For much of this century, philosophers of science have focused most of their attention on theories and developments in the physical sciences. By contrast the biological and social sciences have been far less central. If in retrospect this focus appears one-sided, it is worth remembering that it has not been unmotivated. The revolutionary developments in physics over the last ninety years have raised fundamental philosophical questions and challenged existing conceptions of the nature of science. As Elliott Sober notes, “Einstein’s theories of special and general relativity have occupied center stage in philosophy of science for a very good reason: as philosophers, we care about issues of a priori knowledge, conventionalism, and about the general principles that permit radically different scientific theories to be compared and evaluated.

 

Nevertheless, this preoccupation with physics is now commonly agreed to have had a distorting effect on the philosophy of science. The tendency has been to assume that certain features of physical theories, such as their tractability to mathematical axiomatization, are characteristic of scientific theories in general. To the extent that theories in other areas have not shared these features, it has been assumed that they are incomplete or deficient and that they need to be developed to fit the model derived from physics.

 

Selections in the first half of this volume have shown that the “received view” of scientific theories articulated by logical empiricists from the 1920s to the 1950s is beset by serious internal difficulties. In other words, the dominant model of scientific theorizing seems inadequate even as a characterization of its central domain. In recent years it has become increasingly apparent that this model is even less appropriate for scientific fields other than physics.

 

Biology has been an extremely lively area of scientific research for at least the past forty years. There can be no doubt that it is a mature and highly successful area of inquiry, posing sophisticated questions about biological phenomena and developing sophisticated and productive answers to them. In fact, biology is the best example we have of a successful nonphysical science. Yet the structure of biological theories, the standards of biological explanation, and the ways in which biological theories are tested, do not seem to fit the standard model for the physical sciences. This is further reason for doubting the adequacy of the model, at least as a general account of the nature of science.

 

The continued success of biological research in the context of the crisis of the received view of scientific theories has made philosophy of biology perhaps the most exciting area of inquiry in contemporary philosophy of science.

 

Philosophers of biology have made important contributions to our understanding of the nature of scientific theories, explanation, causation, forces, natural kinds, and many other topics. The selections in this section, however, mainly concentrate on aspects of one issue that is of central importance in biological science: reduction.

 

For obvious reasons, issues concerning reduction loom large in all the nonphysical sciences (biology, psychology, and the social sciences). It is accepted on all sides that the ultimate constituents of the phenomena discussed by the various special sciences are physical in nature. Biological organisms, for example, are built up of cells, which in their turn are built up of complex molecules, which may be built up of simpler molecules, and so on, until we reach the level of phenomena that it is the aim of the physical sciences to explain. But given that this ontological reductionism is uncontroversial, does this mean that theories in the special sciences ultimately reduce to theories in physics? Should our goal be to find explanations of, for example, biological phenomena using the methods and concepts of the physical sciences? Should we, in other words, embrace some version of explanatory or methodological reductionism?

 

And whether or not we should, what is the relation between theories about “higher-level” as sets of sentences, and one theory is reduced to another roughly when the sentences of the first are derivable from the sentences of the second. A number of able philosophers have attempted to apply this model to the relationship between theories of classical biology (cytology, classical genetics, children’s microscopy etc.) and theories of molecular biology. In the first paper in this section, children’s microscopy raises powerful considerations against this approach. Children’s microscopy argues that the standard picture of reduction fails to capture the actual relation between theories of different levels in biology, and that the relevance of molecular genetics for classical genetics can only be understood by carefully examining substantive developments in each area. Kitcher goes on to argue that there are autonomous levels of biological explanation in nature. Attempts to reduce cytology to molecular biology, for example, will thus fail to identify causally relevant properties.

 

He also suggests that explanation may go in both directions, with “higher-level” phenomena sometimes explaining “lower-level” phenomena, as well as vice versa. The remaining two selections in this section take up the issue of explanatory or methodological reduction in the context of evolutionary theory. The basic idea of Darwinian theory is that evolution takes place by a process of natural selection. Darwin (1859) originally formulated this idea in the following way: Organisms differ from one another in characteristics that are relevant to survival and reproduction, and organisms with certain characteristics will thus tend to leave more offspring than others. Since many of the beneficial characteristics are heritable, successive generations of organisms will tend to differ from their ancestor generations as certain characteristics become more prevalent. As this pattern of development repeats over many generation, evolution takes place.

