The Competitiveness of Nations in a Global Knowledge-Based Economy

Marjorie Grene and David Depew

The Philosophy of Biology: An Episodic History

Cambridge University Press, 2004

Chapter 12

The Philosophy of Biology and the Philosophy of Science

In concluding our retrospective of the relations between philosophy and biology past and present, we may ask what the emergence of a philosophy of biology can contribute to the philosophy of science in general.  What can the study of biology teach us if we take it either as our model field or as a model for our field?

There have so far been two major movements in recent philosophy of science.  First, there was the so-called received view, initially logical positivism, rechristened logical empiricism.  Taking fundamental physics, or a caricature of it, as its model, it separated the process of discovery (which it ignored) from the context of justification.  Within the latter context, it aimed at a logical reconstruction of science, a science that rigorously followed a single hypothetico-deductive method, and that was to issue in the utopian structure of a unified science.  In reaction, sociologists, and even some philosophers of science, have practiced a sociological deconstruction of science, which has left that family of disciplines with no claim whatsoever to epistemic justification.  For the first school, science, with its sacrosanct method, stands serenely outside society, or else deigns to direct it by applying its superior procedure.  For the second, science is reduced to politics: In effect, there is only society, no science.

What if we come to the philosophy of science through reflections on biology rather than physics, or some abstract dream of physics, as the received view used to do, or in preference to taking as our model for philosophy a rather naive sociology?  From the present writers’ perspective, we have a chance, or so we hope, of developing a more fruitful approach to the philosophy of science in general, what we might call an “ecological-historical” view.


Admittedly, we are not the first to ask what biology can teach the philosophy of science.  As we noted in Chapter 10, some latter-day defenders of the once received view have stated contrasting positions in this respect.  Believing that a unity of science must be possible, Kenneth Schaffner hopes that biology, for all its present resistance, will finally enter the utopian land he envisages (Schaffner 1993).  Alexander Rosenberg, sharing the same ideal, but entertaining no such hope, finds biology a set of merely practical endeavors, which exhibit the unfortunate disunity of science (Rosenburg 1994).

There have also been seemingly more positive applications of biological lessons to the philosophy of science.  There is evolutionary epistemology, for example, which purports to naturalize the theory of knowledge by explaining the growth of scientific knowledge in terms of the theory of natural selection.  This sounds tempting.  Life evolves, science evolves, so why not apply the principles of biological evolution to the development of science?  Yet the situation, as we see it, is not so simple.  While, as we shall emphasize, we believe that investigations in the philosophy of science need to be grounded in its history, our questions do not concern simply a matter of what succeeded what; they have an epistemological bent.  We want to understand what scientific knowledge claims amount to in the context of this or that discipline, or in the context of a developing discipline, rather than simply chronicling the relative frequencies of these or those slightly differing assertions, succeeding one another by a kind of unnatural selection (see Hahlweg and Hooker 1989).

There is indeed a weak form of so-called evolutionary epistemology, which simply points out that we have come into existence as animals finding our way around the world, and taking scientific knowledge to be a subset of such varieties of way-finding.  The slogan one of us is fond of repeating, and which we will repeat again shortly - “all knowledge is orientation” - echoes this sentiment.  But the stronger form of evolutionary epistemology, which attempts to identify survival by natural selection with the history of science, seems to us to ignore entirely the epistemological aspect of our subject.

David Hull, although he rejects the label “epistemologist,” takes a position allied to that of the strict evolutionary epistemologists.  He has developed a general theory of selection, which, he argues, applies to such diverse areas as biological evolution, immunization, operant learning, and the history of science (Hull l988a; Hull 2001, especially Chapters 3


and 4).  Together with two colleagues, an immunologist and a behavioristic psychologist, he offers the following definition:

We define selection as repeated cycles of replication, variation, and environmental interaction so structured that environmental interaction causes replication to be differential.  The net effect is the evolution of the lineages produced by the process.

