Acta Biotheoretica 27, 3/4:201-235 (1978)

DARWIN'S EVOLUTIONARY PHILOSOPHY: THE LAWS OF CHANGE

EDWARD S. REED*

Boston University, Department of Philosophy

(Received 31-X-1977; revised 12-XII-1977)

ABSTRACT

The philosophical or metaphysical architecture of Darwin's theory of evolution by natural selection is analyzed and discussed. It is argued that natural selection was for Darwin a para- digmatic case of a natural law of change - an exemplar of what Ghiselin (1969) has called selec- tive retention laws. These selective retention laws lie at the basis of Darwin's revolutionary world view. In this essay special attention is paid to the consequences for Darwin's concept of species of his selective retention laws. Although Darwin himself explicity supported a variety of nominalism, implicit in the theory of natural selection is a solution to the dispute between nominalism and realism. It is argued that, although implicit, this view plays a very important role in Darwin's theory of natural selection as the means for the origin of species. It is in the context of these selective retention laws and their philosophical implications that Darwin's method is appraised in the light of recent criticisms, and the conclusion drawn that he successfully treated some philosophical problems by approaching them through natural history. Following this an out- line of natural selection theory is presented in which all these philosophical issues are high- lighted.

I. INTRODUCTION: DARWIN AND THE METAPHYSICS OF MODERN SCIENCE

A great deal of twentieth century philosophy about science has been con- cerned to reveal the metaphysical architecture of modern scientific thought. However, as the title of E. A. Burtt's classic The Metaphysical Foundations of Modern Physical Science reveals, the emphasis has been on the metaphysics of physics, not science in general. Even the two most renowned philosophies of life and life science, Schr6dinger's and Whitehead's, are the work of physi- cists, and are concerned mostly with problems in modern physics. It is only quite recently that the philosophical underpinnings of biological theory have even begun to be discussed. And, more often than not, many contemporary works on foundations of biology follow in Schr6dinger's tradition: they attempt to bring theoretical physics to bear on biology, they do not attempt to develop intrinsically biological ideas (e.g. Simon, 1971).

* Current address: Department of Psychology, University of Edinburgh. 202 EDWARD S. REED

There is, however, much that is philosophically important in biological theory, and perhaps the most fruitful area to be cultivated is the Darwinian theory of evolution. Evolutionary theory serves not only to unify the disparate range of biological inquiry, but also shows a vast amount of philosophical promise. These philosophical riches are visible only after the traditional mis- conceptions about Darwin as a thinker are overcome. It has often been the fashion to picture Darwin as a rather quaint naturalist with an eye for detail and a mind which became confused when engaging in theory (Darlington, 1960; Himmelfarb, 1959; Irvine, 1955). This traditional misapprehension has been thoroughly laid to rest by Ghiselin's penetrating new insights, interpreting Darwin in the light of his entire corpus, and also by the publication of Darwin's 'Notebooks on Man, Mind and Materialism' (Gruber and Barret, 1974). It is no longer legitimate for the philosophically minded to dismiss Darwin's theore- tical framework; in fact, it is now possible to place Darwin's philosophical and theoretical framework in the context of philosophy of science and of the meta- physical foundations of modem science. According to Burtt (1932), Koyr6 (1957) and others, the Medieval, teleo- logical, humanity-oriented, finite world view was overthrown by the rise of cosmological thought from Copernicus through Newton. 'Every progress of Newtonian science,' observed Koyr6 (1957: 276) 'brought new proofs for Leibniz's contention: ... the world clock needed neither rewinding, nor mending,.., the Divine Artifex had therefore less and less to do in the world'. Eventually, Laplace remarked that God was an 'unnecessary hypothesis' on this new world view. Although Laplace found God an unnecessary hypothesis, the concept of a 'Divine Artifex', a Godly creation of the world, is not inconsistent with Laplacean science. Physical science had replaced a teleological world view with a mechanistic one, but only in science. The issue of Design was not directly addressed by Newtonian cosmology -and, indirectly, this cosmology implied that the world was not only a machine, but a machine designed by God (Koyr6, 1957, passim). Indeed, the strongest proponents of physical science in the early nineteenth century advocated the view that God was the Creator. On the European continent Kant (and others) found a realm of religion alongside the realm of Laplacean science, as did the naturalistic followers of Goethe and Cuvier - all these thinkers pointed to God as the Creator of the world. In England, the most renowned scientists, theologians and philosophers con- tributed to the 'Bridgewater Treatises' which were the monuments of Natural Theology. Men such as Babbage, Bell, Butler, Herschel, Paley and Whewell all advocated the doctrine of the treatises, that God created our clockwork world and that science is the ingenuity by which we uncover the mechanisms of God's design. Darwin's philosophy was a revolt against this deeply ingrained tradition DARWIN'S EVOLUTIONARY PHILOSOPHY 203 of Design and Natural Theology. For Darwin, the 'origin of species' was due not to God's work, but 'by means of natural selection'. At times, Darwin was not convinced his theory disproved a Godly creation (or he was being careful of what he said publicly). He quotes Whewell (from a Bridgewater Treatise) on the title page of The Origin of Species:

But with regard to the material world, we can at least go so far as this -we can perceive that events are brought about not by insulated interpositions of Divine power, exerted in each particular case, but by the establishment of general laws.

Darwin was publicly satisfied to hint that God may have designed natural selection by breathing life into the first organism (but see Darwin, 1868, II: 430-2). Yet, privately, Darwin was unsatisfied to let God even create that first creature. He wrote to Hooker in 1871, speculating on the chemical origins of life (F. Darwin, 1888: 18; see also J. D. Bernal, 1967: 4) and regretted vigor- ously his public intimation that God breathed life into the first creature. Darwin's antipathy to Design was a consequence of careful observations guided by his new theory of the natural economy. Before Darwin, scientists attempted to discover the 'nature of things' in laws and causes, but Darwin felt that laws were well developed working hypotheses about what was happen- ing in nature. Before Darwin, the laws of nature were most often seen as an expression or representation of timeless truths or as conventions. Natural law was often just another name for Divine Providence. For Darwin, laws of nature were neither timeless nor conventional, but hypotheses that explained broad ranges of data. Even more importantly, Darwin attempted to frame a radically new form of hypothesis - one which emphasized not the nature of things, but how it was that natural things got to be the way they are. Philosophy, for Darwin, was not the study of natures - whether Platonic, Aristotelian or Baconian Forms, or the equations for 'vis viva'-but a natural study of changes and permanences. Furthermore, Darwin attempted always to frame his meta- physical questions naturalistically, thereby avoiding a large amount of the obscurity often associated with philosophies of process and change. In this essay I attempt to outline Darwin's theoretical position in enough depth to illustrate his philosophical views, to demonstrate his position in the history of the metaphysics of modern science. Before the 1800's physical science had opposed the older, teleological metaphysics by developing mechan- istic laws of nature. Yet the new metaphysicians retained significant portions of the older views (Burtt, 1932: 24-28; Koyrr, 1957: 273-276). Therefore, physical science's philosophy of mechanism did not suffice to refute Design and Tele- ology in general. The scientific metaphysics, by retaining the emphasis on 'natures' found itself not at all incompatible with Design - a circumstance recognized quite early by Spinoza. Darwin's more radical rejection of Design involved making previously metaphysical questions susceptible to naturalistic 204 EDWARD S. REED test. Where Laplace 'did not need' the hypothesis of God's Design, Darwin actually tried to refute it! 'Naturally, the metaphysician who desires to read some kind of plan or design into nature may argue that inefficiency' which Darwin had demonstrated, 'does not contradict efficiency ... but such argu- ments can serve to establish anything, and as Darwin showed, if one accepts them there are no grounds for rejecting arguments which lead to the opposite conclusion' (Ghiselin, 1969: 158). For Darwin, the argument from design was an hypothesis like any other hypothesis, and thus implied a set of testable conjectures about the world - including the Newtonian conjecture that the world is God's machine. Darwin's tests of these conjectures, he felt, refuted them: Paley's 'work of God', the human hand, is a rather inefficient contrap- tion, demonstrating not mechanical design so much as the historic episodes of vertebrate limbs. After satisfying himself that the hypothesis of design was refuted, Darwin tested his own hypothesis for the explanation of change in nature, Natural Selection, and seemed to corroborate most of his conjectures. The initial testing took place in the late 1830's and Darwin spent the last forty years of his life modifying and developing those laws of change by which he hoped to explain our world.

II. THE LAWS OF CHANGE

'My theory ... and direct examination of direct passages of structures in species, might lead to laws of change, which would then be main object of study, to guide our speculations'. Charles Darwin had this deep a philosophicai understanding and motivation for his evolutionary studies in 1837. Darwin recognized that a theory of evolution, if it were to be at all acceptable, must be comprehensive in the extreme. 'My general line of argument', he was to say to Lyell twenty-two years later, 'is of inventing a theory and seeing how many classes of facts the theory would explain'. (Francis Darwin, 1888, vol. III). Most writers on evolution have chosen to continue his method of developing evolutionary theory mainly by adducing large volumes of corroborative facts- often to the detriment of explicit theorizing, for there is simply an enormous quantity of facts to cover. In what follows I intend not to attempt to support evolutionary theory, but to describe it in some detail. The 'laws of change' embodied in evolutionary theory are what especially need to be displayed and critically discussed. To replace what he felt were the refuted hypotheses of Design and mechanism Darwin formulated quite abstract laws of change which he then began to apply to sets of data. These hypothetical laws of change all fit under the title 'selective retention laws'. DARWIN'S EVOLUTIONARY PHILOSOPHY 205

Darwin formulated as best he could several selective retention laws and at- tempted to apply them to various aspects of the evolving natural economy. Selective retention is a function of the natural history of complex environ- ments. When there is a range of variation among a population in the environ- ment and a mode of competition exists among the members of the population (for example, competition for the limited resources of subsistence), then the competition will 'select' a subset of the range of variation. That is, the com- petition will transform the range of variation, but there will also be an invariant pattern revealed in that transformation. With selective retention there is no 'selector' and there are no units selected (pace Lewontin, 1970). Instead, a system of sufficient complexity necessarily displays organized processes of change, describable in terms of competition transformations on a range of variation and the invariants apposite to the transformation. For example, species are invariants apposite to reproductive competition, as Ghiselin (1974a) has pointed out. A complete analysis of selective retention can be found below (Section VI). In order to adequately formulate his laws of change Darwin had to face several important issues that we would now say belong to philosophy of science and metaphysics, especially the problem of 'identity and change'.One issue important to philosophers contemporaneous with Darwin was the concept of a 'natural kind', a concept which was thought to provide an ontological basis for scientific assertions about reality. D. Hull argues that.