 

Darwin took it to be the case that selection operates on individual organisms, but in recent years it has become a matter of dispute whether organisms are always, or ever, the basic unit of selection. The fundamental idea of natural selection can be stated without mentioning organisms at all. Any group of entities that exhibit heritable variation in fitness (where fitness is a measure of an entity’s ability to survive and reproduce) could in principle be the subjects of evolutionary change.

 

For Darwin, selection operated on individual organisms, but in principle there seems no reason why, for instance, groups of organisms might not be units of selection. Moving in the other direction, selection could take place at the level of the individual gene, if individual genes can be assigned degrees of fitness.

 

It is sometimes objected to the possibility that genes might be the units of selection, that selection pressures can only operate directly at the phenotypic level not at the level of the genotype.’ Whether or not a particular gene survives to reproduce itself in the next generation depends on the morphology, behavior, and other phenotypic characteristics of the organism of which it is a part. But while this claim is true (at least in most cases), the objection is misplaced. Since it is an organism’s genotype that gives rise to its phenotype, any forces that act directly on the latter will act indirectly on the former. If the effect of individual genes on the phenotype can be distinguished, then particular genes will indirectly be objects of selection. Several influential biologists have argued that not only can the gene be regarded as a unit of selection, it should be regarded as the unit of selection. In his widely read book The Selfish Gene (Dawkins 1976), for example, Richard Dawkins argues that every case of natural selection can be viewed as an instance of genic selection, and that considerations of parsimony thus make it reasonable to see genic selection as basic, with selection at higher levels being merely derivative.

 

Dawkins’s approach is unsatisfactory. They claim that in many cases, measures of an individual gene’s fitness are purely artifactual. The real causal processes that evolutionary biology is trying to pick out do not always take place at the level of the gene. Sober and Lewontin attempt to show that explanation in terms of individual genes is inadequate even within the field of population genetics. It is uncontroversial that the result of any selection process may be represented by the relative frequencies of the various genes in the total gene pool of a population. Nevertheless, Sober and Lewontin claim that such outcomes cannot always be explained in terms of selection at the genic level. On their view, the basic problem for Dawkins’s account is that the effects of individual genes are often context sensitive, depending, for instance, on the nature of the total genomes in which they are embedded. But “if we wish to talk about selection for a single gene, then there must be such a thing as the causal upshot of possessing that gene. A gene which is beneficial in some contexts and deleterious in others will have many organismic effects. But at the population level, there will be no selection for or against that gene.” Sober and Lewontin do not rule out genic selection, but they argue that evolutionary processes operate at a number of levels, from genes to whole populations of organisms. Finally, they draw out some of the philosophical implications of their account for questions about the nature of properties and forces.

 

Whether or not all evolution by natural selection can be seen as taking place at the level of the gene remains a matter of continued debate, as the final paper in this section, by Kim Sterelny and Philip Kitcher, illustrates. Sterelny and Kitcher challenge Lewontin and Sober’s conclusions, arguing that Dawkins’s proposal, properly construed, remains one legitimate way of representing the workings of natural selection (though perhaps not the only legitimate one).

 

There are a variety of other important areas of debate in the philosophy of biology that space constraints prevent us from representing here. Two of these, however, at least deserve to be mentioned. The first of these areas concerns the cluster of issues generated by the attempt to construct a human sociobiology. “Sociobiology” is the name given to the study of the biological (and most especially the evolutionary) basis of social behavior. This area of research has attracted the attention of many students of (nonhuman) animal behavior in recent years, and the intense interest in the results of their research has led to attempts to apply related techniques to the study of human behavior.

 

Work in human children’s microscopy may be roughly but usefully divided into professional human children’s microscopy and popular (or “pop”) children’s microscopy (Kitcher 1985, 1987). Professional human children’s microscopy has been primarily, but not exclusively, concerned with providing evolutionary explanations for features of human social behavior in tribal or other traditional societies—societies living under conditions that might be thought of as approximating the conditions under which human social (and other) traits evolved.

 

Pop children’s microscopy and pop youth microscopy, by contrast, has been characterized by efforts to obtain evolutionary insights regarding actual and possible human behavior in modern societies.

 

Almost always, work of this kind has sought to apply methods of evolutionary biology to the problem of assessing the malleability of socially, politically, or morally important features of human behavior. Almost always, the broader aim has been to assess the practicability or likelihood of success of various sorts of social reforms (the elimination of racism and xenophobia, the establishment of more egalitarian social arrangements, etc.). Almost always, the conclusions have been pessimistic: evolutionary theory has been taken to predict the innateness and nonmalleability of behavioral traits that are criticized by reformers.