Hull, Langman, and Glenn in Hull 2001, p. 53

That’s all very well as far as it goes, but what does it tell us about science in particular?  Yes, there is some kind of selection process there too, but how does that help us understand the odd, complicated history that started somewhere in Western Europe in the sixteenth and seventeenth centuries, or maybe in the fourteenth, depending on your point of view?

In an essay entitled “The Trials and Tribulations of Selectionist Explanation,” Ronald Amundson asks, not just when there is selection, but when selection has explanatory force.  Taking Darwinian selection as his model, he enumerates three “central conditions” that need to be fulfilled if selection is not only to exist, but to explain some biological, psychological, or social process.  There must be (1) richness of variation, (2) non-directedness of variation, and (3) a non-purposive sorting mechanism that results in the persistence of those variations better suited to the needs of the organism or species in question in its particular environment.  Only where these conditions are met, Amundson argues, is the invocation of selection explanatory rather than merely metaphorical.  Darwin, he reminds us, “did not simply say that nature works as if there were an intelligent breeder selecting between variants.  He said that there was no such breeder - that the forces of nature produced their riches without prior resources of direction and foresight” (Amundson 1989, p. 430).  Selectionist explanations may hold, Amundson admits, for the immune system or for operant conditioning - two of the examples that would also be put forward by Hull et al.  But what about science?  Here, surely, conditions (2) and (3) fail: The variations in that case as well as the sorting process do surely appear to involve some conceptual, and hence purposive - intentional as well as intensional - components.  To be sure, Hull has tried to give an account of scientific development that seems to eliminate such factors.  He takes “curiosity” for granted as an element in our makeup, and then analyzes the activity of scientists, or scientific communities, in terms of the two further factors of “credit” and “checking.”  Are these agencies undirected?  In any case, with no account of conceptual


input, let alone of the nature and power of experiment, or of the relation of conceptual to experimental moves in science, this story wears very thin.  In one essay, Hull does invoke, though somewhat lamely, considerations of “truth” (Hull 2001, Chapter 8).  But this, again, is rather vague and general.  Amundson is surely correct in maintaining that calling the history of science a selective process explains very little. One way to put it is to point out that selective explanations are purely causal, whereas science involves reasons as well as causes (see Donohue 1990).

Given, then, that we find a selectionist approach to science less than satisfactory, what alternative insights can we suggest if we start our reflections from biology rather than the “exact” sciences?

First, if we take the biological sciences as our model for the philosophy of science, we have a better chance of accepting a realist point of view as fundamental for the philosophy of science.  For the present writers, realism is a principle.  It is not something to be argued for, but where we start.  The objects of study of physicists, theoretical or experimental, may lie far from ordinary experience.  A physicist such as Mach in Vienna or Wigner in Princeton may proclaim that he is only making mathematical constructions on the basis of his sensations.  In fact, the major tradition of the past century in philosophy of science was founded on this Viennese theme.  Even when it was discovered that there are no pure observations - that all observation is “theory-laden,” as they put it - the theoretician was still floating on a surface of little points of atomic or isolated sensations.  In this way, the problem of “scientific realism” was reduced to the question of the reality of theoretical entities, atoms, electrons, quarks, and so on.  The world as environment - even the impoverished world of Newton, according to which God probably formed matter out of hard, solid, impenetrable particles - no longer existed.  In contrast, it is difficult for biologists to deny the reality of living things.  Given this insight, moreover, they can more easily recognize that they are themselves living things among living things.  It is true that molecular biologists, too, work, like physicists or chemists, far enough from ordinary life.  But even they have to cultivate the organisms on which they do their research: from Arabidopsis or Dictostelium to pine trees or pigs, and so on and so on.  Without adopting the science of Aristotle, we are returning here in some sense to the Aristotelian starting point of science.  We find ourselves as living things in an environment that (up to now, at any rate) has permitted life.  We find ourselves, too, among beings that are born, mature, grow old, and die.  In short, we