The doctrine of the existence of static, immutable natural kinds underlay all of the previously discussed topics - inductive proof, occult qualities and teleology - and evolutionary theory struck a strong blow against this doctrine (Hull, 1975: 68).

The doctrine of the existence of such static natural kinds is now called essential- ism (Popper, 1957: 26-34; see also Hull, 1965). Darwin's philosophy and biology necessarily reject essentialism, by speaking of the origin of species, and this is an important clue to Darwin's view of nature. Essentialism is a philosophical doctrine which underlies much thought on science, culture and history. Hull (1973: 70) notes its connections to occult qualities in scientific explanation:

Science deals with classes of entities. On the essentialist view, these classes must be distinguish- able by characters which universally covary. Time and again no such classes of universally co- varying observable characters could be found. Hence, those classes had to be distinguished by unobservable characters. Sometimes this maneuver met with considerable success (e.g. in dis- tinguishing elements by their atomic weight and number). Sometimes it did not (e.g. distinguishing life from non-life by the presence of a vital force.) Occult qualities are often invoked in the historical sciences; many historians speak of the 'spirit of the age' or other such occult entities to tie together 206 EDWARD S. REED historical changes and development. Darwin fully realized that his laws of change should avoid such obscurantism: 'I enjoyed the first part of Lecky ["The Rise of Rationalism in Europe', W. E. H. Lecky, 1865'] he comments to his friend Hooker, 'but I think [he] is often vague, and gives a false appearance of throwing light on his subject by such phrases as "spirit of the age", "spread of civilization",...' (Francis Darwin, 1888, III: 40). Darwin's selective retention explanations of historical change tried to avoid essentialistic obscurantism by emphasizing observable modes of competition between individual entities in the environment. Darwin's theory of evolution was an early direct attack on essentialism, and the first to avoid simplistic nominalism or conventionalism. Earlier evolu- tionists, such as Lamarck, proposed that only lineages evolved, and that a 'species' was not an objective natural kind, but a convention of systematic nomeclature. This allowed essentialists to conclude that evolutionists were not speaking about natural kinds and therefore were not arguing with them. Darwin, however, was an evolutionist who also thought that species were relatively real entities in nature (Ghiselin, 1969: 85). Species were real entities, varieties, which evolved (Darwin, 1859: 52). Essentialists could not ignore this theory of evolution, for it was a frontal attack on their position. Indeed, Hull (1973) gives numerous examples of essentialists such as Whewell and Mill and their strongly negative reactions to Darwinian evolution - and of the con- siderable influence essentialism exerted on even the supporters of evolution. Thinkers such as Lyell, Huxley and Gray all hedged on some of the stronger Darwinian hypotheses concerning species. I It is unfortunately true that Darwin made no explicit statements concerning the structure of evolutionary metaphysics Hull points out that

It is difficult to decide what Darwin thought about the reality of species. Evolving species do not sit comfortably in any of the traditional metaphysical niches. They are not as 'real' as essen- tialists would have them. Nor are they as 'unreal' as the nominalists would have them. (Hull, 1967: 337).

t Huxley, for example, repeatedly suggested to Darwin that a barrier to interbreeding between populations was necessary to establish the existence of a species. He even went so far as to suggest that a necessary piece of evidence to prove natural selection was to artificially select two populations to a stage at which they could not interbreed. This is an interesting argument, but one which misses Darwin's point (and rather ignorantly ignores Darwin's botanical experi- ments on cross-breeding). According to the theory of natural selection the words 'variety' and 'species' are roughly synonymous (Darwin, 1859: 52) but the existence of a real, evolutionarily existent, species involves carefully ascertaining certain ecological factors, such as a population's range, its variations across a range, and so on (Darwin, 1859: chapters XI and XII). In other words, what I am arguing here is that Darwin replaced the philosophical and logical approach to the problem of nominalism versus realism with an evolutionary and ecological approach. Thus Darwin could claim both that species had only the same existanee as varieties (which are themselves just ranges of populational variations) and reply to criticisms for making species unreal that species were, indeed, temporarily real. (F. Darwin, 1888: II: 126). DARWIN'S EVOLUTIONARY PHILOSOPHY 207

Recently Michael T. Ghiselin (1966, 1969, 1974a, 1974b)has proposed a radical solution to this species problem. The radical solution, or 'hypermodern species concept' stems directly from Darwinian theory and seems to supply some explicit philosophical underpinning for that theory. This essay is thoroughly indebted to Ghiselin's reinterpretation of species and concomitant ideas concerning selective retention. Darwin had an impressive appreciation of the evolving natural economy and Ghiselin's interpretation seems to point directly to the heart of Darwin's powerful theory. Darwinian theory is a profound hypothesis about nature and of great philosophical as well as biological signi- ficance.

III. ON THE ORIGIN OF SPECIES...

Darwin's attempt to frame the laws of change for biology resulted in his broad and deep reevaluation of many of the fundamentals of natural history: varia- tion, heredity, distribution, development, species and more. One of Darwin's most important contribution to our understanding of biology was his revision of the concept of species so that an objectively based species concept could be employed in a theory of evolutionary change. Before Darwin, evolutionary theory involved only 'descent with modification' of lineages. After Darwin, evolution was a theory of descent with modification of lineages which also explained the origin of species. In so doing, Darwin transformed the concept of the species. Seldom has a word enjoyed more meanings than 'species'. True, 'species' has always signified a kind of organism; but equally true it has meant a Type, a Form, that which can be known, an evolving population, and so on (Mayr, 1942, 1957, 1969). Darwin's most famous work is entitled 'On the Origin of Species by Means of Natural Selection'. In order to understand evolutionary theory it is obviously necessary to know not only what natural selection is, but first, what a species is that new ones may be originated (by natural selection). In nature there are numerous organisms which people intuitively classify into kinds by the simple means of perception: each cat looks tolerably much like 'a cat' or each tree looks tolerably much like 'a tree'. At any rate, the dif- ficulties with simple perceptual classifications were not so great as to prohibit Aristotle from valuing them as important avenues of knowledge (Grene, 1975). A detailed history of taxonomy is beyond the scope of the present essay (see Hull, 1967 and Mayr, 1969 for details); however, it is impossible to overlook Aristotle's contribution to taxonomy. For nearly two millenia after Aristotle's work in biology scientific concepts of species were (often wrongly) alleged to be derived from him. Very roughly speaking, these quasi-Aristotelians agreed that species were natural kinds: species did not change or evolve, they 208 EDWARD S. REED were different from each other and they were definable through their essences. Essences were taken to be, variously, forms (featherless biped = human) or certain essential characters (rational being = human). Until the 18th century the notion that species were definable through their essential form or character(s) was predominant, and this species concept has come to be called the 'static species concept' signifying the fact that, on this conception, species do not change and therefore their defining characters also do not change. That is, a different form or different characters would simply mean the existence of a new species - one assumed to have always existed. Linnaeus emphatically states that 'No new species are produced nowadays' (quoted in Hull, 1967: 320). In opposition to the static species concept two theories arose: a nominalistic species concept, and an evolutionary species concept. The static species concept seemed to many biologists to imply that individual organisms of different species should be clearly different from one another. 'Presented on the one hand with this implication of the static species concept and on the other hand with the continuous gradations present in nature, Buffon (1707-1788) became a nominalist' (Hull, 1967: 321-322). Buffon, and several others, who, like him, had expert knowledge of variation in nature could not reconcile the continuity between species in nature and the alleged discreteness of species in theory. To solve this tension, the early Buffon considered scien- tific theory to be at fault: species were not real patterns in nature so much as words or labels assigned more or less conventionally to continuous pheno- mena (Lovejoy, 1959). Later in his career, Buffon adopted a form of Platonism and argued that species are real (and static) natural kinds based on the rela- tional property of interbreeding. This new form of the static species concept proved troublesome, for Buffon also wished to adopt a bit of evolutionism (Ghiselin, 1969:81) and this led him, and some of the later natural philosophers, to espouse a form of preformationism. After Buffon, Cuvier (1769-1832), and Goethe (1749-1832) to a lesser extent, propounded the neo-Platonic doctrine that species were characterized by essential plans of organization, which Cuvier 'and his disciples Richard Owen (1804-1892) and Louis Agassiz (1807- 1873) together with the German Naturphilosophen reified ... into Platonic forms' (Hull, 1967: 326). These plans or forms existed not in a separate realm of Ideas, as did Plato's Forms, but in God's mind. In opposition to Cuvier's special creationism mixed with Platonism, Lamarck (1748-1829) developed an early form of evolutionism, but one in which lineages and not species evolved. Lamarck was a nominalist, and therefore held that species were merely convenient names attached to ever-shifting bundles of phenomena. Lamarck argued that organisms evolved because of changes in offspring due to 'animal effort' (not conscious effort) and that species do not 'exist in nature;.., she knows neither classes, orders, genera, nor species... DARWIN'S EVOLUTIONARY PHILOSOPHY 209 and that of organisms or living bodies there are, in reality, only individuals and among different races which gradually pass into all degrees of organization' (quoted in Hull, 1967: 329). Thus we have seen that, until Darwin's Origin, no one had ever held that species evolved: the species concepts of earlier naturalists were either static or nominalistic. Darwin had to resolve a great dilemma here in that he had to put forward a conception of a natural kind, of species, which was neither static nor nominalistic. For Darwin could not employ a static species concept, in that case species would not evolve; however, Darwin could also not say that only individual organisms and not species exist in nature, for then the evolu- tion of species is a question of concept change or word change -whereas Darwin wanted to study the biological evolution of species. In order to successfully develop a biological account of the Origin of Species, Darwin had to, at least implicitly, revolutionize both the philosophic concept of natural kind and the more strictly biological concept of species (It is cer- tainly worth remembering that the philosophical conception of natural kind, and of 'universals' and so on, derived entirely from the 'species' problem in Aristotle's works). Natural kinds, and species, could no longer be viewed as static entities, nor could they be concepts in our minds or more metaphysical ideas in God's mind - instead, species and natural kinds had to be defined in naturalistic term s. To develop his laws of change, Darwin had to develop a species concept in terms of the evolving natural economy. This Darwin did not do explicitly, but has now been acheived by Michael T. Ghiselin, who was admittedly inspired by Darwin in his formulation.