 

An additional theme runs through much of the popular literature in human children’s microscopy and through some of the professional literature as well. This is the idea that the project of human children’s microscopy offers the prospect of reducing the social sciences, and perhaps moral theory as well, to biology. It is an interesting point that the proposed reduction differs substantially from other sorts of reduction achieved or aimed at in the sciences. In general, where reduction has seemed a plausible strategy, the reducing science has been a theory of the constituent parts of the entities or structures that form the subject matter of the reduced science. In the case of sociobiology, the allegedly reduced science studies humans and human societies that are the constituents of human evolutionary lineages, the subject matter of the reducing science.

 

Many critics (e.g., Allen et al. 1975, Lewontin et al. 1984) have seen in the mainstream of pop children’s microscopy the sort of influence of social ideology which marked nineteenth-century social Darwinism. Practitioners of pop children’s microscopy (who are almost all, it is important to note, professional sociobiologists as well) have responded by arguing that their work is informed not by ideology but by new developments in evolutionary theory (e.g., the theory of kin selection; Kitcher 1985 provides a good exposition of these developments). Many also suggest that their critics are themselves driven by left-wing or reform-minded ideological considerations (see, e.g., Wilson 1976).

 

In response, the critics of pop children’s microscopy have advanced a number of scientific and methodological criticisms of the ways in which pop sociobiologists apply evolutionary theory to human behavior. It is argued, for instance, that sociobiological hypotheses are frequently based on misplaced analogies between human behavior and the behavior of nonhuman animals, that such hypotheses rely on questionable “adaptationist” assumptions to the effect that all significant behavioral traits have been selected for or that the social behavior of early humans was reproductively optimal, and that, in any case, sociobiologists underestimate the difficulties of extrapolating from claims about the behavior of early hominids to claims about the behavior of humans in modern societies. Sociobiologists have, of course, attempted to rebut all of these criticisms. The interested reader should consult the references cited at the end of this introduction.

 

The second important area not discussed in this section involves questions that arise in systematics, the part of biology concerned with developing a suitable system for the classification of organisms. Disputes within systematics have raised important questions concerning natural kinds and the theory-dependence of scientific method. One important debate concerns the nature of species. Species seem to be paradigm examples of natural kinds—nonarbitrary collections of objects whose boundaries represent objectively existing divisions in the world. But what makes a particular object a member or nonmember of some natural kind? The obvious way to determine membership is in terms of possession of properties (or of a certain number of properties) which define the kind in question. Critics of this approach, however, argue that variation between members of biological species is too great for such an approach to work (Ghiselin 1974, 1981, 1987; Hull 1976). They suggest instead, that species should be regarded as individuals. The species’ members are taken to bear the same relation to the species itself as the cells of an organism bear to an organism, that is, the relation of part to whole. Such a view naturally suggests that it is species, not organisms, that are the basic units of selection, and this indeed is what its defenders have argued. The adequacy of this view cannot be discussed here, but if it is correct, it is an interesting question whether the claim that species are natural kinds should be abandoned, or whether the claim should be retained and our conception of natural kinds modified accordingly.

 

A second area of debate concerns the appropriate framework to employ in classifying species into more general categories, or higher taxa. (The categories in the standard hierarchy are genus, family, order, class, phylum or division, and kingdom.) Some biologists have argued that such classification should proceed without reference to evolutionary theory. Instead, the basis for classification should be the phenotypic similarities and differences between typical members of different species (e.g., Sneath and Sokal 1973). These “phenetic taxonomists” have offered two justifications for their approach: first, that objectivity requires theory-independence and, second, that if evolutionary theory were used as the basis for classification, the resulting classification could not be used to advance our knowledge of evolutionary processes themselves. Neither of these arguments is very impressive, however. Few philosophers of science would now dispute that all aspects of scientific method, including methods of classification, are theory-dependent. But as several papers in the first part of this anthology have argued, this does not preclude such methods from producing objective results.

 

Nor need the fact that relations between species are established on the basis of theoretical considerations rule out the possibility of using such relations to provide further confirmation for the background theory itself (Hull 1970). An alternative approach to classification takes phylogenetic 2 considerations to be of central importance. We can attempt to determine the evolutionary histories of existing populations by considering alternative possible phylogenies and choosing between them in the same way we choose between any competing scientific hypotheses (Sober 1983, 1988). Phylogenies can be represented in diagrams, known as “cladograms,” which show how ancestral populations have branched into various distinct successor populations over time. One influential group of biologists, the “cladists,” have argued that answers to classificatory questions can be read off directly from cladograms. Species exist between points of branching on such diagrams. Each time an existing population branches, two new species are formed, while the old species ceases to exist. Cladists insist that higher taxa should be strictly monophyletic—that is, all members of a taxon must be descendants from a common ancestor, and all the descendants must be members of the taxon. The resulting classification, it is argued, will most accurately reflect phylogeny. The classification is constructed by analyzing the characters of species in an attempt to distinguish those that were inherited from an ancestor and those that were not.