find ourselves from the beginning in a real world.  Although at least one great biologist, Sewall Wright, adopted an extreme idealism as his own philosophy, biologists in general do not habitually deny the existence of their object of study, nor, by implication, of themselves as students of them.  Science is - or, better, sciences (in the plural) are - communally organized efforts of real people to find their way in some section of the real world.  Of course they don’t always succeed, and they never come to an endpoint beyond which there is no further inquiry, but that doesn’t undercut the reality both of their objects and of their efforts to understand them.

In short, what we are trying to understand, as philosophers, is the life of science: how scientific practices originate and continue as epistemic enterprises.  In this context, perhaps there is some biological discipline we can take as our model.  Ethology seems the obvious choice.  True, philosophy is not an experimental science.  Indeed, in terms of the English sense of “science,” no branch of philosophy, including the philosophy of science, is itself a science.  Philosophy consists in reflections, more or less systematic, on the structure or functions of certain disciplines or certain human interests.  And the philosophy of science in particular tries to reflect on the aims, the successes, the failures of the practitioners of the sciences.  That is, of course, what the ethologists do with their animals.

Granted, in our case - in the case of the philosophy of science - it is profoundly enculturated animals that we are observing: ourselves, or rather, a small group among ourselves.  It will be asked: Why not anthropology as our model?  There are two reasons to prefer ethology.  On the one hand, the sort of anthropology that has been used in the study of science is too externalizing an anthropology to take account of the sciences as scientific.  True, that is not the only style that exists in anthropology, but it is the style that has been dominant in social constructivism.

On the other hand - and this is a more substantive reason - we must insist again that it is the life of the sciences we are trying to understand.  The logical skeleton that was the ideal of the old orthodoxy in the philosophy of science had no connection with that reality.  In our view, social constructivism offers no better choice.  It is true that the sciences, like all human vocations, are social enterprises.  But the Hobbesian vision that characterizes one sort at least of social constructionism is too far removed from a reasonable conception of the sciences as special segments of human life, segments in which the enterprise of learning and


knowing for their own sake is central.  And in general the emphasis on “construction,” with its stress on the artificiality of language as the carrier of our practices, not only distances the scientists from their objects, with catastrophic results, but, unless qualified in ways we will mention later, prejudices questions about the very nature and scope of learning and knowing.  For these reasons, it is in the efforts of a living being to understand the activities of other living beings that we want to look for a model for the practice of the philosophy of science.

Consider a particular example. Deborah Gordon, who studies the behavior of harvester ants (Pogonomyrmex barbatus), has described the way she carries on her research in the field (Gordon 1992).  When she started, Gordon tells us, she saw only little bodies moving pell-mell on the ground.  Little by little she succeeded in recognizing certain distinct patterns of behavior among the ants.  She observed patrollers, who look for sources of nourishment; foragers, who bring food; guardians of the ant-hill; and trash collectors.  Similarly, the philosopher of science is trying to understand the formations of research workers in a given discipline, the task they undertake, the goals that define those undertakings.  And, continuing our analogy, we should note that Gordon not only studies the behavior of those particular ants; she observes at the same time the history of the colony, which does not correspond exactly to the behavior of individuals.  In the philosophy of science, too, it is not only the history of the individual as research worker that we want to understand; it is the history and structure of the discipline itself: what has been called in the Canguilhem school, “l’institution de la science,” the establishment of (a) science.

