Accelerating developments in molecular biology since 1953 have strongly encouraged the advocacy of reductionism by a number of important biologists, including Crick, Monod, and E. O. Wilson, and strong opposition by equally prominent biologists, especially Lewontin, along with most philosophers of biology.

Reductionism is a metaphysical thesis, a claim about explanations, and a research program. The metaphysical thesis that reductionists advance (and antireductionists accept) is physicalism, the thesis that all facts, including the biological facts, are fixed by the physical and chemical facts; there are no nonphysical events, states, or processes, and so biological events, states, and processes are “nothing but” physical ones. This metaphysical thesis is one reductionists share with antireductionists. The reductionist argues that the metaphysical thesis has consequences for biological explanations: they need to be completed, corrected, made more precise, or otherwise deepened by more fundamental explanations in molecular biology. The antireductionist denies this inference, arguing that nonmolecular biological explanations are adequate and need no macromolecular correction, completion, or grounding. The research program that reductionists claim follows from the conclusion about explanations can be framed as the methodological moral that biologists should seek such macromolecular explanations.

Introduction

Recombinant tissue plasminogen activator (rtPA) benefits patients who have had an acute stroke, but a delay in treatment for over 3 hours raises the risk of intracerebral hemorrhage. Reperfusion of blood into an ischemic region, while preserving metabolic function, results in the production of molecules that may damage the injured tissue. The cerebral microvasculature is a major site of injury during reperfusion with a biphasic disruption of the blood–brain barrier (BBB) seen after reperfusion. Multiple factors have been implicated in the damage to the microvasculature by ischemia with reperfusion, including free radicals, blood products and proteases. Matrix metalloproteinases (MMPs) are a gene family of neutral proteases. Once formed and activated, the MMPs attack the basal lamina around the cerebral blood vessels, leading to the opening of the BBB. MMPs are induced in cerebral ischemia.

MMPs are secreted in a latent form that requires activation. Plasminogen activators (PAs) are serine proteases involved in angiogenesis, neuronal growth and regulation of other proteases through activation processes. Brain cells produce PAs in response to an ischemic injury. Urokinase-type plasminogen activator (uPA) is secreted by microglial cells in culture. Urokinase generates plasmin, which activates MMPs. Latent MMP-2 (gelatinase A) is activated by a membrane-type metalloproteinase (MT-MMP), which is activated by plasmin. Latent MMP-9 (gelatinase B) is activated by stromelysin-1 (MMP-3), which also requires plasmin for activation.

In this chapter I defend Quine's naturalistic epistemology [Quine, 1969], extend it and respond to its critics. In doing so I have borrowed freely from the work of philosophers who had no thought to defend or extend a Quinean naturalistic epistemology (hereafter QNE), and I have criticized the views of some exponents of (non-Quinean) naturalistic epistemology.

The main theses of QNE are well known: traditional epistemology, a largely or wholly a priori and foundationalist discipline, is to be replaced by an empirical inquiry which will be a “chapter” mainly of psychology. This compartment of psychology is to examine the relation between an epistemic agent's “meager sensory input” and its “torrential output” of descriptions of a three-dimensional world “in order to see how evidence relates to theory.”

Naturalizing epistemology is an inevitable consequence of Quine's rejection of the analytic/synthetic distinction, his repudiation of a difference between a priori and a posteriori and his attack on modality. These three commitments, together and separately, lead through a variety of arguments to the conclusion that philosophy and science are continuous with one another: the former is just very general and abstract scientific theorizing. Accordingly, a compartment of philosophy that deals with psychological states, as epistemology traditionally does, should be viewed as very general and abstract psychological theorizing. Epistemology must be continuous with psychology because philosophy is continuous with science. If science cannot be a priori, neither can epistemology. If there are no truths in virtue of meanings, no appeal to meanings can underwrite epistemological claims, nor can there be a range of necessary truths to which we have epistemic access, and for which an epistemological account must be provided.

In this chapter I want to argue that one particular science faces limits that do not confront other sciences, and that these limits reflect a combination of facts about the world and facts about the cognitive and computational limitations of the scientists whose business it is to advance the frontiers of this science. The science is biology, and the limitations I claim it faces are those of explanatory and predictive power. In the first part of this chapter I advance a contingent, factual argument about the process of natural selection which destines the biology in which we humans can take an interest to a kind of explanatory and predictive weakness absent in our physical science. I then go on to show how these limitations are reflected in at least two of the ruling orthodoxies in the philosophy of biology: the commitment to the semantic approach to theories, and to physicalist antireductionism.

If I am correct about the limits to biological knowledge, we must face some serious issues in our conception of what scientific adequacy and explanatory understanding consist. My claim is that biology is far more limited in its ultimate degree of attainment of scientific adequacy than are the physical sciences, because the only generalizations of which biology is capable will not provide for the sort of coordinated improvement in explanation and prediction which characterizes increasingly adequate science. This fact about biology reflects as much on the biologist as it does on the phenomena the biologist seeks to explain and predict. Were we much smarter, physics and chemistry would remain very much as they are, but biology would look much different.

