The Philosophy of Philip Kitcher

It would be an exaggeration to say that this handsome edited volume provides a full overview of Philip Kitcher’s oeuvre; there is no discussion, for a start, of his major works on James Joyce or Thomas Mann. But even limiting the book’s scope to his narrowly philosophical concerns, Kitcher’s phenomenal erudition creates unique difficulties for the editors. It seems likely that no single reader, excepting perhaps Kitcher himself, will be able to engage at a professional level with all the debates touched on here, by a stellar cast of contributors. But equally, the great majority of philosophers will find at least something of significant professional interest in this volume, which aims at comprehensive coverage of Kitcher’s philosophical work by touching on philosophy of science, philosophy of biology, philosophy of mathematics, political philosophy, ethics, epistemology, and philosophy of religion.

As a reviewer, then, one can’t really hope to discuss everything in the book knowledgeably. Like all those readers, I’m an interested but substantially uninformed onlooker for much of what goes on. Fortunately, the editors and authors have clearly recognised this danger, and pitched the entries accordingly. The philosophy of maths is not, I think, a topic on which many non-specialists feel comfortable, so it makes for a somewhat daunting start to the book. Gideon Rosen’s opening essay on ‘Kitcher against the Platonists’, fortunately, is a really exemplary case of writing with one eye on a wider audience than one’s subject would usually command, and the result is simultaneously a superbly readable introduction to the debate over the existence and nature of mathematical objects, and a serious critical engagement with the formalist take on the question that Kitcher has pursued over three decades. Though it’s the best such introduction I’ve seen, I can’t pretend that it equips me to judge the success of Rosen’s efforts to defend a ‘moderate Platonism’ against Kitcher’s charges. What I can say, though, is that the exchange between Rosen and Kitcher is the sort of philosophical writing we teach students to aspire to: collegial and constructive in tone, combative and robust in substance.

The next two papers concern technical issues in the philosophy of biology, a field to which Kitcher has made seminal contributions. The first is by Karen Neander, than whom as Kitcher notes in his response, ‘no contemporary philosopher has thought harder […] about the concept of biological function’ (p. 68). Neander begins by noting that ‘perhaps everyone who has thought much about the issue is now a function pluralist in some form or other’ (p. 45), but takes issue with the particular form of pluralism that Kitcher advances. On the one hand, Kitcher adopts something like an etiological theory, holding that something’s function is what that thing is designed to do. But this breaks down into a more and a less direct notion of design; for Kitcher, something may have an activity as its function either—as etiological theorists like Neander have claimed—because that activity is what it has been naturally selected or intended by its designer to do, or because it happens to contribute that activity to the naturally selected or designer-intended functioning of some larger complex system of which it forms a part.

This second option of a Cummins ([1975])-style ‘causal’ function is necessary, according to Kitcher, because the etiological account is ‘too onerous, ontologically and epistemically’ (p. 46) to be useful to physiology. That is, working biologists are rarely concerned with the evolutionary history of a trait, but with how it works. He gives the analogy of a piece of machinery, which operates only because of a particular connection made by a screw unwittingly and unintentionally dropped into its workings during assembly; there may be an interesting story about how the screw got there, but it’s simply not relevant to an operational analysis of the apparatus. However, as Neander notes, there are two ways to take this claim about ‘connection’. If the screw is connecting things by screwing them together, then the machine is working as it does precisely because the screw is doing what screws are designed to do; the operational analysis needs the aetiology. On the other hand, if the connection made by the screw is, say, an electrical one made in virtue of its metallic composition, it makes an entirely accidental and unintended contribution to the machine’s operation. So the question is whether we should say that the screw in this case has the function to make the electrical connection, and whether etiological analysis has anything interesting to say about this. On the sort of ‘ultra-strong’ etiological analysis Kitcher favours, nothing can be said; there is no selective reason for the connection to be made in this way rather than any other, and thus a non-etiological way of analysing function is also necessary. But etiological requirements need not, Neander points out, be so strict; a weaker, ‘middling-strong’ etiological analysis might require only that comparatively recent selection explains the presence or maintenance of this arrangement  over some of the available alternatives. If engineers copying the apparatus choose to keep putting screws in this place rather than soldered joints, then that is a fact of aetiology relevant to functional analysis.

It’s a subtler debate than I have summarized it here, pursued sensitively and generously by both parties. For me Neander has the better of it since, as she points out, causal accounts of function lack a normative element and so face notorious difficulty in explaining dysfunction. Since biological dysfunction is certainly something that interests physiologists, this poses a problem for Kitcher; in the cases where he thinks the causal account is necessary, it does a poor job of explaining their actual practises and interests, which he regarded as an essential requirement of a satisfactory theory. So if the ‘middling-strong’ theory can be sustained, it looks greatly preferable.

Another theorist who has developed the idea of ‘relatively recent’ selection as crucially significant for the concept of function is Paul Griffiths. Here he focuses on a different issue, the concept of biological information. Phenotypes uncontroversially require both genes and environment for their development; why, then, do we persist in talking as though genes carry a ‘recipe’ for the organism in some sense that the environment does not? For Kitcher, the answer is simply ‘the usual preference for overly simple, often monocausal explanations’ (p. 76), and the solution is what he terms ‘causal democracy’: the patient and careful giving to every causal factor of its due. Griffiths has previously opposed this view, holding with Susan Oyama and others that the discourse has conflated the statistical Shannon-information encoded in the genome with intentional information ‘about’ the organism. Here, however, he takes a different approach, building on more recent work he has done with Karola Stotz and others to develop a new ‘teleosemantic’ view of genetic information, resting on Ernst Mayr’s venerable distinction between proximate and ultimate explanation in biology. This view, he shows, helps to vindicate Kitcher’s advocacy of causal democracy. It’s a generous and constructive line of discussion, which is reciprocated in Kitcher’s response; together, the two provide a marvellously readable summary of a decade’s progress and convergence in the understanding of biological information.