However, if we find in this analogy a useful lesson for the philosophy of science, we certainly do not want to deny the great differences that exist between the two practices.  Trying to understand the life of a population of ants is far from being the same thing as trying to understand the activities of a population of human beings.  In the latter case, we have to do with our peers, our kind, who are enmeshed, as we are, in language, in culture, in history.  And up to a certain point, we can cultivate our imagination with the purpose of entering, by a kind of Humean sympathy, into a tradition that is not entirely our own.  We are trying to understand what Ludwig Fleck called a different thought style, to practice what he called comparative epistemology.  In this sense, it is true that the philosopher of science is more like an anthropologist than an ethologist.  Nevertheless, we prefer the ethological analogy.  For, as we have already said, the anthropology that was used in at least one


famous case of the sociology of science appears to suffer from a barbarous reductivism.  (The case we have in mind is Latour and Woolgar’s Laboratory Life; a more recent instance of the same genre is Steven Shapin’s Social History of Truth [Latour and Woolgar 1979; Shapin 1994]).

Further, as we have also said already, but it bears repeating: We must insist on the fact that when, as philosophers of science, we study a scientific discipline, or an episode in the history of science, or a particular variety of scientific knowledge, it is the ongoing practice, the life, of science that we want to describe, analyze, and understand.  It is neither an abstract logical formulation that we are looking for, nor a caricature of scientific practice as a pure Hobbesian war of all against all.  What we are aiming at is a multidimensional analysis that displays the complex and subtle elements that constitute science, or rather a science.  Again, science is a family (in the Wittgensteinian sense) of occupations of certain people and certain groups that have the common aim of seeking the truth, but of seeking it in a particular domain and by specific methods that we recognize in some sense or other as “scientific.”  Knowledge is a form of orientation, finding one’s way in an environment.  For men and women of science, that means orientation in a discipline, a language, a type of laboratory, a style of experimentation, and so on.  There is no single, all-inclusive formula for such activities.  It is a question of immersing oneself in the detailed history of some particular scientific enterprise, and, it is to be hoped, gaining philosophical insight from that study.

Richard Burian’s study of the work of Jean Brachet may serve as an example of this kind of work (Burian 1997).  Burian examines in some detail the exploratory work of Brachet and his colleagues on the localization of nucleic acids and part of the pathway that led him and his coworkers to consider the problem of protein synthesis.  Burian writes:

The tools he devised, appropriated, and adapted, were put to different uses in the service of a variety of problems, explored in parallel.  In all of these studies, he sought to employ a wide range of organisms and the widest possible range of techniques in order to provide a basis for reconciling the findings obtained and for achieving broad understanding of the process in question.

Burian 1997, p. 32

This wide-ranging style of experimentation, based in biochemistry and biochemical embryology, and using techniques from a variety of sources


on a variety of organisms, contrasts strikingly with those of the more celebrated pioneers of molecular biology, who “typically employed tools imported from physics in combination with the new methods of genetic analysis applied to microorganisms” (Burian 1997, p. 39).  Yet, by difficult and devious paths, these different styles of investigation converged on the same entities.  Brachet’s methods themselves shed important light on one variety of scientific practice, in which problems and techniques from a number of disciplines are brought to bear on a developing field of investigation.  But this story also suggests a broader lesson.  Comparing Brachet’s research with that of other workers engaged in the same search, Burian observes:

The very fact that work in three such different styles (and with aims as diverse) as Brachet’s, Crick’s, and Zamecnik’s could be brought into concordance in about one decade serves as an important marker of the mobilization of the entities and phenomena in the enormous domain covered by these three distinctive research programs.  One of the most important tasks in the philosophy of biology - indeed, the philosophy of science generally - is to understand how we achieve concordance in the interpretations of the findings of workers who, in [Hans Jörg] Rheinberger’s terminology, work with such different ‘epistemic objects’ as those that preoccupied these three key figures.  It is in handling such topics as this that the philosophy of experiment will find its liberation from the excessive theory-centrism, now waning, of recent philosophy of science.