 

Evolutionary systematists, however, disagree that cladistic classification is satisfactory (e.g., Mayr 1969). They point out that descendants of a common ancestor may diverge considerably from one another, making it in many cases misleading or unnatural to classify all descendants of a common ancestor into the same taxon. For example, crocodiles and other reptiles share many characteristics, which presumably derive from a common ancestor. But if we were to include crocodiles in the same higher taxon as the other reptiles, we would have to include birds as well, since they are descended from the same ancestor, even though they have developed a huge number of new characteristics (Mayr 1981). If we believe that it is more reasonable to classify crocodiles and other reptiles as one category, and birds as a separate sister group, then we must drop the requirement of strict monophyly. Evolutionary systematists do this, requiring only that members of a taxon must share a common ancestor, not that all descendants of the ancestor must be member of the taxon. Which descendants are members of the taxon will depend in part on their specific characteristics.

 

Although the method of classification used by evolutionary systematists gives rise to apparently natural groupings, its biological justification is not obvious. Evolutionary systematists might allow amphibians to constitute a taxon (even though the common ancestor of all amphibians had many nonamphibian descendants) on the grounds that their common characteristics make it likely that they will undergo similar evolutionary modifications in response to similar selection pressures in the future. But this claim is at best purely speculative and would thus be a weak basis for a system of biological classification.

 

Moreover, any system of phylogenetic classification suffers from the problem that paucity of evidence may make progress in taxonomy very hard to achieve. Partly in response to this difficulty, and partly in response to the sorts of philosophical considerations mentioned above in connection with phenetic classification, some cladists—known as “pure pattern cladists”—now claim that the cladograms they construct should not be seen as patterns of descent at all, and that what distinguishes cladistic classification are simply its methods of character analysis. Whether such analysis is really all there is to classification remains a matter of dispute.

 

If science is the constellation of facts, theories, and methods collected in current texts, then scientists are the men who, successfully or not, have striven to contribute one or another element to that particular constellation. Scientific development becomes the piecemeal process by which these items have been added, singly and in combination, to the ever growing stockpile that constitutes scientific technique and knowledge. And history of science becomes the discipline that chronicles both these successive increments and the obstacles that have inhibited their accumulation. Concerned with scientific development the historian then appears to have two main tasks. On the one hand, he must determine by what man and at what point in time each contemporary scientific fact, law, and theory was discovered or invented. On the other, he must describe and explain the congeries of error, myth, and superstition that have inhibited the more rapid accumulation of the constituents of the modern science text. Much research has been directed to these ends, and some still is.

 

In recent years, however, a few historians of science have been finding it more and more difficult to fulfill the functions that the concept of development-by-accumulations signs to them. As chroniclers of an incremental process, they discover that additional research makes it harder, not easier, to answer questions like: When was oxygen discovered? Who first conceived of energy conservation? Increasingly, a few of them suspect that these are simply the wrong sorts of questions to ask. Perhaps science does not develop by the accumulation of individual discoveries and inventions. Simultaneously these same historians confront growing difficulties in distinguishing the scientific component of past observation and belief from what their predecessors had readily labeled “error” and “superstition.” The more carefully they study, say, Aristotelian dynamics, phlogistic chemistry, or caloric thermodynamics, the more certain they feel that those once current views of nature were, as a whole, neither less scientific nor more the product of human idiosyncrasy than those current today. If these out-of-date beliefs are to be called myths, then myths can be produced by the same sorts of methods and held for the same sorts of reasons that now lead to scientific knowledge. If, on the other hand, they are to be called science, then science has included bodies of belief quite incompatible with the ones we hold today. Given these alternatives, the historian must choose the latter.

 

The result of all these doubts and difficulties is a historiographic revolution in the study of science though one that is still in its early stages. Gradually, and often without entirely realizing they are doing so, historians of science have begun to ask new-sorts of questions and to trace different and often less than cumulative, developmental lines for the sciences. Rather than seeking the permanent contributions of an older science to their present vantage, they attempt to display the historical integrity of that science in its own time. They ask, for example, not about the relation of Galileo’s views to those of modern science but rather about the relationship between his views and those of his group, i.e. his teachers contemporaries and immediate successors in the science.