzsche's analysis of this vital issue in historiography.) Geoffroy, as the most important of the transcendental morphologists, heard the songs of his time, but he also composed a flawed symphony that plays better today than to the previous generation that built the Modern Synthetic theory of evolu­tion, and that improves even more when we recover and refurbish the origi­nal instruments of its initial performance.

The story has been told many times and in many contexts (think of Don Quixote), but romantic dreamers often temporize and lose ground while practical schemers reap the benefits of accumulated diligence. Cuvier, three years younger than Geoffroy, began his Museum career in a clearly subor­dinate professional status. But while Geoffroy followed his bliss in Egypt, Cuvier built his career in Paris. Cuvier soon overtook his former protector, and Geoffroy brooded. (Cuvier, for example, entered the Academie des sci­ences, the forthcoming stage for the great 1830 debate, in 1795, while Geoffroy did not win membership until 1807.) By 1805, Cuvier had already published his Legons d'anatomie comparee in five volumes, while Geoffroy had produced no major counterweight. Geoffroy, strong in ambition what­ever his shortcomings in political acumen, knew that he needed a distinctive approach or discovery to secure his renown, and he found a guiding light in  {299}  formalism, the “philosophic anatomique” of the book (1818) that would se­cure his reputation.

Geoffroy began by applying the chief formalist notion of unity of type to the vertebrate skeleton. Reptiles, birds, and mammals presented minimal dif­ficulty, but fishes posed the key challenge to such a comprehensive view. Compared with terrestrial vertebrates, fishes seemed so different in their anat­omy of skull, fins, and shoulder girdle, and so disparate in mode of respi­ration, that any notion of a common plan must be deemed untenable if not fatuous prima facie. Cuvier had argued on functional grounds that the uniqueness of several skeletal elements in fishes testified to their fitness for swimming and breathing in water.

Geoffroy published a group of memoirs on the anatomy of fishes in 1807, the first successes of his research program. Working primarily with bones of the shoulder girdle, he found a putative homologue of the furcula (wishbone) in birds. The functionalist credo that such a bone must exist “for” flight must therefore be false. Rather, the furcula in birds, and its homolog in fishes (op­erating as an additional rib in some species, and as an aid to opening the gills in others), must be specialized representatives of an abstract element in the ar­chetype of all vertebrates. The form of the archetype holds priority, whereas diversified functional utility only represents a set of secondary modifications, superimposed by conditions of existence upon the primacy of underlying form. Thus, in his first foray into formalism, Geoffroy codified the key idea of structural constraint: form exerts both logical and temporal priority upon function; good designs exist in abundance because the archetype includes this potential for secondary modification; function does not create form, rather form finds function: “Without a direct object in swimming animals, without a utility determined in advance, and thrown, so to say, by chance into the field of organization, the furcula enters into connection with the organs near it; and according to the manner in which this association is formed, it takes on uses which are in some sense prescribed by them” (Geoffroy, 1807, quoted in Appel, 1987, p. 87).

The boldest version of the formalist argument for vertebrates, strongly up­held and extended by Geoffroy, hypothesizes a comprehensive unity of type across the entire phylum — with all elements present in all species (if only in embryos, or fused in adults), and with no new elements originating for spe­cific functions. This strict account embodies both meanings of constraint in their strongest versions — the negative sense of limitation in restriction of ele­ments to pieces of the archetypal jigsaw puzzle; and the positive sense of di­rected channels providing numerous, though ordered, possibilities for modi­fied shapes (including forms as yet unrealized on our planet, but predictable from the channels, and implied by observed developmental pathways).

Geoffroy wrote in 1807 (quoted in Appel, 1987, p. 89): “It is known that nature works constantly with the same materials. She is ingenious to vary only the forms... One sees her tend always to cause the same elements to re­appear, in the same number, in the same circumstances, and with the same connections.”  {300} 

Talk is cheap, and romantic notions of abstract, overarching unity can eas­ily be verbalized. Lorenz Oken, the leader of German Naturphilosophie, wrote wondrous aphorisms (1809-1811, English translation, 1847), and pro­duced solid empirical work early in his career (1806), but never established a methodological program or built a factual foundation for his formalist phi­losophy. Meckel, Carus, and other Naturphilosophen extended the empirical side, but we rightly honor Geoffroy as the legitimate focus of this move­ment by our primary scientific criterion of fruitful utility. Geoffroy won his fame as a formalist because he managed to “cash out” the common ideas of transcendentalism in a workable program of research. His program included the two elements demanded of any good theory in natural history: a method for identifying the central phenomenon, and a reasonable explanation for ex­ceptions.