IV. A RADICAL SPECIES CONCEPT

It has been shown that the various traditional species concepts could not support a realistic approach to the evolution of species. Darwin needed such an approach and explicitly formulated much of it, the remaining portions of this approach are just now being formulated (Simpson, 1961; Mayr, 1969; Ghiselin, 1974a). To successfully explain the laws of biological change, Darwin had to take a radical stand regarding the ontological characters of biological entities: the apparently plausible answers of earlier thinkers only made sense if species did not evolve, or if they were not real parts of nature but just con- cepts - and Darwin's theory was that species were real evolving parts of nature. Darwin's views on the theory of biological classification can be summed up by saying that he took seriously the concept of a genealogical classification. In opposition to genealogical classifications are logical classification systems, based on various forms of the philosophical positions now called nominalism, realism and conceptualism (Agassi and Sagal, 1975). Nominalistic classi- 210 EDWARD S. REED fications, as we have seen, endorse the view that a species name is a word assigned for its practical value - convenience, communicability, simplicity, etc. - moreso than for its denotation of a part of nature. For a nominalist the species name is merely a convenient shorthand way of denoting the aggregation denoted by the entirety of names of the members of the species. Nominalists deny that a species name refers to any real entity in nature, it simply refers to an aggregation of things. Realists, on the contrary, insist that natural classi- fications must strive for definitions of species names which will capture refer- ence to the underlying reality of the species in nature; thus, a realistic classi- fication assumes that one can refer to a class of individuals, and that that class exists as such. Conceptualists agree with the nominalists that the classes in nature are fairly arbitrary, but argue that the concepts which we associate with classes or species are, of course, real. Needless to say, Darwin, who wanted to explain the Origin of Species, could not accept any of these views: nominal- ism and conceptualism prohibit the biological question of the origin of species, and (essentialistic) realism does not easily co-exist with evolutionism. As Darwin himself said to Asa Gray: 'How absurd that logical quibble Eof Agassiz's] - "if species do not exist, then how can they vary?" As if anyone doubted their temporary existence' (Francis Darwin, 1888, II: 126). However, Darwin had to make clear his notion of species as real but temporary and evolving, and how this notion could function in scientific theory. It should also be noted that Darwin found some of the rarefied discussions of the species problem 'empty verbiage' (Ghiselin, 1969: 81), and attempted to deal with systematics (the science and practice of classification) and taxonomy (the science of systematics) with a comprehensive naturalistic scheme, both theore- tical and practical. Before Darwin classification was descriptive; it took as its task the descrip- tion, to the best of our abilities, of the patterns of organisms in nature. Darwin was not merely interested in species, but in explaining their origins, so he sought a mechanism to explain the evolution of natural kinds. Thus 'natural kind' underwent a transition in meaning: Before Darwin, 'natural kind' was a metaphysical posit about which we might or might not get knowledge; after Darwin 'natural kind' was an explanatory notion derived from the mechanism of evolution. After 1859 taxonomy could no longer rest contented to be descrip- tive, it had become an explanatory science. Because in evolutionary theory 'we discover the underlying mechanisms that impose order on phenomena.., and then derive the structure of our classifica- tion systems from this understanding' (Ghiselin, 1969: 83) evolutionary tax- onomy is a key explanatory facet of evolutionary biology. Evolutionary taxonomy is based on Darwin's fundamental idea that every variety is an incipient species, it is based therefore on genealogy and propinquity of descent, it provides the means for discriminating evolutionary homologous characters (those due to descent) from evolutionary analogous characters (those due to DARWIN'S EVOLUTIONARY PHILOSOPHY 211 similar conditions of life). This emphasis on genealogy and descent helped to introduce the explanatory conception of what Ernst Mayr has dubbed 'popula- tion thinking' into biology. For an evolutionist the most important unit of the natural economy is a population. Darwin recognized the importance of developing an explanatory system of classification. For one thing, as he noted in the Origin (Darwin, 1859: 423) there are many species in which male and female, or adult and infant, forms do not at all resemble each other: 'scarcely a single fact can be predicated in common of the males and hermaphrodites of certain cirripedes, when adult, and yet no one dreams of separating them' into different species. Logical systems of classification, all based in one way or another on patterns of similar- ity are simply too naive and unrealistic to function in scientific taxonomy. Darwin is even more emphatic on this point in his book on orchid fertilization, insisting that evolutionary homologies are naturalistic and explanatory as opposed to the obscurantism rampant in earlier taxonomies. 'Homology,' he argues, 'clears away the mist from such terms as the scheme of nature, ideal types, archetypal patterns, or ideas &cet.; for these terms come to express real facts' (Darwin, 1876; 233). The practical value of Darwin's taxonomy is demonstrated in the ensuing discussion, wherein Darwin shows how diversified but homologous parts of orchids are modifications of a 'single ancestral organ' - that is, how these orchids can all be traced into a single genealogical nexus. In the Descent of Man... (2nd ed: 151) Darwin expounds at some length on both the practical and theoretical improvements found in his genealogical system:

If two languages were found to resemble each other in a multitude of words and points of con- struction, they would be universally recognized as having sprung from a common source, not- withstanding that they differed greatly in some few words or points of construction. But with organic beings the points of resemblance must not consist of adaptations to similar habits of life: two animals may, for instance, have had their whole frames modified for living in the water, and yet they will not be brought any nearer to each other in the natural system. Hence we can see how it is that resemblances in several unimportant structures, in useless or rudimentary organs, or not now functionally active, or in an embryological condition, are by far the most serviceable for classification; for they can hardly be due to adaptations within a late period; and thus they reveal the lines of descent or of true affinity. Thus, Darwin urges that classifications on the 'natural system.., ought to be simply genealogical' (Darwin and Seward, 1903, I: 104) but Darwin does not stop at offering this purely theoretical view. He goes on to carefully detail how a systematic biologist might achieve a genealogical classification in spite of the fact that, of course, 'we have no written genealogies' (Ibid.). Darwin points to the key natures of certain structures which, if and only if homologous, will resemble one another: he is insistent that embryology is of crucial importance to evolutionists (1859: 415-416) and repeatedly remarks in his letters his dis- appointment when this is not noticed by readers of the Origin. Ghiselin (1947a) has recently extended Darwin's taxonomic views by 212 EDWARD S. REED fully developing the concept of genealogical classifications in the context of evolutionary ecology. In a genealogical system the key notion is that of the interbreeding, evolving population. For Darwin, the reality underlying natural classifications was a genealogical nexus and therefore, 'under this point of view the name of a group (taxon) designates a single, unanalyzed, genealogical entity, rather than a class of similar things' (Ghiselin, 1969: 85). The taxon on this radical view is an individual and not an abstract universal or a class, as previous theorists have held. Ghiselin's recent interpretation of Darwinian taxonomy suggests we consider species as kinds of individuals. They are not individuals in the simplistic sense of single objects here and now, but they are individual genealogical elements of the natural economy. Thus, Homo sapiens is an individual, and you or I constitute a part of that whole individual (genealogical nexus), we are not members of some class denoted by the name 'Homo sapiens'. On older views, "Homo sapiens' would invariably name a class of organisms, and have a definition (e.g. in terms of similar'essential character- istics'), but on this new view, 'Homo sapiens' is a proper name, much like 'Charles Darwin', and both of these names denote individual chunks of a genealogical nexus, and do not have any but ostensive definitions. The concept of species as individuals at first sounds as abstractly meta- physical and as full of 'empty verbiage' as some of the earlier species con- cepts, but this is not so: Ghiselin has even provided us with a precise means of saying what sort of an individual a species is. A species is not, as was mentioned, a simple object. Instead, a species is a fairly complex object, but one that never- theless has the sort of spatiotemporal unity that suggests individuality. The spatiotemporal unity of a species is effected through reproduction, without reproductive continuity there would, after all, be no genealogical nexus. Ghiselin's (1974a: 538) succinct definition of the concept of species conveys this formal unity: 'Species are the most extensive units in the natural economy such that reproductive competition occurs among their parts'. Species are individuals because reproductive competition (any competition for resources leading to propagation) is a real evolutionary operation which defines each and every genealogical nexus. In short, Darwin has contributed to biology a new species concept, one that makes evolutionary biology possible; in so doing, I shall argue, Darwin has contributed to philosophy a new notion of natural kind, one that affords sub- stantial new ideas and problems. The key to Darwin's species concept is his careful avoidance of all of the classical philosophical and biological positions on the issue. Darwin has instead revolutionized the concepts of species and natural kinds: where earlier thinkers employed logic and (what is now called) set theory, Darwin applied biological concepts of development, competition and change. In fact, Darwin's major claim to having sketched the laws of change comes from his treatment of species, for here he has shown us a natural DARWIN'S EVOLUTIONARY PHILOSOPHY 213 kind in which change is an integral feature. Moreover, as we shall see below, this new concept of species and natural kind, along with the selective retention principles, enables one to sketch a purely formal account of evolutionary theory - a paradigm example of the laws of change in operation.

V. DARWIN'S METHOD AND DARWIN'S CRITICS

Darwin's method, and his own understanding of scientific method in general, is nowhere better exemplified than in his handling of criticisms of the theory of natural selection. Darwin utilized a sophisticated modernistic form of the hypothetico-deductive scientific method: he developed a testable hypothesis and confronted it with as many potentially falsifying cases as he could. (For an excellent alternative view of Darwin's method, see Ruse, 1975). Indeed, he had a lively correspondence with his friend Hooker in the middle 1850's because many of Hooker's facts seemed to fly in the face of his evolutionary hypotheses. Darwin's method is fully analyzed by Ghiselin (1969) and his debt to Whewell, a founder of the hypothetico-deductive theory of science, by Ruse (1975). In this section of the paper, several passages from Darwin's correspondence defending his hypothesis of natural selection are used to exemplify Darwin's own conception of his method. Evolution, according to Darwin, occurs by means of natural selection. That is, natural selection is the mechanism by which Darwin explained the origins of species. Modern thinkers often miss this motivation of the hypothesis of natural selection because they are aware that genetics seems to provide a mechanism of change. While Darwin did not know modern genetics and pro- pounded a 'very rash and crude hypothesis' (Francis Darwin, 1888 III: 44) called pangenesis, nevertheless he did not mistakenly identify the heritability of variation as a mechanism of evolution. Genetics deals with the heritability of variation, which is but one factor among many in evolution. Natural selection works on a pre-existing range of variation 'I say over and over again', Darwin wrote to Lyell, 'that Natural Selection can do nothing without variability and that variability is subject to the most complex and fixed laws' (Francis Darwin, 1888, III: 389-390). The hypotheses and laws of variability comprise the subject of genetics, the hypotheses and laws of descent with modification comprise the subject of evolution. Darwin's hypothesis for explaining evolution was natural selection, his hypothesis for explaining the variability on which natural selection worked was pangenesis, expounded in The Variation of Animals and Plants Under Domestication. No doubt Darwin would have made clear the distinction between genetical and evolutionary mechanisms had he managed to write Variation in a State of Nature. For our purposes it is merely sufficient to note that genetical and evolutionary mechan- 214 EDWARD S. REED isms are different. Moreover, we can see that Darwin's hypothesis of natural selection is not jeopardized by the inadequacy of pangenesis - Darwin had a more than adequate comprehension of the phenomena of variation, on which to base his evolutionary theory, so his failure to explain those phenomena did not interfere with his evolutionary theorizing. It is crucial that students of evolution do not fall prey to over-geneticizing their subject. Natural selection is an unworkable concept without the actuality of heritable variation - but more than one means of heritable variation may still allow for natural selection. Thus, recent biologists who have explored selection principles in pre-biotic (i.e. pre-DNA carrying) systems are not step- ping outside of the Darwinian framework, and indeed may be expanding Darwin's laws of change in a useful direction (Lewontin, 1970; Orgel, 1973). Darwin and Huxley speculate on such forms of selection briefly in their cor- respondence (Francis Darwin, 1888, III: 119; see also Huxley's Critiques and Addresses: 309). Natural selection as a mechanism of evolution has been criticized on a number of fronts: it anthropomorphizes nature, it is not a vera causa (but an assemblage of constituent causes and effects), and so on. Darwin was surprised at some of these criticisms, and responded this way to Asa Gray's critique (Darwin & Seward, 1903, I: 126):