Burian 1997, pp. 40-41

Similar case-based philosophizing can also be found elsewhere.  Almost classically by now, there is what Fleck did with the Wassermann test, deriving from that history his concepts of thought style and thought collective.  Hans Jorg Rheinberger bases his philosophy of experimental systems and epistemic things on a detailed study of protein synthesis in Paul Zamecnik’s laboratory.  On a more general scale, but still through the study of particular cases, Jean Gayon also furnishes philosophical insights through his account of the destiny of Darwin’s hypothesis of natural selection (Gayon 1998).  In such studies, history is fundamental.  Just what were the disciplinary backgrounds, interests, techniques involved in each case must be carefully described and analyzed.  Even when the activity under study is contemporary, it is still history, like Thucydides’ Peloponnesian War - or better perhaps, especially for readers of this book, like “natural history” from Aristotle to


Cuvier - that is, the study of life styles in their concrete distinctness from one another. [1]

So far, so good.  But to say that our work is like that of an ethologist is not enough.  Ethologists have different ways of approaching their subject matter.  Some have been radically reductive in their methods, like Fraenkel and Gunn (Frankel and Gunn 1940).  It is simply motions they claim to be studying.  As we have indicated, it is more than such purely externalist description we are after.  On the other hand, there are ethologists, such as Cheney and Seyfarth, who want to penetrate if they can “inside the mind of another species” (Cheney and Seyfarth 1990).  Following Thomas Nagel’s famous inquiry, “What is it like to be a bat?” they want if they can to penetrate the subjectivity of their object of study, in this case vervet monkeys (Cheney and Seyfarth 1990; see Nagel 1974).  A similar intention underlies the work of Donald Griffin, whom Cheney and Seyfarth cite approvingly (Griffin 1984; 1992).  Moving from a vain hope for total objectivity, such writers turn to a search for subjectivity, for a “secret inner something,” to set against purely external, or purely material, appearances.  However that seems to us an equally vain hope - Cheney and Seyfarth certainly admit its difficulty - and, when carried to extremes, a foolish one.  Thus Griffin, for example, wants bees to be conscious of their directed food-searching, as we are when we go to the supermarket.  Surely the evolution of the central nervous system has made some difference in the nature of animals’ experience.

In short, if we are not looking for pure locomotor descriptions, neither is it inner feels we are after in our reflections on the sciences.  However, there is, we believe, a third, more promising, way to go.  Another name for ethology is behavioral ecology (Cheney and Seyfarth 1990, p. 10).  We are trying to understand certain ways of coping in, and with, certain environments.  If we read it in terms of the principles of ecological psychology, Cheney and Seyfarth’s subtitle, “How Monkeys See the World,” may suggest a direction for such investigations.  Traditionally, “seeing,” and other perceptual systems, have been described in terms of isolated, private sensations from which we somehow infer hypotheses about what is out there beyond us.  The relatively new discipline of ecological psychology takes a different approach. [2]

1. Clearly there is no reason why careful case studies should be limited to biology.  For example, Friedrich Steinle considers Charles Dufay’s work in electricity as a case of exploratory experimentation similar to the Brachet case treated by Burian (Steinle 2001).

2. The basic text is J. J. Gibson 1979.


Whatever any animal perceives entails three equally essential aspects.  First, there are always things or events occurring in its environment.  Again, we are considering real animals in real situations.  No science fiction or possible world nonsense about it.  Second, there is always information in that environment that the animal has the capacity to pick up.  This takes the form of invariants: constancies within change which the animal’s perceptual systems have evolved to be able to pick up.  On the perceiver’s side, such information pick-up consists in a process of differentiation within a structured context.  It is important to stress this feature.  On the older view of perception, it was a question of somehow associating together small, meaningless sensations.  In terms of the ecological approach, on the contrary, perceivers, from early infancy on, are understood to be discriminating between, or differentiating, distinguishable features within their environment.