The traditional intersection between the philosophy of social science and moral philosophy has been the problem of value-freedom: whether theories in social science unavoidably reflect normative judgments, whether they can be or are free of them, and whether either of these alternatives is a good thing or not, given the aims and methods of empirical science in general. Despite its importance, this is a debate to which little both novel and reasonable has recently been added.

But the recent resurgence of interest in “moral realism” holds out the hope of a new approach to this issue, one likely to lend support to the opponents of moral neutrality, support from a direction in which they could not have expected it – the “scientistic” approach to theories about human behavior. For once moral realism is embraced, the attractiveness of deriving moral claims from descriptive social theories becomes very great.

REALISM AND NATURALISM

Moral realism is not just the thesis that at least some normative claims are definitely true or false. It also requires that at least some of them are in fact true. A moral realist who held that all ethical claims are just false could hardly be said to have offered a defense of ethical theories and claims. But if at least some moral claims are true, it behooves the moral realist to show how we can know which ones they are, and what the evidence for them is. But providing the metaphysics and epistemology moral realism needs has long been a problem. As Sayre-McCord [1988, p. 13] notes, “the common (mistaken) assumption is that the only realist positions available in ethics are those that embrace supernatural properties and special powers of moral intuition.”

To some economists, evolutionary theory looks like a tempting cure for what ails their subject. To others, it looks like part of a powerful defense of the status quo in economic theory. I think that Darwinian theory is a remarkably inappropriate model, metaphor, inspiration, or theoretical framework for economic theory. The theory of natural selection shares few of its strengths and most of its weaknesses with neoclassical theory, and provides no help in any attempt to frame more powerful alternatives to that theory. In this chapter, I explain why this is so.

I begin with a sketch of the theory of natural selection, some of its strengths and some of its weaknesses. Then I consider how the theory might be supposed to play a role in the improvement of our understanding of economic processes. I conclude with a brief illustration of the problems of instantiating a theory from one domain in another quite different one, employing the most extensive of attempts to develop an evolutionary theory in economics. My pessimistic conclusions reflect a concern shared with economists who have sought comfort or inspiration from biological theory. The concern is to vindicate received theory or to underwrite new theory against a reasonable standard of predictive success. Few of these economists have noticed what the opponents of such a standard for economic theory have seen, that evolutionary theory is itself bereft of strong predictive power (see McCloskey 1985, p. 15).

Two things to note and set aside at the outset are the historical influence that economic science has had over evolutionary theory from before Darwin to the present day, and the profit that biologists have taken in recent years from developments in economic theory.

More than a generation of active analysis of the structure and foundations of genetic and evolutionary theory has produced more unsettled questions than it has generally accepted conclusions. Even such apparently straightforward matters like the definition of fitness and the cognitive status of the principle of the survival of the fittest remain vexed. More issues such as the number of levels of irreducible organization, the units of selection, are becoming more complex as research proceeds. As for the relations between molecular genetics, classical genetics and the rest of biology, the most widespread view attempts to combine a thoroughgoing physicalism with a resolute antireductionism. The controversy about whether there are biological laws remains with us, as do basic questions about the relation of models in population biology to one another and to some actual or possible theory.

All of these debates presuppose a series of distinctions fundamental to common sense, biology, and its philosophy: distinctions such as those between innate and acquired, organism and environment, genetic information sources and environmental informational channels, gene versus individual, individual versus group, development versus inheritance, replicator versus interactor. If these distinctions do not divide nature at the joints, then it is unlikely that debates carried on in their terms will illuminate much more than our conceptions about nature, as opposed to biological processes themselves.

A recent paper by Griffiths and Grey (1994) in effect sets out to undercut all these distinctions, and to thereby undercut all of the biology based upon them. In particular, their arguments threaten the nomological possibility of a subdiscipline of genetics – both as a separate study of heredity and a separate study of development.

In the Museum of Science and Technology in San Jose, California, there is a double helix of telephone books stacked in two staggered spirals from the floor to the ceiling twenty five feet above. The books are said to represent the current state of our knowledge of the eukaryotic genome: the primary sequences of DNA polynucleotides for the gene products which have been discovered so far in the twenty years since cloning and sequencing the genome became possible.

THE ALLEGORY OF THE PHONE BOOKS

In order to grasp what is problematical about the human genome project (HGP), I want you to hold on to this image of a stack of phone books, or rather two stacks, helical in shape. Imagine each of the phone books to be about the size of the Manhattan white pages, and that the two stacks of phone books reach up a mile or so into the sky. Assume that the books are well glued together, and that there are no gusts of wind strong enough to blow the towers down. The next thing you are to imagine is that there are no names in these phone books, or on their covers. Only numbers. We do know that each phone number is seven digits long and we know the numbers have been assigned to names listed alphabetically, but without the names we can't tell to whom a number belongs. Moreover, the numbers are not printed in columns down the pages that will enable you to tell where one phone number ends and the next begins. Instead of being printed in columns down the page, the numbers begin at the top left and fill up the begin at the top left

Following the model of The Origin of Species, the books I have written are each one long argument. The essays collected together and slightly revised here have not figured in any of these long arguments. But they do reflect a common theme: the implications of a broader naturalism and a more specific Darwinism for issues with which nonbiologists concern themselves. The oldest of them goes back only about ten years, and in each case I still find myself happy with their conclusions, if not altogether any longer satisfied with their expression and argument. Though written by a committed naturalist and one of Darwin's latter-day “bull dogs,” these papers give voice to the recognition that there are important limits to the power of these two inspirations to solve problems in philosophy, science and policy.