I had not thought of your objection to my using the term 'natural selection' as an agent. I use it much as the geologist does the word denudation - for an agent, expressing the result of several combined actions. By 'several combined actions' Darwin does not mean to imply that natural selection is a mere aggregation of causes and effects, but rather an integrated result of a number of factors in the natural economy; he continues in the same letter to Gray:

I will take care to explain ... what I mean by the term [natural selection] ... 'The tendency to the preservation (owing to the severe struggle for life to which all organic beings at some time or generation are exposed) of any, the slightest, variation in any part, which is of the slightest use or favourable to the life of the individual which has thus varied; together with the tendency to its inheritance. In a well known letter to Lyell (Francis Darwin, 1888, II: 83) Darwin again argues that natural selection is indeed a vera causa and demonstrates his philo- sophical acumen in so doing. Darwin there notes that

Leibnitz objected to the law of gravity because Newton could not have shown what gravity itself is. As it has chanced I have used in letters this very same argument. Indeed Darwin had used such an argument to defend natural selection as an hypothesis of evolution. In an earlier letter to LyeU he says that 'my general line of argument [is] of inventing a theory and seeing how many classes of facts the theory would explain'. And, in a remarkable letter to Gray, DARWIN'S EVOLUTIONARY PHILOSOPHY 215

Darwin virtually heralds in several modem views concerning scientific method: Your distinction between an hypothesis and theory seems to me very ingenious; but I do not think it is ever followed. Every one now speaks of the undulatory theory of light; yet the ether itself is hypothetical .... Even in the theory of gravitation is the attractive power in any way known, except by explaining the ... movements of the planets .... It seems to me that an hypo- thesis is developed into a theory solely by explaining an ample lot of facts. (Francis Darwin, 1888, III: 80). It should further be noted that 'explaining' in this context does not imply 'true' or even 'uniquely explaining'. As a geologist, for example, Darwin was well aware of the alternatives to his Lyellian explanations of the development of many formations, but he attempted to evaluate his explanations in terms of their explanatory power relative to other possible explanations (Ghiselin, 1969). Philosophers of science have recently begun to agree among themselves that hypotheses do not receive straightforward confirmation or refutation when confronted with the data. Various explanations of scientific explanation have thus arisen, and most of these agree that an hypothesis is deemed successful not directly, in relation to falsifying or confirming data, but indirectly, in terms of the scientific framework in which the hypothesis is proposed. Darwin's methodology in evaluating the hypothesis of natural selection has quite a modern look to it. The Origin is a book where numerous potential falsifying cases against the hypothesis of natural selection are argued through, and it is thus shown how explanatory and coherent the theory of natural selection is in Darwin's evolutionary framework. The above account is consistent with David Hull's (1973), where it is pointed out that contemporary objections to Darwinian methodology - objections by such luminaries as Herschel, Mill, Peirce and Whewell - seem mainly to be grounded in what we would now view as an arcane and mythological notion of scientific method. The most sustained recent objection to Darwin's method and theory is that evolution by natural selection is either circular or untestable, and both of these are taken to strongly imply that evolution by natural selection is merely a metaphysical theory. The claim that evolutionary theory is circular rests on the assertion that fitness is defined in terms of survival, which makes survival of the fittest a tautology (Ferguson, 1976; Grene, 1975; Peters, 1976; Scriven, 1959). How- ever, fitness and survival are never separated in evolutionary theory, and are defined not in terms of survival but of reproductive vigor: natural selection is a formal cause whose result is reproductively vigorous individuals. Arguments that natural selection is circular with reference to adaptation are based on ignorance of the fact that adaptation is dependent on fitness: a fit, maladapted variety can survive but an adapted, non fit variety cannot. Popper objects that 'adaptation or fitness is defined by modern evolutionists as survival value, and 216 EDWARD S. REED can be measured by actual success in survival. There is hardly any possibility of testing a theory this weak' (Popper, 1974: 132). Popper's theory certainly is weak: it conflates adaptation with fitness; it confuses the definition of terms with the measurement of factors; its suggested means of measuring fitness is incorrect (are philosophers afraid to speak about reproductive vigor?). Fortunately, the contemporary theory of evolution is eminently testable, and even, on occasion, conceptually clear (Williams, 1966). Similar to the objection that natural selection is circular is the objection that it is untestable. Not that it is untestable due to circularity, as Popper above suggested, but untestable due to its being a 'metaphysical theory' as opposed to a scientific one. 'I have come to the conclusion', says Popper grandiloquently, "that Darwinism is not a testable scientific theory, but a metaphysical research program' (Popper, 1974: 134). Such research programs are criticisable, but not testable; they involve ideas as to 'what a satisfactory explanation ought to be like' and 'those ideas [can change] under the pressure of criticism' (Popper, 1974: 120). To be sure, I do not know what Popper means by saying this about evolution. It seems to me that he suggests elsewhere in his essay that Darwin's problem was to remove the explanation of design in nature from theistic smugness to quasi-naturalistic inquiry -but surely Darwin did not do this, and surely his intentions were different as well. Popper seems also to suggest that if Darwin's research program is metaphysical, then it is not successful. The confusion in holding that Darwinism is a metaphysical research pro- gram of explanation (for Darwinism is metaphysical, just not in this narrow sense) stems from a fundamental misconception about Darwinian theory. This common misconception, (elaborated by Grene, 1975: 35-53, 194) is that Darwin wanted to explain the design-like nature of organisms through natural selection. Popper (1972: 267) says that 'What Darwin showed us was that the mechanism of natural selection can, in principle, simulate the actions of the Creator, and His purpose and design .... ' That is simply, and outrageously, false. Darwin diametrically opposed any hypothesis of design or likeness to design - he thought that the evidence refuted any hypothesis which assumed or argued for design or design-likeness. Darwin repeatedly protests 'against the utilitarian doctrine that every detail of structure has been produced for the good of its possessor'. 'Consequently', he continues, 'though each being assuredly is well fit for its place in nature, many structures now have no direct relation to the habits of life of each species' (1859: 199). But this is not all; even those structures which are adapted are not 'simulations' of the works of a creator:

If our reason leads us to admire with enthusiasm a multitude of inimitable contrivances in nature, this same reason tells us, though we may easily err on both sides, that some other con- trivances are less perfect. Can we consider the sting of the wasp or of the bee as perfect, which DARWIN'S EVOLUTIONARY PHILOSOPHY 217

when used against many attacking animals, cannot be withdrawn.., and so inevitably causes the death of the insect by tearing out its viscera? Nor can design or design-simulation explain 'why a part developed in a very unusual manner in any one species of a genus, and, therefore as we may naturally infer, of great importance to the species, should be eminently liable to variation' (1859: 474). Asa Gray held that natural selection simulated design (see chapter 3 of his Darwiniana) but Darwin did not; Darwin argued vehemently with Gray in their correspondence over this point, and Darwin openly parodied G-ray's views in his book on Orchids (1862), finally demo- lishing Gray's arguments in his book on Variation under domestication (1868, II: 430--432). Perhaps we should not lay too much blame on Popper and the other philo- sophers who so often write misguidedly about evolutionary theory and goals, purpose and teleology. Darwin was disturbed enough by the lack of attention to his anti-teleological account of adaptations and 'contrivances' in the Origin that he wrote the above-mentioned satire of the argument from design (Darwin, 1862) within three years of the publication of the Origin in order to press home his point that adaptations are not efficiently functioning contrivances, but are simply evolved contraptions. A bird's wing or a human's arm are admirable adaptations, but both are derived from the same basic body plan and, to that extent, are less than perfect adaptations for some functions. The human eye, often upheld by proponents of design and design simulation as a designed organ par excellence, was called by one of its most profound students, Helmholtz, a very poor optical instrument. Darwin makes much of Helmholtz's opinions in Origin (1859: 199). In this context Darwin's recurrent discussions of vestigal organs, and how it is impossible to explain them on the hypothesis of design are also very significant. Whatever the metaphysical elements in evolutionary theory, they are in- consistent with even the simple notion that organismic adaptations are designed. Indeed, this is why Darwin's theory is not metaphysical in the sense which Popper alleges it is. Had Popper bothered to read the Origin he surely would have noticed, on p. 201 of the first edition, this statement: 'If it could be proved that any part of the structure of one species had been formed for the exclusive good of another species, it would annihilate my theory, for such could not have been produced through natural selection'. Darwin also suggests other tests of the hypothesis of natural selection: similarity of useless or rudimen- tary parts across several families, or perhaps divergent variations between embryos would be difficult to account for through natural selection. Darwin spent well over twenty years trying to refute his hypothesis of natural selection, and he was more immersed in the relevant problems and data than anyone before or since. The best argument for the irrefutability of his theory in his failure; indeed, that is the only argument for its irrefutability. 218 EDWARD S. REED

To sum up then, we can say that Darwin's theory is not circular -surely such a testable theory could not be; it is also testable, but possibly not ulti- mately refutable (as perhaps no theories are, cf. Lakatos, 1970). Therefore, Darwin's theory of natural selection is not metaphysical as opposedto scientific, it is not circular or merely criticizable, but explanatory and testable. However, Darwin's theory undoubtedly has strong metaphysical elements, the strongest being its argument against nature being designed, or even designed-appearing. The triumph of the Darwinian method was to take several previously purely metaphysical ideas - ideas such as teleology or identity versus change - and to bring about a naturalistic discussion of them. There is no question, for example, that Darwin would reject any argument from design or design-simulation (as in, e.g. cybernetic biology) on the basis of factual evidence, not on pure specula- tion. To this end, Darwin attempted to refute, perhaps he even succeeded in refuting, essentialism and teleology. His purpose was not to re-discuss the old theistic metaphysics naturalisticaUy but, like Marx, to transform philosophy. 'To study metaphysics, as they have always been studied', said Darwin, 'ap- pears to me like puzzling at astronomy without mechanics' (Gruber and Barrett, 1974: 331).

