Handmaiden to the Science?
Re-engineering Philosophy for Limited Beings: Piecewise Approximations to Reality. William C. Wimsatt. xviii + 450 pp. Harvard University Press, 2007. $49.95.
Integrating Evolution and Development: From Theory to Practice. Edited by Roger Sansom and Robert N. Brandon. xiv + 334 pp. The MIT Press, 2007. $34.
Evidence and Evolution: The Logic Behind the Science. Elliott Sober. xx + 412 pp. Cambridge University Press, 2008. $29.99.
Half a century ago, philosophers of science paid little real attention to the nitty-gritty reality of what happens in the laboratory or field. Idealizations ran rife; the only subject considered was physics; and the only tenable philosophy was a logic-intoxicated form of positivism. But in the 1960s—thanks to people like Thomas Kuhn, who insisted in The Structure of Scientific Revolutions (1962) that philosophers must take seriously the real life of science—the field began to welcome those who took pains to study and understand the actual practice of empirical inquiry.
No area saw a greater sea change than the philosophical study of biology. The 1950s and 1960s were good to the science itself, what with the discovery of the double helix at the molecular level and the growth in confidence of evolutionary studies at the organismic level. Looking for areas to conquer, philosophers of science turned to the life sciences, with aims beyond simply showing that organisms are more than just machines or that one must invoke vitalistic modes of understanding to capture the mystery of the organism.
The first biennial conference of the (American) Philosophy of Science Association, held in Pittsburgh in 1968, marked the first real awakening of this new way of doing things. I was in attendance, one of a small group who were determined to look conceptually at the life sciences and make the philosophy of biology a genuine and worthwhile branch of the larger discipline.
Looking back now, I see that beneath the excitement was a troubling issue that remains with us today. If you are an empirical scientist, the reasons for what you do are fairly obvious: You want to understand the nature and the workings of the world of experience. But what if you are a philosopher, specifically a philosopher of science? What are you doing and why? Is your job primarily to aid science, to be a kind of high-powered theoretician? In the words of the great British empiricist John Locke, is your task that of being a "handmaiden" to the sciences? Or do you have aims of your own, and if scientists do not care for what you are doing, tant pis ? Should your primary focus be on the traditional problems of philosophy—epistemology (theory of knowledge) and ethics (theory of morality)? Given the heated nature of the discussions that philosophers of science have over beers at the end of the day, probably there is no resolution to this problem; different people will simply have different aims. This diversity of goals is borne out by several recent books showing that although the seeds planted four decades ago in Pittsburgh still flourish and bloom today, there is still some confusion about what philosophers of biology should or should not be doing.
One person at that 1967 meeting in Pittsburgh was William (universally known as "Bill") Wimsatt, then a graduate student, who outtalked us all, had more energy than the rest of the group put together and knew more biology than seemed rather decent. Wimsatt's great abilities and enthusiasm were recognized shortly thereafter when he was awarded a postdoctoral fellowship at the University of Chicago, where he later joined the faculty. Particularly important in his intellectual development were Richard Levins and Richard Lewontin, who were both for a while at Chicago.
Wimsatt has been a strong voice arguing that the philosophy of biology exists primarily for the benefit of biology itself. For someone like myself, who is in the other camp—frankly, my dear, I don't give a damn what the biologists think—but who finds such intellectual differences stimulating, it is rather exciting that Wimsatt, on the verge of retirement, has recently published a long-awaited book. Re-Engineering Philosophy for Limited Beings: Piecewise Approximations to Reality is a volume based on much that he has written before, but it is more than simply a collection of articles. Rather, it is a new, overview take on things, with earlier material embedded within it.
The overall theme is that we need new ways of dealing with a messy world—ways that do not guarantee immediate truths but that help us to get by and to grope our way slowly to better modes of understanding. In the volume, which is aimed at achieving these ends, Wimsatt covers many topics. For example, in a very good paper titled "False Models as a Means to Truer Theories," he analyzes in detail an episode in the history of science—the ways in which geneticists found methods of mapping the genes on a chromosome—showing the messy but productive fashion in which the protagonists proposed limited models, ever refining their concepts until they achieved a reasonable grasp of reality.
A somewhat dated section titled "Reductionism(s) in Practice" takes up nearly half the book. Here Wimsatt goes over some of the debates about whether ideas or theories at one level can be shown to be consequences of ideas and theories at other levels. Many of the references are from the 1970s.
Finally, in an epilogue, "On the Softening of the Hard Sciences," Wimsatt encourages us to move from straw-man simplistic reductionism to messy complexity. At this point, more than anywhere, I sensed the influence of Lewontin, who as a Marxist has long decried the reductionistic tendencies of modern science and urged us to a more holistic, if more complex, view of nature.
To illustrate both the strengths and the weaknesses of Wimsatt's thinking, let me focus on the notion of heuristics , a concept that plays a big role in his discussions and is the subject of two of the book's four appendices. Paradoxically, it is not entirely easy to grasp the nature of the beasts, but heuristics are very much like biological adaptations—methods that help you to understand the essence of a problem and to overcome it, not necessarily perfectly but usually adequately.