Third, there are affordances, opportunities the environment affords the perceiver, or dangers it presents.  It may offer the chance of a mate, the threat of a predator, and the like.  “Affordance” is a coinage of J. J. Gibson’s, which, he says, he substituted for the term “value,” with its subjective connotations (J. J. Gibson 1966, p. 285).  It is not a question, as Köhler put it, of “the place of value in a world of facts.”  The world itself exhibits values, or meanings: relations between perceivers and features of their environments that offer them goals to seek or avoid.  An animal’s world is, from the beginning, a world full of meanings, and evolution has endowed it with the potentialities to respond to such a world.

In other words, there is a real world in which the animal finds itself, and to the constitution of which it in turn contributes through its activity.  This world is structured: In the flux of events there are constants, invariants, stable proportions that characterize for the terrestrial animal, for example, the ground, the horizon, and so on.  Again, it is these invariants that constitute the information that the animal picks up in its environment.  And it is these same invariants that allow the animal in its turn to perceive affordances - that is, the advantages and the dangers that its environment presents to it. [3]

This situation exists for all animals.  However, there are also peculiarities, so far as we can tell, in the human situation.  Although perception as such is direct, there are also three kinds of indirect perception mediated

3. We are here following an account by Eleanor J. Gibson (E. J. Gibson 1983).  See also Gibson and Pick 2000.


by our cultural inventions.  (They are nevertheless still to be found in the real world, in nature; there is only one world, within which culture arises.)  These three inventions are tools, language, and the use of pictures.  From birth, the perceptions of the infant, then of the child and of the adult are saturated by these human and cultural ingredients.  But the fundamental structure of perception remains the foundation of these accomplishments - and, as we shall argue, if not the foundation, at least the analogue, of all knowledge (see J. J. Gibson 1979).

What lessons does this new theory of perception offer to the philosophy of science?  We believe there are three.

First, the ecological approach assists us in maintaining our realist position.  We can definitively finish with the phenomenalism that has haunted the philosophy of science since its inception.  We can finally forget the picture of Mach counting his sensations, and try to understand the situation of scientific workers as engaged, each in his or her discipline, in an ongoing dialogue with reality. [4]  We need to see ourselves, and the scientists among us, each in his or her own discipline or emerging discipline, as live, if enculturated animals in a complex, ongoing environment, full of meanings, whether soothing or startling, beckoning or alarming.  The traditional theory of perception set the observer, let alone the thinker, apart from his milieu, isolated with a congeries of meaningless sensations that had somehow to be correlated with expectations in a constructed, perhaps fictional, “out-there.”  This has surely contributed a great deal to the difficulty of understanding scientific practice and formulating a more concrete and more adequate picture of such practice.

Second, if we adopt the ecological point of view, we find that perception plays an important part even in our more cerebral, or more enculturated, knowledge.  Consider the three categories of indirect perception that characterize human knowledge: tools, language, and the use of pictures.  Up to the beginning of the computer age, tools served chiefly to improve perception.  Hooke looked at his cells, and Leeuwenhoeck at his little animals.  Physicists still have to look at the traces in a Wilson cloud chamber.  We never wholly escape the basic need

4. We owe this expression to Dr. Frank Quinn of the Virginia Tech mathematics department.  Michael Polanyi spoke of the scientist’s confidence of being “in contact with reality” (Polanyi 1958).  The concept of a dialogue further implies something like what Polanyi called “indwelling.”  The language in such a dialogue will itself be partly established, partly as yet developing.


of picking up information that allows us to perceive the affordances of our environment, be it in a laboratory or a field.  Perception is always the fundamental knowledge on which other knowledge rests.  Depiction clearly also involves perception, if, as J. J. Gibson, describes it, a kind of double vision:  We are always aware both of picture and frame (J. J. Gibson 1979).

The case of language seems at first sight more difficult, since there is a long tradition that separates language from reality.  Yet language does not separate us radically from the essential process of perception.  Rather it enriches it; naming draws our attention to objects.  Moreover, language itself has to be perceived: heard, seen, or in the case of the blind, touched.  It is true that when the child begins to speak, it enters into a world not previously known, and from that moment its perceptions are caught up in its linguistic life.  This is an infinitely subtle relation, and difficult to analyze.  But it does not remove the foundation of all human activity in the perceptual processes by which we conduct ourselves in the world, at once cultural and natural, that surrounds us.