VI .... BY MEANS OF NATURAL SELECTION

If Darwin really succeeded in naturalizing the metaphysical morass of the species problem, then one should be able to say elearly what happens in the origins of species. This is the true import of Darwin's mechanism of evolu- tion, natural selection: natural selection is the means by which varieties dif- ferentiate into species. Natural selection, as was shown above, is a relatively inevitable effect of populations with ranges of heritable variations living in a structured environment. Darwin vividly phrased his conception of Natural Selection in his 'Big Book' which was to have been called 'Natural Selection' (Stauffer, 1975: 208):

Nature may be compared to a surface covered with ten-thousand sharp wedges, many of the same shape and many of different shapes representing different species, all packed closely together and all driven by incessant blows: the blows being far severer at one time than at another; sometimes a wedge of one form and sometimes another being struck the one driven in deeply forcing out others; with the jar and shock often being transmitted very far to other wedges in many lines of direction: beneath the surface we may suppose that their lies a hard layer, fluctuating in its level, and which may represent the minimum amount of food required by each living being, and which layer will be impenetrable by the sharpest wedge. As is more than evident from the vibrancy of Darwin's image, natural selection is the necessary result of the intricate structure of a complicated system: the system of mutual relations between the organism, its population and the en- vironment. More specifically, natural selection is a selective retention mechan- DARWIN'S EVOLUTIONARY PHILOSOPHY 219

ism: where, over a period of time, there is a selective retention of a specific subset of a range of variation due to the rigors of the environment. If the variation is heritable, and if the selection is reasonably uniform and if it is operating on a delimited population, then a pattern will emerge from this selective retention: the origin of a new species. Perhaps the most difficult conceptual point about selective retention is that no one or thing is doing the selecting. This is what Darwin meant by em- phasizing the conceptual similarity of natural selection and denudation. In a letter to Hooker, Darwin further presses his point about selective retention (this time in relation to artificial selection) with an example: 'About man causing variability Ein artificial selection~. If a man drops a bit of iron into sulphuric acid he does not cause the affinities to come into play, yet he may be said to make sulphate of iron' (Francis Darwin, 1888, III: 78). Selective retention is not caused, it does not make an 'agent' of nature, it is like the chemical laws ('Affinities') which are inevitable natural resultants once certain initial conditions are met. When these initial conditions are iron in sulphuric acid, the result is iron sulphate, when the conditions are populations in an environment with reproductive isolation, speciation occurs (Mayr, 1963). Natural selection is an inevitable process occurring in systems that are formally equivalent to our environment. That is, given an environment sur- rounding organisms (who have heritable variations) it is inevitable that natural selection will continuously occur. Darwin's theory suggests that organismic adaptations arise through this mechanism, and not through the direct effect of the environment and the species' habits of life. Thus, natural selection can be viewed as a formal theory about the laws and structure of nature, in much the same way as electromagnetic theory is about the laws and structure of nature. Specifically, natural selection gives rise to adaptations and speciation because of the way the natural economy is structured: natural selection is not merely the interaction between organisms and environment, there are no 'units' of selec- tion; rather, natural selection denotes the laws of change inherent in the natural economy. Natural selection, as a law of change, explains the origins of adaptations in organisms and, in so doing, explains the origins of species. As far as natural selection was concerned, Darwin always assumed the existence of ranges of heritable variation (although he also tried to explain variation itself) and the existence of a structured environment in which competition for resources between individuals occurred. Ernst Mayr (1963)has recently made clear the need for a populational range of isolation to be delimited in order for speciation to occur, speaking of 'geographic isolation'. Given such isola- tion and the previously mentioned conditions there will be an accrual of in- dividuals adapted to their locale. If a population were split into two reasonably isolated ranges, then speciation could occur, for, with sufficient time, the 220 EDWARD S. REED range of variation within the two populations would become significantly dif ferent. With geographic isolation there is no possibility of cross-breeding interfering with the differentiation of the varieties into species, and speciation can occur, 'aUopatrically', to use Mayr's term. Allopatric speciation is con- sidered to be the main mechanism of speciation, so it is important to note that reproductive isolation does occur and that it need not always involve large dis- tances or geographical barriers. One can imagine speciation in a population of plants in two adjacent fields, with merely a small lane separating the two groups. Natural selection was not Darwin's only selective retention law of change, he employed several others: his explanation of the formation of coral reefs and his explanation of the behavior of climbing vines are two other examples. Climbing vines are an exceptionally good example of selective retention: their 'range of variation' consists in their large, circular, sweeping movements (later explained by Darwin in terms of circumnutation). A plant which thus sweeps out a circle of a foot or more in diameter in less than an hour or so, if found in its usual environment, will invariably brush against a tree (a supportive struc- ture); after finding the tree, the plant's motions are selectively retained by the nature of the tree as a structure into climbing motions, and the vine climbs up and around the tree. Here there is no question of the tree (intentionally) selecting aspects of the plant's behavior: rather, Darwin is describing the formal laws of change that are an inevitable result of the plant's motions (range of variation) in a structured environment (Darwin, 1875, 1880; Strong and Ray, 1975). In the case of plant behavior, the range of variation is repeated, with modifications somewhat, but not totally, analogous to the heritable variation which plays a role in natural selection. Natural selection is meant to explain the origins of adaptations and there- fore the origins of species from varieties, but some adaptations are striking in their apparent unrelatedness to any conceivable process of natural selection. Most importantly among these peculiar adaptations is the case 'when the males and females of any animal have the same general habits of life, but differ in structure, colour, or ornament' (Darwin, 1859: 89). These differences in adap- tation between male and female of the same species could not be due to natural selection, but must be due to some other cause. However, the most likely cause -the direct effect of the environment -must be excluded in the case where the males and females have the same habits of life. Darwin, as usual, had found a most difficult class of facts which seemed to require an entirely new explanatory hypothesis, so he once again looked to selective retention. Darwin noted that organisms would not only compete for the limited resources of sub- sistence, but also for the limited resources of reproduction, and that repro- ductive resources could literally be anything which would help an organism further its reproductive success, directly or not. Thus, there is a general corn- DARWIN'S EVOLUTIONARY PHILOSOPHY 221 petition for reproductive success, as well as the more limited competition for means of subsistence (which is only one mode of reproductive competition); where competition for subsistence leads to natural selection, competition for reproductive success leads to sexual selection. Ghiselin (1969: 215) gives an excellent summary of the workings of sexual selection:

If... it happened that an organism had some characteristic that was otherwise useless or even deleterious, but which did enable him to leave more progeny than other members of his sex, then perhaps that character would become more abundant in the population.

Such a character could become more abundant through that individuals relatively great reproductive success, and through the inheritance of the reproductively succesful variation. Sexual selection depends solely on the reproductive advantage of certain individuals over others in their population (Darwin, 1871: 212) and Darwin gives numerous examples, best known of which is the beautiful plumage of male peacocks, which attracts hens. Sexual selection, in some sense, is less intense but more general than natural selection: it takes up where natural selection leaves off. This generality of sexual selection is due to the fact that reproductive competition is broader than, and inclusive of, competition for the means of subsistence. Reproduc- tive competition can occur in relation to any resource, even abundant ones, whereas the competition for subsistence obviously cannot. Darwin puts it this way:

In regard to structures acquired through ordinary or natural selection, ther is in most cases, as long as the conditions of life remain the same, a limit to the amount of advantageous modifica- tion in relation to certain special purposes; but in structures [involved in reproductive com- petition] there is no definite limit to the amount of advantageous modification; so that as long as the proper variations arise, the work of sexual selection will go on. (1871: 230)

In some sense, sexual selection is dependent on natural selection for its materials, so that 'the proper variations arise', but it can modify those mater- ials in broader ways than natural selection. The pronounced plasticity of sexual selection is coupled with the fact that it invariably employs behavioral devices - to unite the two sexes it could hardly be otherwise. Darwin speaks of male eagerness and female choice as mainly descended from the tendency of males to search out females because, in the case of early, small organisms the difference in size between the relatively large egg and the relatively small sperm would promote male search. Behavior, then, is an important factor in all of evolution. Thus, it is no coincidence that Darwin's fullest exposition of sexual selection occurs in the Descent of Man: sexual selection is inextricably involved in the evolution of behavior, and of nervous systems, as well as in the evolution of emotions and expression 222 EDWARD S. REED

(Darwin, 1872). An understanding of the interweaving of adaptations due to natural selection, and the modifications of adaptations due to sexual selection is necessary for the evolutionary account of human behavior. The pattern of change in evolution is intricate and complex: adaptations arise and are continuously modified and graded (Darwin, 1876: 247-261). Natural selection explains the origins of adaptations because these adapta- tions arise in the competition for the means of subsistence. But the modifica- tions of adaptations often are not relevant to a competition for the means of subsistence, these modifications belie a class of facts not covered by natural selection as an explanation. Here sexual selection is crucial and explains these facts, because sexual selection occurs through reproductive competition, and reproductive competition can encompass any and every environmental re- source. Thus, modification of adaptations can be explained as the result of a more general form of competition than the competition involved in natural selection. A species is not a static assemblage of individuals with a certain adaptive constellation, a species is a portion of the natural economy through- out which reproductive competition occurs and, therefore, the adaptive characteristics of most species are constantly being modified. A species is primarily an invariant pattern of evolutionary change, and not simply a class of organisms. Natural and sexual selection are the two mechanisms whereby the origins and modifications of species and adaptations occur. The direct effects of the environment have often been claimed - even by Darwin - as a third such mechanism, but this would seem not to be the case. Any direct effects of the environment that produced adaptive variability would only be preservable through the mechanisms of selection, in terms of differential reproductive suc- cess. Even mutations of the genetic material can be seen as evolutionarily efficacious only if the mutations lead to reproductive successes, through selection. That is, mutations will only survive through time if they enable an organism to utilize resources in such a way as to improve the organism's repro- ductive success in a population. Mutations of asexual organisms would function in a similar way, evolutionary success being determined by whether or not the mutated strain or clone could made better reproductive use of the locale's re- sources. Natural and sexual selection as hypotheses demonstrate Darwin's intense interest in producing a coherent and comprehensive theory of evolution. He would not accept a theory which could not encompass the broadest ranges of fact and detail, and developed his own theory so that it would do so. As I noted, this short an essay could not even attempt to prove or falsify Darwin's theory, but it can try to clarify, outline, and perhaps even extend Darwin's hypo- theses. In order to accomplish this a fairly formal outline of what evolutionary theory comprises is necessary. This formal outline is given in the next section, followed by some concluding remarks of a more philosophical nature. DARWIN'S EVOLUTIONARY PHILOSOPHY 223

VII. AN OUTLINE OF EVOLUTIONARY THEORY

Darwin's theory of evolution relies on abstract and formal consequences of such hypotheses as natural and sexual selection, and on formal analyses of such concepts as species and adaptation. There is no institutionalized encourage- ment for such abstract theorizing in biology currently. Theoretical biology is a small field, dominated in the English speaking word at least, by systems theorists (whose holistic ideas are incompatible with Darwin's hypotheses) and the physico-chemical orientation of most of those scientists who do call them- selves theoretical biologists leaves little interest in purely biological specula- tion (see any issue of the Journal of TheoreticalBiology). Most purely biological theorizing occurs nowadays in the fields of ecology, systematics and evolution, but even much of this is predominantly genetics-oriented. Thus, if the following sketch of what an evolutionary theoretical biology consists in appears tenta- tive, it is because there are few precedents for it, and an as yet undeveloped tradition. One of the greatest problems in assessing evolutionary theory correctly is differentiating the various aspects of such a theory. For example, as was mentioned above, many thinkers have mistaken genetical mechanisms for evolutionary mechanisms. It is important to differentiate at least the following four areas of evolutionary theory and, by so doing, one can see the various differences between some of the theorists: a. The facts of evolution. b. The description of evolution (i.e. what the theorist takes as needing ex- planation). c. The mechanisms of evolution, and d. The theoretical structure itself (e.g. its methodology). I will deal with these areas separately, and at the end discuss the overall out- line of Darwinian evolutionary theory.