This mutual generosity is similarly evident in the following three papers on general philosophy of science, which I shall not attempt to summarize in detail here, and Kitcher’s responses. Chapter 4, by Michela Massimi develops and defends Kitcher’s ‘real realism’ about scientific entities and theories; Chapters 5 and 6, by Jim Woodward and Michael Strevens, respectively, engage with Kitcher’s views on the unity and autonomy of the sciences, and the possibility of inter-theoretic reduction. These are detailed, productive exchanges between major figures, which bring influential debates right up to date, and will be important reading for many in the field. A great part of the success of these chapters—and the volume as a whole—lies in the tone of Kitcher’s replies, which do him nearly as much credit as the various encomia of his assembled peers. The commentary-and-response format depends greatly on the temperament of the responder, and can result in truly painful reading; thankfully, no such catastrophe is evident here.

The next three chapters, by John Dupré, Lorraine Daston, and Nancy Cartwright and Alexandre Marcellesi, engage in various ways with Kitcher’s views on—to quote the title of his ([2001]) monograph—Science, Truth, and Democracy. The problem can be put succinctly, if fatuously: ‘the public have had enough of experts’. Public policy needs to be informed by scientific expertise, if it is to be effective; scientific expertise must in turn sell itself to the democratic public if it is to justify the funding it needs. But the ‘social technologies’ of science and democracy have become dangerously out of kilter.

Kitcher offers an optimistic—though not, he is careful to say, a utopian (p. 246)—view of these matters; by striving to approximate an idealized democratic conversation between all parties, we could in principle hope eventually both to align scientific research with the reflectively justified preferences and priorities of the polis, and to align those preferences and priorities with what the best research tells us is actually the case, and actually likely to work. His interlocutors here all strike more sceptical notes. Dupré, examining cases like ‘racial science’, climate change, GM crops, and the anti-vaccination movement, finds it doubtful that any significant progress can be made by the methods Kitcher suggests. Daston, reviewing scientific history, questions whether the ‘disinterested’ value-autonomy of science really arose in the way Kitcher maintains, and can therefore play the role he claims it does in stifling properly democratic discussion of scientific objectives. Cartwright and Marcellesi discuss the harrowing Peter Connelly (Baby P) and Daniel Pelka abuse cases, showing how lazily ‘linear’ causal reasoning by policy-makers both leads to failures by the child protection services, and ensures that their lessons will not be learned.

If anything, and despite Kitcher’s spirited rejoinders, in the months since the volume’s publication these chapters no longer seem pessimistic enough. Each morning’s Fresh Horror from across the Atlantic bespeaks a regime revelling in the discovery that it can inhabit its own world of ‘alternative facts’ so long as it determinedly and actively disregards the normal scientific and journalistic fact-checking filters. Nor is anything so complicated or esoteric as non-linear causal models responsible for the disconnect between experts and electorates; polling around the ‘Brexit’ referendum showed that  Britain’s voting public simply refused to believe the overwhelming consensus among economists that leaving the EU would be financially ruinous.

Curiously almost absent from the discussion in these chapters is the role of the internet in all of this. Itself a gift of scientists to the public, it has radically democratized the flow of information; everybody, now, can be his own Gutenberg. But this profusion of publication has meant that the normal epistemic safeguards of the non-expert may now be counter-productive; in the sort of self-contained informational bubble that is now possible, consisting entirely of those who share one’s opinions, attempting to triangulate those views with others’, and to seek corroboration from what are regarded as authoritative sources, may simply create an unwarranted confidence in one’s existing prejudices. If democracy is indeed a ‘social technology’, then these developments have revealed a potentially lethal vulnerability, all too easily exploited by the cynical, which Kitcher’s imagined dialogues seem hopelessly inadequate to repair. In a trolling culture, where any attempt at civil engagement may simply be taking the bait, perhaps a ‘well-ordered’, values-driven science needs to find its figurative way into the business of punching fascists instead.

The final two essays are by Michael Smith on ethics and meta-ethics, and by Daniel Dennett on the philosophy of religion. Smith’s piece, which tackles Kitcher’s The Ethical Project ([2011]), is gently phrased, yet perhaps the most devastatingly critical in the whole volume. For his own stated purposes, Smith argues, Kitcher would be better served with a non-cognitivism than his professed pragmatism, or even with a Kantian view like Smith’s own; both positions that Kitcher rejects. Dennett’s essay is… a Daniel Dennett essay on the philosophy of religion. Few readers will be surprised at the contents; some will find it a bracing and fearless takedown of cant and obscurantism, others a strident and ungenerous polemic, whose pugilistic insistence on the bad faith of its targets has no place in academic debate. More, I think, will regard such a stark and unnuanced dichotomy as an unhelpful way to frame debates; they should appreciate the characteristically nimble and good-natured manner in which Kitcher manages to sympathize with Dennett’s aims, while yet resisting endorsement of his most more incendiary charges.

Inevitably, with a scope this broad, there will be fascinating issues that won’t be addressed; I would have been interested to see more, for instance, on The Ethical Project’s discussion of the ‘function’ of ethical reasoning in light of the debates around Kitcher’s view of the function concept. But the coverage is remarkably inclusive, and remarkably accessible; while few potential readers, as I said, will be professionally interested in everything here, I imagine similarly few will find any of the chapters uninteresting or incomprehensible. The editors and contributors—Kitcher himself included—have done a sterling job, and the volume stands as an impressive testament to a major figure.