Finally - and this is the most significant implication of the ecological approach for the philosophy of science - the three components of the perceptual situation hold as well, analogically, for more cerebral kinds of knowledge.  The term “perception” is used metaphorically for forms of insight other than strict sense perception, and with good reason.  In his induction into any discipline, the student finds himself in a new world, surrounded by events and objects formerly unfamiliar.  What he learns, in this new ambiance, is to pick up information in the form of invariants, and with their help to perceive the affordances - the meanings - available in this new environment.  A beginning medical student, observing an X-ray, sees only lines; he learns to read these lines as an infected lung or a fractured limb or whatever.  This is still a question of sense perception; but a similar process of increasing awareness marks the initiation into any new discipline, or sub-discipline.  There are objects and events from whose constancies over change we pick up information that allows us to grasp formerly hidden meanings within the new world we have now come to inhabit.

This is very like a child’s perceptual learning - mediated by language as well as tools and depictions - except that with scientific exploration, the process remains open-ended.  While some features within the discipline become routine, it is the still not-so-clear features the scientist is always groping for.  As François Jacob put it, “Unpredictability is in the nature of the scientific enterprise.  If what is to be found is really new,


then it is by definition unknown in advance.  There is no way of telling where a particular line of research will lead” (Jacob 1982, p. 67).  There is no single, over-all algorithm for such a process; styles of investigation will differ with the context.

In every discipline, however, or as Rheinberger puts it, within every experimental system, there will be some procedures, some entities, that have become sufficiently established to serve as technical tools in the next stage of the investigation, while other entities or relations are still unclear.  Rheinberger calls the upshot of this process “the production of epistemic things” (Rheinberger 1997).  That is a puzzling phrase, since we usually think of scientists as discovering things, not making them.  But what they make, presumably, are objects of knowledge: When we seize on the meaning of a phenomenon previously unknown, we are making into an object (for us) what was previously at most a shadow, foreshadowed in our search, but not “objectified.”  In a way, this is a Kantian insight, stressing the role of the knower in knowledge, but certainly without the sweep of the Kantian principles.  Again, there is no overall rationale to be found here; we are restricted in every case to a given historical context, in a way that goes far beyond the dreams - or better, the nightmares - of the sage of Konigsberg.  Still, through careful case studies, we can gain insights into the ongoing practices of the sciences in ways that are philosophically revealing.  Abandoning the goal of a grand overall synthetic view of science, or of “the scientific method,” we can strive, like scientists though in a different style, to find ourselves, as we hope, engaged in a dialogue with reality.

Moreover, from careful case studies and carefully limited generalizations gained from comparing them, we can gain insights into the ongoing practices of the sciences that are philosophically revealing.  To take an example from an earlier chapter, we found Cuvier insisting that only the careful study of each animal for its own sake was worth pursuing, whereas his colleague Geoffroy thought the search for large underlying generalities would set comparative anatomy on a new and more fruitful path.  Yet by the middle of the nineteenth century, their opposing theses were taken as complementary rather than contradictory models for biologists’ work.  Contemporary investigations such as those of Burian or Rheinberger, which we have mentioned, also illustrate the variety of differing experimental systems, as well the possible interactions among them.  Such interactions can be fruitfully studied without the old insistence on an ultimate unity of science of a monolithic and reductive sort.  The growth of new disciplines also illustrates such interactions,


as, for example, in the current development of bio-informatics, which combines the insights and methods of cell and molecular biology, mathematics and computer science, and statistics to achieve an understanding of gene structure and function that biologists would not have attained without the help of their neighbors - who in turn must learn something of the material their expertise is called in to interpret.