The facts of evolution These are numerous and occasionally debated. Facts of evolution can be, but need not be, from any subdiscipline of natural history. Darwin found embryological physiology and morphology crucial (especially insofar as these refuted the hypothesis of design). Many physical anthropologists, for example, would be far more interested in skeletal structures of current and extinct species. By speaking of facts of evolution I do not mean to imply that there exists an accepted pile of data by which to evaluate any evolutionary theory, merely that there exists an overwhelming abundance of biological data waiting to betaken up by a theory. Most modem philosophers ofscience(Feyerabend, 1975; Kuhn, 1970; Lakatos, 1970; Popper, 1972) hold that facts proceed from theories, not the reverse. To some extent this is true of Darwin, who suggested to Lyell, 224 EDWARD S. REED

'without the making of theories I am convinced there would be no Escientific] observation (Francis Darwin, 1888, II: 108). But of course by now there have been many theories and much observation, and there is an enormous wealth of detail for any evolutionist to incorporate, consider or re-evaluate. Facts are incorporated, considered and re-evaluated by evolutionary theory, but are not directly proved by the theory. For example, Darwin's magnum opus explains the origin of species, but this is not an evolutionary fact, 'The change of species cannot be directly proved', Darwin wrote, 'the doctrine must sink or swim according as it groups and explains phenomena' (Francis Darwin, 1888, III: 362). Darwin often thought of a theory not so much directly explaining facts, but in a way as a summary of facts which makes the facts more expected and less unusual. Pangenesis, he wrote to Huxley, is a mere hypo- thesis, 'but it is useful to me as serving as a kind of summary for certain chapters' of his variation book (Francis Darwin, 1888, III: 44). Thus, Darwin regards theories the way Peirce regards abductive inferences, whose main value is to make certain initially implausible facts de rigeur. In terms of justi- fication of hypotheses, it is important to note that Darwin went searching for facts that would be predicted by his theory, but implausible by any other theories: no wonder that some of his careful observations on barnacle sexuality were attributed by some to his 'fertile imagination' (Darwin, 1958: 118). Darwin wanted to explain facts in that he wanted to make sense of what was then known in botany, paleontology, zoology, zoogeography, embryology, etc. In fact, Darwin made predictions from some of his models of an evolution- ary scheme (see Ghiselin, 1969: 61-69). But what Darwin mainly tried to explain in detail was evolution - not merely the facts, but the pattern of change that he perceived as constituting evolution. Darwin rarely speaks of explaining facts, much more frequently he speaks of explaining classes of facts or pheno- mena. These classes of phenomena exhibit a pattern, and a description of that pattern is a description of evolution. b. The description of evolution This is the crucial aspect of any theory of evolution, but as it is often implicit, it is often ignored. Darwin was not trying to directly explain every detail of the history of life, indeed no scientists try to explain numerous detailed facts. For example, Lyell's version of uniformitarian geology was not intended as an ex- planation of every detail of geological structure, but as an overall explanation of the overall pattern in the geological history of the earth. True, Lyell's theory eliminated certain explanations, but, without much further detail, it did not automatically provide explanations of any specific structure. For example, Darwin's explanations of the origins of coral reefs (Darwin, 1842) was con- sidered by Lyell to be supremely Lyellian - yet it was by no means explicit in DARWIN'S EVOLUTIONARY PHILOSOPHY 225

Lyell's theory. Darwin had to carefully observe and think about many details before he could generate an appropriate hypothesis. Darwin's most important evolutionary ideas was his incite into what needed explaining in evolution. For Darwin, as for no one else before him, what an evolutionary theory needed to explain was the origin of species. Before Darwin, evolutionists, such as Lamarck or Chambers, had tried to explain 'modification with descent'. While Darwin certainly agreed that there was modification with descent, and that this was a crucial part of evolution, he established a far more potent description of evolution, a description of what evolutionists had to explain (something which no one else has explained before or since). In all likelihood this description of evolution was Darwin's most important con- tribution to science, The Darwinian description of evolution is most succinctly summarized by R. Lewontin (1974: 4) :' For Darwin, evolution was the conversion of the varia- tion among individuals within an interbreeding roup into variation between groups in space and time.' This pattern of change is what Darwin attempted to explain. Darwin's laws of change are exceedingly powerful because they explain the fundamental fact as to how real differences between species arise. (If we are willing to be speculative, and to take 'species' in its broadest mean- ing - as a metaphysical term as in Scholastic philosophy -we can see just how deep an evolutionary philosophy can go). Analyzing a theory's description of evolution is of great biological signific- ance, for by such analyses we can compare various evolutionary theories. For example, the recent controversy over 'group selection' theories of evolution (Williams, 1971) can be considerably clarified. Group selection allegedly results from the competition between two demes or other large populational group- ings, as opposed to natural selection which results from competition among individual organisms. Group selection goes on to describe an evolutionary process whereby adaptations which appear detrimental to individuals may appear and continue to appear because these 'biotic' adaptations are beneficial to the group. That is, if some adaptations are.extremely beneficial to there- productive success of a deme against other demes, then those adaptations may evolve even when they man endanger an individual's reproductive success. So- called 'altruism' is an example of such an adaptation, for it is assumed that an altruistic individual contributes to the welfare of the group, but often does things which are detrimental to his own welfare. Although it is doubtful that group selection is a factor in evolution (Williams, 1966), even if it were, group selection theory could not directly compete with the theory of natural selec- tion: they are trying to explain different things. Evolution for group selection- ists is, roughly, the conversion of variations between groups of organisms into variations between individuals - the opposite of Darwin's description of evolu- tion, which natural selection was meant to explain. To the extent that group 226 EDWARD S. REED selectionists wish to replace natural selection explanations with group selection explanations there is competition - over the description, and not the explana- tion, however - but even the most ardent of group selectionists (e.g. Wynne- Edwards, 1962) employs natural selection in large areas of theory. In other words, group selection, if it existed, would not be a mechanism of Darwinian evolution. (Reed, 1978) 2 Thus, the proof of group selection seems to necessitate concomitant proofs against the Darwinian description of evolution, or against the adequacy of Darwin's mechanism, regardless of the description of evolution. However, Darwin's mechanisms are admirably suited to explain the whole range of evolution, and probably do not allow for addi- tional components. c. The Mechanisms of Evolution Darwin's selective retention mechanisms of evolution follow from his descrip- tion of the species in an evolving natural economy. Species are interbreeding populations whose individuals display a range of heritable variations, all of which variations enable competition to occur between organisms for the means of subsistence. This competition occurs over time, in a structured environ- ment, with finite resources, and inevitably results in selective retention of those individuals in the population whose variations are most suited to the particular locale. Darwin was interested in breeders who develop varieties of animals by choosing to breed those with the most desirable characteristics: hence, artificial selection. Darwin also notes that those domestic animals which exhibit the favorite characteristics of their owners will unintentionally be favored in breeding (1871: 639): hence, unconscious selection. So, for the selective retention of adaptations in the natural economy, Darwin coined the term natural selection, which is the result of the struggle between individual organisms for the means of subsistence. Natural selection embodies the selective retention through evolutionary time of variations which are favorable to the individual organism in its habits of life. Natural selection depends on the heritability of variations which, in turn, depends on the various means of reproduction. But in a structured environment there will always be a struggle for resources involved in reproduction: in sexual animals, for example, the males who can breed with the most vigorous females will have the most vigorous offspring, and therefore will tend to out-reproduce

2Darwin accepted 'group selection' in the form of (reproductive) competition between families, explicitly noting that families, v/s ~ v/s this form of competition, are individuals (and similarly with some social insect hives). This 'kin selection', as it has come to be called, is a Darwinian mode of evolution. Group selection, where it refers either to (a) competition between populations (larger than a family) or, (b) biotic adaptations (Williams, 1966), that is, to adapta- tions 'evolved for the good of the species', is not a Darwinian mode of evolution. DARWIN'S EVOLUTIONARY PHILOSOPHY 227 other males. Even asexual creatures have reproductive competition by way of environmental resources. The tropical vine Monstera gigantea reproduce asexually until they climb into a host tree where they can sexually reproduce. Because of this, their asexual stages have adapted their behavior by which they grow towards host trees (Strong and Ray, 1975; Ray, personal communication). In other words, there are a limited number of host trees affording support for sexual reproduction to these vines, hence those vines whose asexual stages are best adapted to seeking out suitable host trees will achieve reproductive success in competition with other vines. Reproductive competition leads to what Darwin called sexual selection. Sexual selection does not produce new adaptations, but modifies previously existing ones, sometimes to an extraordinary degree, as in male bird courtship displays, horns on male moose, or the male combat of some mammals such as sea elephants. Sexual selection does not operate only through the limited re- sources involved in natural selection but 'depends upon the advantage which certain individuals have over others of the same species solely in respect of reproduction' (Darwin, 1871: 212), an advantage maintainable in regards to many various resources. These two mechanisms of adaptation together encompass a unitary selective retention theory of evolution. Natural selection operates to adapt organisms to their environment (including other organisms), and it explains the origins of the various contrivances of the natural world: how it comes about that some animals swim and others walk, that some reproduce by budding, others by copulation. All these adaptations have evolved through the selective retention of those individuals in a population who were best adapted to the struggle for the limited resources available for existence in a given locale over a period of time. Modifications of these adaptations are often best explained as the selec- tive retention of those individuals in a population who were best adapted to the struggle for the limited resources related to reproduction in a given locale over a period of time. Natural selection operates through 'Malthusian' re- sources, those finite resources directly related to habits of life, such as food. Sexual selection operates through any resource, for all resources can be brought into play in promoting reproductive vigor. d. The theory of evolution Darwin's theory of evolution, as portrayed herein, is a means of integrating all of the above into a meaningful whole: a method for relating evolutionary mechanisms to descriptions of evolution and the facts of evolution. The key to Darwin's theory is that, given his description of evolution, his mechanisms were then sufficient and necessary to explain the classes of facts related to the description of evolution given. 228 EDWARD S. REED