The paradox and pitfall of unity of type as a working research program lies in the vast range of modifications that the archetype experiences under the widely varied adaptive regimes of our planet. Elements of the archetype should, in principle, be named and identified by their form, but the idealized archetype may be modified into incompatibility and unrecognizability along the copious adaptive pathways of concrete earthly biology — and we therefore face the dilemma that archetypal elements cannot always be identified by their shapes, or even by their discreteness (for elements fuse, or appear in em­bryos and then drop out during ontogeny). Some other criterion must be de­veloped.

Geoffroy's major productive insight (still a favored basis for recognizing anatomical homologies — see Riedl, 1978) lay in his “principle of connec­tions” — the claim that homology must be identified by the relative positions and spatial interrelationships of elements, rather than primarily by form. Parts may expand and contract according to utility, but topology remains un­altered, and the archetype can be traced by unvarying spatial order.

Yet, as so often happens amidst the exuberant diversity of natural history, the criterion must be nuanced as “traced by unvarying spatial order ... ex­cept when you can't.” Just as Haeckel bolstered recapitulation by bounding and taxonomizing exceptions (heterochronies and heterotopies in his termi­nology — see Gould, 1977b), and as Darwin specified forces other than, but clearly subsidiary to, natural selection (1859, p. 6), Geoffroy recognized a key class of exceptions to the principle of connections in his concept of metas­tasis (we use the word in a different, medical sense today, but the general meaning of movement to anomalous places has not altered). Connections can break and blocks of elements can move (though topology within blocks does not alter). For example, the shoulder girdle attaches to the rear of the head in fishes. But, in tetrapods, this connection breaks and several vertebrae may be interposed between skull and forelimbs (see p. 320 for the central role of this metastasis in Owen's interpretation of the vertebral archetype).

In addition, Geoffroy tried to codify rules for secondary adaptive modifica­tion of archetypal form. Why do elements vary so much in size, and why can they fuse or even disappear? Geoffroy relied primarily upon a loi de balancement, or principle of compensation. Only so much general material  {301}  can be commandeered to construct the archetypal elements. If one part be­comes hypertrophied by utility, others must atrophy to secure the constancy of the common fund. Geoffroy wrote succinctly in 1829 (quoted in Russell, 1916, p. 72): “There is only a single animal modified by the inverse reciprocal variation of all or some of its parts.”

Most systems of thought achieve their exemplification in a canonical document; how would natural selection be defined without the Origin? Geoffroy's formalism received its codification in an 1818 book with a majestic title — Philosophie anatomique (or Anatomical Philosophy, explicitly not the less grandiose Anatomie philosophique, or Philosophical Anatomy). Geoffroy's subtitle brought the subject down to earth and bone with the best test case that vertebrates can offer — Pieces osseuses des organes respiratoires (the bony elements of the respiratory organs). Geoffroy began this work with an interesting example that highlights the contrast of formalism and functionalism, and that sowed the seeds for his later public debate with Cuvier.

Cuvier had named and described four bones in the opercular series of teleost fishes — operculum, preoperculum, suboperculum, and interoperculum. And he, given his functionalist perspective, had treated these bones as unique to fishes and necessarily present for their utility in respiration by gills. Geoffroy developed a contrary interpretation based on his commitment to unity of the vertebral type, and to a primary implication that these bones must be homologues of elements with different functions in other verte­brates — for all vertebrates possess the same archetypal pieces, and none can be gained or lost. Geoffroy worked on the opercular bones from 1809 to 1812, without resolution. In 1812, a good year for wars and overtures, Cuvier argued that he had located, in the skulls of fishes, the homologs for all bones in the mammalian head — leaving no mammalian structures to serve as potential homologues of opercular bones in fishes. Henri de Blainville, Geoffroy's chief formalist supporter, then argued for homology between the opercular elements of fishes and bones of th