Thoughts on current research in analytic philosophy

In the year 2024, 140 years after Frege’s Grundlagen, almost 100 years after Carnap’s Aufbau, and 85 years after Stebbing’s Thinking, in what state do we find the research produced in academic analytic philosophy? As I’ve said before, I think things are better than they’ve been for eighty years or so. There is still a great deal to do to rectify access to the discipline and improve its culture—though again things have improved significantly in these respects thanks to the initiatives of MAP, SWIP, as well as tireless work by other groups and individuals. And many of the academic departments that foster and fund much of the research itself are under threat, while the outlook for humanities in higher education more broadly is, to put it mildly, bleak. So it might seem a strange moment at which to celebrate the work coming out of this discipline. But in fact I think it’s important precisely because the value I see in current work is what these cuts—those already made and those sadly to come—are intent to throw away. So I think it’s important to say that the research produced is, I think, of very high average quality at the moment.

Mehrangarh Fort, Jodhpur, Rajasthan

Partly, that’s because analytic philosophers are more often drawing on, incorporating, and showing proper respect towards the insights of traditions with which they’ve less often interacted in the past, taking their arguments and frameworks into account in the way they’ve long taken scientific findings and arguments and frameworks into account: traditions such as classic Chinese philosophy, critical theory, black feminist thought, and many more. Partly it’s because those who got their doctorates in the past decade or so have brought a slew of new questions to the table. And partly, I would say, it’s because the work done using the methodology the discipline has always used and on the questions it’s always tackled is currently extremely well done.

It’s the third I’d like to talk about here, plus a little of the second. In particular, I’d like to talk about three considerations that are often adduced in favour of pessimism about the state of philosophical research: (i) the literature is chockfull of epicycles; (ii) analytic philosophy makes no progress; (iii) the incentive structure makes it impossible to introduce truly novel ideas. All of these are taken as evidence of a research programme in decline.

What is an epicycle? People usually use the metaphor to mean the sort of paper that takes a previous attempt to answer a question, notes that it doesn’t quite work, perhaps because it fails to cover a particular case that’s offered as a counterexample, and then proposes a new answer that is very closely related but incorporates some small amendment that allows it to account for that case. “Here’s Prof X’s account of love/art/knowledge/justice/adaptive preference/harm/the wrong of epistemic injustice; but lo, see how it founders on my ingenious counterexample; and yet lo once more, see how my equally ingenious amendment accommodates it.” That sort of thing. An interesting thing about such a style of publication is that it’s pretty common in other subjects, and in the ones of which I’m most aware it’s well received. In mathematics, we tend not to get counterexamples to previously published proofs, but we do get a series of gradual, sometimes small improvements on previously established theorems: “Here’s Prof Gower’s lower bound on this quantity; but lo, see my proof of this slightly higher lower bound of the same,” “Here’s Prof Tao’s result for this class of cases, but lo, see my slight generalisation to a larger class.” And, in the sciences, we very often find papers showing a previous theory fails to accommodate a new experimental result and then a small amendment is proposed to cover that result as well as all previous evidence. And the same in the study of history or economics. And in those subjects I think this is viewed positively, as progress. So why dislike epicycles in philosophy?

Sometimes, there is a very reasonable frustration you feel when you read a paper that presents a counterexample that seems so obvious that the original proponent of the theory it falsifies should really have seen it themselves and addressed it. There’s no doubt that can make you feel that journal space has been unnecessarily squandered. And of course, unlike scientists, we don’t need to do new costly experiments to come up with our counterexamples, nor like historians have to make time-consuming trips to the archive. So often it does seem that the epicycle on the original theory is needed, but should really have been foreseen and made by the original author. But this isn’t so common, and in fact I think is made less common by another feature that people tend to bemoan, namely, referee-proofed papers. While it might be rather tedious to read, papers that anticipate and try to explain away objections that first come to mind actually forestall a lot of the back-and-forths that people don’t like to see occurring in the pages of the journals.

I think another frustration arises when the series of epicycles is enormously long. Analyses of the concept of knowledge are the usual example here. I’m not hugely interested myself in the analysis of concepts for their own sake, and indeed I think knowledge should not have the central role in epistemology it’s often take to have. But I tend to think of the value of this literature as helping us identify what people take to be the sources of epistemic value. Is it just truth? It seems not because knowledge, surely the most valuable of all epistemic states, requires more than true belief. Is it then justification also? Again, knowledge requires more than that. So truth, justification, and some sort of modal conditions, such as safety and sensitivity? But, but, but. And so on. However, if that’s what this literature is getting at, it’s surely valuable! And, moreover, it’s little wonder it’s taking us a while to get to the bottom of it! Our patterns of valuing things are very complex! They’re unlikely to be fully mapped by a thousand or so people over seventy years! World is crazier and more of it than we think.