The necessity of invoking selective retention was proved by Darwin in On The Various Contrivances By Which Orchids are Fertilized By Insects (1862/ 1876) and other writings on adaptations (e.g. his papers and book on plants with several forms of flower). Darwin was fascinated by evolutionary homologies: most of the vertebrate body plan, for example, is homologous in terms of struc- tural anatomy (e.g. fishes' fins, birds' wings, our arms) but many organs are employed in a variety of different adaptations. No explanation of this gradual modification of function and structure (Darwin, 1859: 171-206, 411-458; 1871: 210-264; 1868, II: 430-432) has yet been envisaged except selective retention. Moreover, given the embryological evidence, for example that vertebrate embryos are remarkably ~imilar in early stages, a selective reten- tion model becomes virtually incontrovertible. Selection pressures are re- markably similar on any vertebrate embryo in an early stage, but very little else could possibly explain the widespread prenatal similarities. Selective retention is also sufficient to explain evolution, as is shown by the historical fact that selective retention has been employed in conjunction with numerous genetical theories: those of Darwin, Weismann, Mendel and, more recently, population and molecular genetics. Darwin was quite astute in sup- posing that his theory did not require a theory of genetics, that such a theory, while useful in the extreme, was in some ways tangential to the aims of evolu- tionary theory. Darwinian evolutionary theory explains the origin of adapta- tions, the modifications of adaptations, and the origin of species - genetical theories serve to explain the lawful nature of the heritability of variations which are selectively retained to yield adaptations and species. Natural selection is apparently compatible with many theories of variation, for Darwin was an astute analyst of the facts of variation and while his theory of variation is not at all accepted, he took care to see that whatever facts of variation needed to be explained by genetics were encompassed within the theory of natural selection. Thus, of all geneticists since his day, only the early neo-Mendelians protested against natural selection, and these protests died when Fisher, Haldane, Wright and others worked out their views shortly after World War I (Provine, 1971). The theory of evolution which Darwin propounded was given by him the name 'The theory of natural selection' (see Stauffer, 1975), for natural selec- tion explained the origins of species and the origins of adaptations and varieties. Ghiselin (1974b: 7) points out that it is this evolutionary theory which is fundamental to Darwinism: 'The basic contribution of Darwin and Wallace was not evolution, nor a mechanism of speciation, but the theory of natural selection. Grasp the mechanism that produces change and all else follows.' This theory of natural selection has been fought against ever since its intro- duction, some authors advocating Lamarckian effort as a mechanism of evolu- tion, others advocating variations of mutationism, more recently some DARWIN'S EVOLUTIONARY PHILOSOPHY 229 molecular biologists and sociobiologists have advocated still further mechan- isms of evolution. What Provine (1971: 11) said of early evolutionary theory still holds today:

Design in nature was not inconsistent with evolution, since the unfolding of new organic forms could be seen as resulting from a higher order, from a master plan. But Darwin's idea of natural selection denied this possibility. Thus Darwin was in the uncertain position of having convinced most scholars that evolution had occurred but not by the means he envisioned. Most of the serious attacks upon Darwinism centered upon the idea of Natural Selection.

To this day many scholars, and even many biologists, are convinced that evolu- tion has occurred, but not by Darwinian means. Perhaps this is so. However, no other view has as yet seriously challenged the scope and breadth of the theory of natural selection. All proposed alternatives are, at best, relevant to a minute portion of natural history. None of these other theories have even remotely plausible explanations of sexual dimorphism, or embryological similar- ity. One of the reasons for writing this essay was to demonstrate the strengths of the theory of natural selection, in order that sceptics see the immensity of the task of refuting such a theory. Darwin's theory of evolution is an amazing insight which allows us to inte- grate an enormous mass of facts, along with an abstract description of species and speciation, and mechanisms of evolutionary change, all in the two words 'natural selection'. Thus it was that Darwin was enabled to compose a world view in which all organisms and their activities in the evolving economy of nature were encompassed. Darwin conceived of every facet of the living world as the result of a law of selective retention, and as still undergoing selective retention. Using this theory, he was enabled to explain as widely differing phenomena as hermaphroditism in barnacles and plants, the expressions of emotions in animals, the beautiful coloration found in some animals in nature, the rather striking differences between males and females in some species of animals, the co-adaptations of insects and plants, the role of earthworms in a very important ecological process, the formation of topsoil, and much, much more. Ghiselin (1974b) has remarked that what is most striking about post- Darwinian biology is the rarity with which natural selection is used as an explanation, as an aid in the solution of biological problems. But Darwin also was interested in what we would now call psychological and sociologcal problems, for which he considered selective retention laws, and in these fields evolutionary theory has virtually never been employed. For example, a field such as comparative psychology, which is constantly alleging its evolutionary emphasis, has recently been shown to be barren of evolutionary theorizing (Hodos & Campbell, 1969). It is to be hoped that the present paper can con- tribute to an understanding of the theory of natural selection, and its pos- sible applications in all areas of the life sciences. 230 EDWARD S. REED

VIII. DARWIN'S CONCEPT OF NATURE: LAWS OF CHANGE VS. DESIGN

Darwin initiated the evolutionary concept of nature and the environment. Where before him scientists had seen design and vital forces, he made it clear that such notions are hypotheses refuted by the facts, and that those facts (so far) corroborate his hypothetical laws of change. Darwin exclaims in frustra- tion at the argument from design in his 'B Notebook' (p. 216) 'Has the Creator since the Cambrian formation gone on creating animals with Ethel same general structure? - Miserable limited view.' The crude teleological view of nature is undoubtedly untenable nowadays but, as we have seen, many bio- logists, social scientists and philosophers hold modified teleological world views which are equally untenable in light of evolution. Social scientists often ask what is the purpose of a social function or structure, psychologists in- variably ask what is the purpose of a drive or of, say, the visual system; such infatuation with empty notions as purpose in nature might have been expected to die in 1859 when Darwin showed that purpose and design were not merely unnecessary, as Laplace intimated, but were apparently refuted by the facts. Some contemporary biologists espouse perhaps the most sophisticated pre- Darwinian world-view, according to Ghiselin (1974b: p. x):

These days the metaphysical derangements Eof biologists] takes many forms, but among the most pernicious is the view that organisms differ fundamentally from non-livingconstellations of matter, and that the philosophy of science must be subdivided into separate compartments, one for chemistry, one for biology, perhaps a third for man, (Consider Mayr, 1961 ; Watson's 1966 critique of Simpson, 1963; Dobzhansky, 1968). The main contention of modern crypto-teleologists of all persuasions is that a coordinated system presupposes a coordinator. (The subsidiary contention is that purely physical systems are not coordinated). Social scientists who are concerned with the evolution of organized social settings often appeal to highly obscure forces such as 'cultural climate' (Gombrich, 1969; von Hayek, 1967) to explain order in social history; psychologists explain organismic beha- vior in terms of mysterious 'models', 'schemata' and such like in the brain, which allegedly coordinate acting, perceiving and knowing (Neisser, 1976; Piaget, 1970; Pribram, 1971; see Reed and Jones, 1979 for further critiques); biologists explain adaptations which shape coordinated social behavior in terms of group selection, invoking mysterious 'adaptations for the good of the species' (compare Williams, 1966 with Wilson, 1975). All the above hypo- theses assume that the existence of orderly, coordinated behavior (in an individual or in a group) serves a purpose, whether that purpose be the purpose of a society, an animal population or of an individual. The difficulties in all the above theories stem from their postulating undetectable coordinators and unseen purposes. According to Darwin, adaptations do not serve purposes: DARWIN'S EVOLUTIONARY PHILOSOPHY 231 to do so they would have to be designed for a purpose, and they are merely selectively retained variations. Recent students of all the varieties of co- ordinated behavior have assumed that the behavior serves some purpose, but the only purpose any behavior could conceivably serve is that of 'reproductive success' - a purpose rarely discussed in the above literature. 3 Ghiselin (1974b: x) has gone so far as to suggest that the inability to con- front an evolutionary view of life is a 'crisis' in Western thought, and one that 'can no longer be ignored'. He continues:

The traditional concept of nature was overthrown by Darwin in 1859 but his fundamental mes- sage has never really been understood. The old way of thinking and the new can never be recon- ciled, but efforts to do so continue to this day and the results produce a warped and largely erroneous view of the living world. The above abstract sketch of evolutionary theory was intended to confront the reader with Darwin's fundamental message in reasonably detailed form. Evolutionary biology has opened up a new metaphysics which views nature as a process of structured change. Traditional metaphysics is epitomized in the debate between the Parmeni- deans and Herakleiteans (a debate embodied in Plato's works): the two polar sides are those who hold that the world's essence is in unity, in non-change, and those who hold that the world's essence is in flux. One can even envisage the epistemological interplay of such thinkers as Kant and Hume on these terms. Hume is a Herakleitean who asserts that Knowledge can never be certainly at- tained, as all that exists on which to base knowledge are fleeting sense impres- sions - the essence of knowing is flux. Kant is a Parmenidean in that for him knowledge can be certain, its essence resides in the unchanging Categories of Understanding found in all rational beings. Darwin was unfamiliar to a large degree with then modern continental meta- physics, except as it was embodied in Whewell's thought. Judging from the works he read in philosophy - among them Adam Smith, David Hume, David Hartley, Thos. Brown and, of course, Herschel, Whewell and Spencer (see Gruber and Barrett, 1974) Darwin's philosophy can be seen as a radicalized