Wimsatt notes that one very important thing about adaptations is that they often "serve to transform a complex computational problem about the environment into a simpler problem, the answer to which is usually a guide to the answer to the complex problem." For instance, many plants use the shortening day as a guide to the onset of cold weather. They could be mistaken—the weather could turn unseasonably chilly early in the fall, while the days are still relatively long—but by and large the trick is good enough for this world. This tactic is likewise the way of science. Scientists in many or even all fields make reductionistic heuristic assumptions, believing that complex situations are best treated by breaking them up and looking at the parts.
Wimsatt has nothing against this approach—indeed, he admits that it is a good way to go. However, he warns that we should realize that we are using heuristics and not a God-given tool for finding the truth. Heuristics can mislead. For instance,
commonly found in simple models of systems (and even in not-so-simple ones) is the assumption that the system is isolated (in effect, that it has no environment) or that its environment is constant in space and time. This asymmetry in simplifications is indicative of the kinds of biases induced by using reductionist problem-solving strategies.
By reminding us of this shortcoming, Wimsatt offers a kind of meta-critique of science, showing how, in disputes, differences are often misunderstandings based on hidden assumptions, ones unseen even by the protagonists.
To illustrate his case, Wimsatt refers to work done by his former colleague Michael Wade, a population geneticist. At the end of the 1970s, Wade plunged into the dispute between those who think that natural selection always works at the individual level and those who think that it can promote the good of groups at the expense of the individual. Wade's experiments on flour beetles suggested that group selection has a lot more power than many then thought, and Wimsatt digs into this finding—one that goes against the reductionism of making all causes work at the lowest or smallest level—arguing that the reason the individual selectionists were unprepared for the discovery was that they were making hidden assumptions that are "biologically unrealistic and incorrect," each of which "independently has a strong negative effect on the possibility or efficacy of group selection."
Wimsatt is making a really interesting point—despite irritating me in the process, as he does when he characterizes unfairly the supporters of individual selection. The people he names as partisans all had more nuanced views on the individual-group selection issue than he implies. The late John Maynard Smith, to take one example, openly agreed that group selection was probably responsible for the maintenance (not the origination) of sex. But I am sufficiently convinced by Wimsatt that I want to go back to the individual-group selection debate and rethink it in the light of what he claims.
Wimsatt really engages with his subject, so I am surprised at his rather gloomy assessment that "An alien power seeking to undercut our civilization could do no better than to teach most of our science and its history an philosophy as we do now." I cannot help feeling that this is the voice of Pittsburgh still speaking.
In my view, things have changed very much for the better, as evidenced by a new collection of essays to which Wimsatt himself contributes, Integrating Evolution and Development: From Theory to Practice , edited by philosophers Roger Sansom and Robert N. Brandon. One of the most exciting new areas of study in the biological sciences is evolutionary development ("evo-devo"). Once virtually ignored (by such great evolutionists as Theodosius Dobzhansky and Ernst Mayr, for example), it has now become one of the hottest subjects in the life sciences.
Philosophers have followed in the wake of the biologists' discoveries, and this collection shows really informed and subtle work being done on uncovering assumptions and pointing to new avenues of understanding. I particularly liked Sansom's essay on adaptability (a notion that is a great favorite of Richard Dawkins) and the essay by Paul Griffiths on the importance of evo-devo for evolutionary psychology. Griffiths shows beautifully how philosophy can take a broader view—linking development and behavior—and can point to connections that might escape the notice of the working scientist. In other words, despite Wimsatt's gloominess, I want to congratulate those philosophers of biology who have tried to be relevant to science on their solid successes.
But I also want to note that those oriented to more traditional philosophical issues, such as epistemology and ethics, have produced some terrific thinking. Elliott Sober, a philosopher of science at the University of Wisconsin-Madison, has long been a leader in this school, and his latest work, Evidence and Evolution: The Logic Behind the Science , shows why he commands our attention. He is interested in the question of evidence for theories, and he shows through a careful analysis of statistical thinking (particularly Bayesian thinking) how one can make informed decisions about claims made in biology.
Of particular interest is Sober's critique of "intelligent design" theory. He brings new ideas to the subject, particularly through the application of probability theory, so that one reads and learns things of value quite apart from the critique as such. But I am not at all sure that I agree with his argumentation. One of the most interesting questions about evolutionary biology is whether, before Charles Darwin, Alfred Russel Wallace and the discovery of natural selection, one could get away from the God hypothesis. Many (Dawkins, for instance) argue that the design-like nature of the world—the hand and the eye—calls out for an explanation, and Dawkins maintains that before Darwin it was impossible to be an intellectually fulfilled atheist.
This is not to say that people of that early era had not criticized inferences about God made on the basis of design. To this day, there is no more withering polemic than that of David Hume, who argues in his Dialogues Concerning Natural Religion that, if there is a God, given the pains of gout, the deity must be a pretty unpleasant chap. However, against Sober and with Dawkins, I am not convinced that Hume shows that there is nothing at all—or rather, Nothing at All. (Hume himself rather admits this at the end of the dialogues.) For my money, no natural selection, then no atheism. But I agree with Sober against Dawkins that, given natural selection, one does not at once plunge into atheism.
I am a lucky person. I was excited by philosophy of science four decades ago and remain so. I congratulate those like Wimsatt who think that the philosophy of biology should be a handmaiden to the science, and also those like Sober who show that there is more to be said than this. Both ways, there is terrific work being done, and we all should celebrate this fact.