And I think the same goes for other debates that see many epicycles. Think of the debate over when adaptive preferences should be considered genuine preferences that others should respect and attempt to fulfil. Or the literature on what counts as a harm. These are profoundly complex normative and conceptual questions and it’s little wonder that answering them results in a lot of trial and error. That trial and error is sometimes undertaken by a single researcher as they pursue their own account, but it’s sometimes undertaken in an exchange that occurs in the literature. And why not? Journal articles are in any case just letters we write to the academic community to tell them where we’ve got to in our investigations at particular milestones.

Are all these epicycles interesting to read? Even to people deeply steeped in the relevant literature and deeply invested in the question, they are not, or not always. But I think we mustn’t expect every step along the path to answering important questions to be very interesting. It isn’t in the sciences. A question can be fascinating, important, substantial, profound, crucial while at least some of the components of its answer and some of the work we must do to discover that answer are not.

Of course, such epicycles will be even less interesting if you don’t find the questions they try to answer interesting in the first place. And that is fine! Philosophy is a subject remarkable in its breadth. From substructural logics through the interaction of aesthetics and ethics to the existence of abstract objects by way of the nature of oppression and the epistemology of testimony. Few will find it all interesting. We do have a real problem if the gatekeepers to the journals and monograph presses find only some subjects interesting or important, as for instance when much seminal feminist philosophy was kept out of the major journals, since part of the choice an editor makes is whether the material submitted to them sparks sufficient interest. And so it behooves editors to ensure they aren’t rejecting work that falls outside their own personal areas of interest.

Next up, progress: is philosophy making any? We certainly aren’t establishing many unconditional facts with certainty, as we might think mathematics is, nor even with high credence, as you might think biology or ecology or linguistics is. But then it’s hard to see how we ever could. What premise that could ground a philosophical argument attracts widespread assent? There are no undisputed foundational premises in our discipline. And yet I think we are making profound and substantial progress in understanding issues.

Take the area I know best, which is epistemic utility theory or accuracy-first epistemology. The idea is that the sole fundamental source of epistemic value is accuracy or proximity to truth, and all norms that govern what we should believe can be derived from this. In the last twenty five years we’ve come to understand a huge amount about which norms might be established in this way and which might not, and that in turn has shed light on these norms, because one comes to look differently at a norm of belief when one sees that it can’t be shown to further the goal of accuracy.

Or take another area about which I’m beginning to learn: welfare ethics. Building on Harsanyi’s groundbreaking theorem, and proceeding by adding what we might call epicycles, we’ve gained remarkable insight into what principles about group decision-making under uncertainty one must reject if one wishes to reject utilitarianism. We now know in much greater detail the costs incurred by egalitarianism and prioritarianism, for instance.

In neither of these cases—accuracy-first epistemology or welfare ethics—have we established any unconditional facts with any degree of certainty. It is always open to people to reject veritism and it is open to them to reject the principles that give utilitarianism, and it’s hard to see how we could ever show they are irrational for doing so. In the former case, though, we’ve learned what we get if we accept it, and in the latter we’ve learned what we must be prepared to sacrifice if we reject utilitarianism. And this is progress. It’s substantial progress, I think. It’s the sort of progress we can make on these questions.

Finally, novelty: is there enough? Perhaps different areas have different amounts, but in epistemology, I’d say there’s a great deal, much of it generated by the very people who are typically taken to be under pressure from the publication system to keep ploughing the same furrows as their predecessors, namely, people who completed their doctorates in the past ten years and don’t yet have tenure, if they’re in the US system. It’s awkward to name names because it sounds like you think those you don’t list aren’t innovative, so let me be clear that this is entirely off the top of my head and reflects more what I’ve been working on myself recently than any judgment about people I don’t list. But let me mention Julia Staffel’s work on degrees of irrationality and transitional attitudes, Rima Basu’s work on the moral dimension of beliefs, Georgi Gardiner’s work on attention, Jane Friedman’s work on inquiry, Kevin Dorst’s work on rationalising apparent irrationalities, Amia Srinivasan’s work on identifying political implications of what seem like purely theoretical epistemological positions, and almost anything Rachel Fraser writes, but particularly her work on narrative testimony. Many of these are opening up whole new research programmes. I can’t think of another ten year period since the 1940s in which epistemology welcomed so many new avenues for research.

So, in sum, I think research in epistemology is stronger than at most periods in the past eighty or so years. Going back further, of course, you enter another golden age—that of Carnap, Stebbing, Hosiasson-Lindenbaum, Ramsey, de Finetti, Russell, and the Wittgenstein of On Certainty. Naturally, I can’t speak for all of philosophy here. Perhaps epistemology is an outlier? It’s possible, but I doubt it.

I’m no fan of the current publication system and particularly its prestige ranking of journals and glacial pace of refereeing, which benefits those in the most comfortable positions. But innovative work is certainly published. Of course, perhaps the examples I give aren’t as radically innovative as others would like to see. Here, I think, we’d just have to look at examples, that is, ideas that are not getting out into journal or monograph publication. We’d need to get specific. And that’s difficult to do by its very nature, because the ideas aren’t in the public domain to be considered.