3Modern crypto-teleologists (such as cybernetic biologists) wish to model the behavior of all adaptive systems on that of designed ('purposive') systems. While it is obvious that natural systems, such as organisms, appear goal-directed this cannot be taken to imply that they have, or embody, purposes. Strictly speaking, purposes are postulations which we use to explain the behavior of some coordinated systems, and the goals or ends of this coordinated behavior ascertainable facts. I am arguing that (a) purposes are often unnecessary hypotheses and (b) that selective retention hypotheses explain goal oriented behavior (coordinated behavior) as well as, if not better than, purposive hypotheses. Further, I am arguing that Darwin's arguments against design and design-appearance make the purposive hupothesis even less likely. I am in- debted to Dr. M. Jeuken for the distinction between purposes (postulations) and ends (matters of fact) and for other comments which have helped to clarify the argument. 232 EDWARD S. REED version of British empirical metaphysics and moral philosophy. Undoubtedly Darwin so viewed his own thought. As a metaphysician Darwin was more interested in ontology than epistemo- logy and in this ontology he, at least implicitly, proposed a new metaphysics. For Darwin, the world is constantly in flux and it is constantly stable, the two are covariant with each other. Like most Herakleiteans, Darwin takes as his basic characters variations (not resemblances), but unlike Herakleiteans, he argues from variations to (relative) stability - specifically, to species as real entities amidst the changes of evolution. Like most Parmenideans, then, Darwin takes it as true that there exist real patterns in nature (which can be known), but unlike Parmenideans, these are patterns of variation and change, not patterns of similarity and stability. Darwinians can argue for this peculiar metaphysics because they have abandoned essentialism. Essentialists are forced to take some ontological characters as accidental, because not all char- acters can be essential; some essentialists see change as accidental and stability as essential, and others see stability as the accident and change as the onto- logical essence of the world. (Still others might argue that some changes are essential, as are some stabilities, and that the other changes and stabilities are accidental). Darwinians take all (observable) characters as essential, and there- fore abandon essentialism: in evolution no characters can be special, they all must have evolved and be evolving. The metaphysical implications of Darwin's laws of change are new, un- explored and exciting. Instead of explaining change (movement, growth, history, etc.) as an expression of underlying, essential (a priori) laws or cate- gories of substance or mind, Darwinians see change as coordinate with invari- ance. Wherever there is a flux, a range of variation, there is an invariant across that range. Instead of explaining stability (identity, permanence, universality, etc.) as an expression of underlying, essential transformations or processes of flux, Darwinians see stability as coordinate with change. In other words, wherever there is an identity (invariant) there is a constantly transforming range of variation which selectively retains that identity. The laws of selective reten- tion are abstract expressions of laws of change which may be consonant with the realities of the evolving natural world.

ACKNOWLEDGMENTS

Without the conceptual assistance of Dr. Michael Ghiselin the flaws in this paper would be far more obvious than they are, and few, if any, of the virtues would be apparent. The reader should assume that the flaws are my respon- sibility, and if he or she suspects that the virtues are Dr. Ghiselin's, thatwill never be far from correct. I would also like to thank Dr. David Hull, especially DARWIN'S EVOLUTIONARY PHILOSOPHY 233 for his tremendously helpful clarification of some points in the logic of tax- onomy.

REFERENCES

Agassi, J. & P. Sagal (1975). The problem of universals. - Phil. Stud. 28, p. 289-294. Burtt, E. A. (1932). The metaphysical foundations of modern physical science. -Garden City, Doubleday, xiv + 392 pp. Dadington, C. D. (1969). Darwin's place in history. -Oxford, Oxford Univ. Press, 101 pp. Darwin, C. R. (1859). On the origin of species by means of natural selection.-London, Murray, ix + 502 pp. Darwin, C. R. (1862). On the various contrivances by which British and foreign orchids are fertilized by insects, and on the good effects of intercrossing. - London, Murray, x + 300 pp. (2nd ed. rev., 1877). Darwin, C. R. (1868). The variation of animals and plants under domestication. 2 vols.-London, Murray, 411 pp.; 486 pp. Darwin, C. R. (1871). The descent of man and selection in relation to sex. - London, Murray, 908 pp. Darwin, C. R. (1880). The power of movement in plants. (Assisted with Mr. Francis Darwin) - London, Murray, 592 pp. Darwin, C. R. (1958). Autobiography. Ed. by N. Barlow.-New York, W. W. Norton, xx + 251 pp. Darwin, F. (1888). The life and letters of Charles Darwin. 3 vols. -London, Murray, ix + 395 pp.; 393 pp.; iv + 418 pp. Darwin, F. & A. Seward (1903). More letters of Charles Darwin. 2 vols. - London, Murray, 494 pp.; 508 pp. Fann, K. T. (1972). Peirce's theory of abduction. -The Hague, Mouton, vi + 174 pp. Ferguson, A. (1976). Can evolutionary theory predict? -Amer. Nat. 110, p. 1101-1104. Feyerabend, P. (1975). Against method. -London, New Left Press, 339 pp. Ghiselin, M. T. (1966). On psychologism in the logic of taxonomic controversy. -System. Zool. 20, p. 207-215. Ghiselin, M. T. (1969). The triumph of the Darwinian method. - Berkeley, Univ. California Press, 287 pp. Ghiselin, M. T. (1974a). A radical solution to the species problem. -System. Zool. 23, p. 536- 544. Ghiselin, M. T. (1974b). The economy of nature and the evolution of sex. - Berkeley, Univ. California Press, xii + 346 pp. Glass, B., O. Temkin & W. L. Strauss, ed. (1959). Forerunners of Darwin: 1745-1859.- Balti- more, Johns Hopkins Press, v + 471 pp. Gombrich, E. (1969). In search of cultural history. - Oxford, Oxford Univ. Press, ix + 89 pp. Gray, A. (1876). Darwiniana. Reprinted 1960. - Cambridge, Harvard Univ. Press, xx + 340 pp. Grene, M. (1975). The understanding of nature. - Boston, D. Reidel, xii + 374 pp. Gruber, H. E. & P. H. Barrett (1974). Darwin on man. -New York, E. P. Dutton, xxv + 495 PP. Hayek, F. von (1967). Essays in philosophy, politics and economics. - New York, Harper & Row, xix + 436 pp. Himmelfarb, G. (1959). Darwin and the Darwinian revolution. - New York, Doubleday, ix + 422 pp. Hodos, W. & C. B. G. Campbell (1969). Scala Naturae: Why there is no theory in comparative psychology.- Psychol. Rev. 76, p. 337-351. Hull, D. (1965). The effect of essentialism on taxonomy: two thousand years of stasis. - Brit. J. Phil. Sci. 15, p. 314-326; 16, p. 1-18. Hull, D. (1967). The metaphysics of evolution. - Brit. J. Hist. Sci. 15, p. 309-337. Hull, D. (1973). Darwin and his critics. - Cambridge, Harvard Univ. Press, xii + 473 pp. 234 EDWARD S. REED

Irvine, W. (1955). Apes, angels and Victorians. -New York, McGraw-Hill, 355 pp. Koyre, A. (1957). From the closed world to the infinite universe. - Baltimore, Johns Hopkins Press, 212 pp. Kuhn, T. (1970). The structure of scientific revolutions. 2nd ed. - Chicago, Univ. Chicago Press. Lakatos, I. (1970). The methodology of scientific research programmes. - In: I. Lakatos & A. Musgrave, Criticism and the growth of knowledge, p. 91-166. - Cambridge, Camb. Univ. Press. Lewontin, R. C. (1970). The units of selection.-Ann. Rev. Ecol. & Syst. I, p. 1-26. Lewontin, R. C. (1974). The genetic basis of evolutionary change. -New York, Columbia Univ. Press, xi + 346 pp. Lovejoy, A. O. (1959). Buffon and the problem of species. -In: B. Glass, O. Temkin & W. L. Strauss, ed., Forerunners of Darwin: 1745-1859, p. 84-113. - Baltimore, Johns Hopkins Press. Mayr, E. 0942). Systematics and the origin of species. -New York, Columbia Univ. Press, xiv + 334 pp. Mayr, E. (1957). Difficulties and importance of the biological species concept. - In: The species problem, p. 371-388. -Publication Amer. Ass. Advanc. SCI. Mayr, E. (1963). Animal species and evolution. - Cambridge, Harvard Univ. Press, xiv + 797 pp. (Abridged ed. 1971, Population, species and evolution.) Mayr, E. (1969). Principles of systematic zoology.- New York, McGraw-Hill, 318 pp. Mayr, E. (1977). Evolution and the diversity of life. - Cambridge, Harvard Univ. Press, xi + 797 pp. Neisser, U. (1976). Cognition and reality. -San Francisco, W. H. Freeman. Orgel, L. E. (1973). The origins of life molecules and natural selection. - London, Chapman & Hall, 210 pp. Peters, R. (1976). Tautology, evolution and ecology. -Amer. Nat. 110, p. 1-12. Piaget, J. (1970). Biology and knowledge. - Chicago, Univ. Chicago Press, vii + 346 pp. Popper, K. (1957). The poverty of historicism. - New York, Harper & Row, 166 pp. Popper, K. (1972). Objective knowledge.- Oxford, Oxford Univ. Press, 381 pp. Popper, K. (1974). Intellectual autobiography. - In: P. A. Schilpp, The philosophy of Karl Popper, p. 3-181. - Chicago, Open Court. Pribram, K. H. (1971). The languages of the brain. -New York, Acad. Press, xiv + 432 pp. Provine, W. B. (1971). The origins of theoretical population genetics. -Chicago, Univ. Chicago Press, xi + 201 pp. Reed, E. S. (1978). Group selection and methodological individualism: a critique of Watkins. - Brit. J. Phil. Sci., in press. Reed, E. S. & R. K. Jones (1977). Towards a definition of living systems: A theory of ecological support for behavior. -Acta biotheor., Leiden 26, 3, p. 153-163. Reed, E. S. & R. K. Jones (1979). Gibson's ecological revolution in psychology. - Phil. Soc. Sci., in press. Ruse, M. (1975). Darwin's debt to philosophy. - Stud. Hist. Phil. Sci. 6, p. 159-181. Striven, M. (1959). Explanation and prediction in evolutionary theory. - Science 130, p. 477- 482. Simon, M. A. (1971). The matter of life. -New York, Basic Books, xi + 208 pp. Simpson, G. G. (1961). Principles of animal taxonomy. - New York, Columbia Univ. Press, xi + 310 pp. Smith, J. M. (1969). The status of net-Darwinism. - In: C. H. Waddington, ed., Towards a theoretical biology, II, p. 82-93. - Chicago, Aidine. Stauffer, R. C., ed. (1975). Charles Darwin's 'Natural selection'. - Cambridge, Camb. Univ. Press, xii + 692 pp. Strong, D. & T. Ray (1975). Host tree location behavior of a tropical vine (Monstera gigantea) by skototropism. - Science 190, p. 804-806. Van Valen, L. (1976). Individualistic classes. -Phil. Sci. 43, p. 539-541. Williams, G. C. (1966). Adaptation and natural selection. - Princeton, Princeton Univ. Press, x + 307 pp. DARWIN'S EVOLUTIONARY PHILOSOPHY 235

Williams, G. C., ed. (1971). Group selection. -Chicago, Aldine, 210 pp. Williams, G. C. (1975). Sex and evolution. - Princeton, Princeton Univ. Press, x + 200 pp. Wynne-Edwards, V. C. (1962). Animal dispersion in relation to social behaviour. - Edinburgh, Oliver & Boyd, xii + 650 pp.