What is meant by ‘naturalism’ is something the first of these essays, “A Field Guide to Recent Species of Naturalism,” more fully recounts. But briefly, naturalism in latter-day philosophy is founded on a commitment, voiced initially by W. V. O. Quine, to let the sciences be our guides in epistemology and metaphysics. The renaissance in evolutionary biology and philosophers’ increased recognition of its relevance to human affairs have made a generation of naturalists into defenders and exponents of the theory of natural selection as naturalism's most informative guide. This is a conviction which I share. But I also recognize that naturalism leaves hostages to philosophical fortune: problems of justification that cannot be ignored. Detailing the structure of naturalism in the philosophy of science and how it deals with outstanding questions, especially of realism and antirealism, enables us to gauge its strengths and weaknesses.

The “trolley” problem was introduced by Judith Jarvis Thomson in two papers, “Killing, Letting Die and the Trolley Problem,” and “The Trolley Problem” [Thomson, 1986; see also Foote, 1967]. The puzzle raised by the “trolley” problem is that of finding a principle that will reconcile two strongly held intuitions: that in the case of a trolley hurtling down the tracks towards five innocent persons, it is permissible to shunt the trolley onto a siding, thereby killing one person so that the five may live; while in the case of five patients requiring body parts, it is impermissible to anesthetize an innocent person and transplant the needed organs.

The problem is serious because the moral principle that most obviously seems to underlie the “trolley” case, viz., it is permissible to kill one in order to save five, is the direct denial of the principle that seems to underlie the “transplant” case. Indeed, the two cases are often supposed to place sharply in contrast consequentialist and deontological moral principles. For, in “trolley” the greater good for the greater number seems to decide the case, whereas in “transplant,” the rights of the innocent trump the welfare of the patients. Like Thomson, other philosophers have sought a principle that will reconcile our intuitions about “trolley” and “transplant.” Optimally, such a principle should explain why there is almost universal agreement on the two cases – that killing the one to save the five is permissible in “trolley” and impermissible in “transplant” and justify the moral distinction that we draw between them.

THE CONSENSUS ANTIREDUCTIONIST POSITION IN THE PHILOSOPHY OF BIOLOGY

The consensus antireductionist position in the philosophy of biology begins with a close study of the relationship of classical genetics (Mendelism and its successors), to the molecular biology of the nucleic acids, and their immediate protein products. This study reveals that there are in fact no laws of Mendelian genetics to be reduced to laws of molecular biology, and no distinctive laws in molecular biology to reduce laws of Mendelian genetics, that the kind terms of the two theories cannot be linked in general statements of manageable length that would systematically connect the two bodies of theory; and that nevertheless, biologists continue to accord explanatory power to Mendelian genetics, while accepting that Mendelian genes and their properties are “nothing but” nucleic acids and their properties.

The first three of these observations serve to completely undermine the thesis once held in the philosophy of biology that Mendelian genetics smoothly reduces to molecular genetics in accordance with some revision of the postpositivist account of reduction. The last two observations have been joined together as “physicalist antireductionism” – so called because it attempts to reconcile physicalism – the thesis that biological systems are nothing but physical systems, with antireductionism – the thesis that the complete truth about biological systems cannot be told in terms of physical science alone.

Social and behavioral scientists – that is, students of human nature – nowadays hardly ever use the term “human nature.” This reticence reflects both a becoming modesty about the aims of their disciplines and a healthy skepticism about whether there is any one thing really worthy of the label “human nature.”

For some feature of humankind to be identified as accounting for our “nature,” it would have to reflect some property both distinctive of our species and systematically influential enough to explain some very important aspect of our behavior. Compare: molecular structure gives the essence or the nature of water just because it explains most of its salient properties. Few students of the human sciences currently hold that there is just one or a small number of such features that can explain our actions and/or our institutions. And even among those who do, there is reluctance to label their theories as claims about “human nature.”

Among anthropologists and sociologists, the label seems too universal and indiscriminant to be useful. The idea that there is a single underlying character that might explain similarities threatens the differences among people and cultures that these social scientists seek to uncover. Even economists, who have explicitly attempted to parlay rational choice theory into an account of all human behavior, do not claim that the maximization of transitive preferences is ‘human nature’.

I think part of the reason that social scientists are reluctant to use “human nature” is that the term has traditionally labeled a theory with normative implications as well as descriptive ones.