There’s an old joke I tell sometimes when I try to explain philosophy to people outside academia. It’s about a drunk fumbling about under a street lamp looking for his keys. A well-meaning passerby asks him if he remembers where he dropped the keys. He points some distance down the street. “Then why are you looking here?” asks the passerby. “Because the light’s better.”

At the beginning of November, David Bourget and David Chalmers published the results of the 2020 PhilPapers Survey, a sequel to a previous philosophy survey they conducted in 2009. The survey aims to provide an overview of the state of the discipline by polling active philosophers on their stance toward a series of familiar philosophical questions and debates. The primary stated aim of the survey is sociological: to get empirical answers to the question of what philosophers think. The authors speculate that the survey results might also provide some indication of where the truth lies: “If philosophy has any tendency to converge to the truth, then philosophers’ views might provide some guidance about the truth of philosophical views.”

The 2020 survey consists of forty “main” questions as well as sixty “additional” questions of a more specialized nature. The questions usually offer two or three options, allowing respondents to state that they accept, lean toward, are neutral about, lean against or reject each option. For instance, respondents can state their preference for physicalism or non-physicalism concerning the mind, for communitarianism, libertarianism or egalitarianism in political philosophy, or for classical or non-classical logic. Some survey questions ask about particular philosophical dilemmas, like whether you should take one or both boxes in Newcomb’s Problem. Others ask in a more general way about philosophical methods or philosophical progress. All of the questions allow for a number of alternative answers, such as “Accept an alternative view,” “The question is too unclear to answer” and “There is no fact of the matter.” The survey collates responses from 1,785 philosophers working in B.A.-granting departments and who publish in English.

I have misgivings about this project but I should begin by acknowledging its virtues. The survey is a part of a broader project represented by PhilPapers and its associated websites. These websites are a tremendously valuable asset to the profession and the work that Bourget and Chalmers have put into establishing and maintaining them deserves everyone’s gratitude. The survey itself is an interesting document. As the authors note, “today’s sociology is tomorrow’s history, and these results may be of some use to future historians of philosophy.” To that I would only add that the survey itself is an interesting record of the sociology of philosophy in the early 21st century. That the survey exists, and that it exists in the form that it does, is revealing, as are the ways that the 2020 survey differs from the 2009 survey. For instance, the addition of questions concerning the philosophy of gender and race show a shift in what questions have the attention of the philosophical mainstream.

The survey makes some strong suppositions about what philosophy is and how it’s conducted. What philosophy is and how it should be conducted are of course themselves perennial philosophical questions, and the survey polls philosophers on these questions. But the very nature of the polling already encodes certain ideas about philosophical practice. Philosophy, the survey seems to imply, consists of a set of clear questions whose answers can usually be bundled into two or three options. Discerning which of these answers is correct is difficult. But that, the survey suggests, is where the philosophical action lies. Philosophers are in the business of developing arguments in support of one or another answer to a set question.

A lot of the philosophical work that I find most meaningful doesn’t take this form. Instead, it challenges the questions themselves. Wittgenstein maintains that this-or-that disputes over philosophical isms fall apart under pressure. For example, in both the Tractatus (5.64) and the Philosophical Investigations (§402), he argues that the attempt to articulate a thesis of solipsism doesn’t end up saying anything that distinguishes it from realism. Heidegger, especially in his later work, suggests that the impulse to grasp hold of answers bespeaks a lack of patience that inhibits genuine questioning. The proliferation of isms, Heidegger writes, attests to the expectation that all ideas should be easily publicly communicated and exchanged, an expectation that requires they be flattened and packaged for easy public consumption. Imagine Wittgenstein or Heidegger being asked to fill out the PhilPapers survey and you get some idea of my misgivings.

This picture of philosophy as formulating definite answers to set questions has critics closer to the Anglophone mainstream. Cora Diamond’s “Eating Meat and Eating People” doesn’t offer a novel answer to the question of whether it’s permissible to eat animals or animal products. It’s pretty clear that Diamond is deeply troubled by the eating of animals. But the main thrust of her paper is that the most familiar arguments against eating animals grossly misconstrue what’s at issue. John McDowell’s “Non-Cognitivism and Rule-Following” attacks ethical non-cognitivism, but the paper is hardly a call to vote for cognitivism instead. The paper tries to show that this way of framing the question is misguided. And a number of philosophers have challenged the very idea of thinking about ethics through the sorts of abstract decision procedures you find in the trolley problem. For instance, Martha Nussbaum and Iris Murdoch both insist that ethics is more a matter of perception than decision. Our ethical options aren’t laid out in advance for us like a fixed decision tree. The real ethical work, as they see it, lies in seeing a situation sensitively and responsively. On this view ethics demands something more like a creative act than a selection from a fixed menu of options.

The authors of the survey acknowledge the issue. They report: “Respondents from non-analytic philosophical traditions often reported feeling somewhat alienated by the questions, and even respondents from analytic traditions sometimes reported that the questions reflected a fairly traditional conception of philosophy that did not fully represent philosophy as it is done in 2020.” They add that they’re “especially sorry for bringing about feelings of alienation” and pledge to do better next time.

But it’s not clear to me that they grasp the nature of that alienation. They write, “We made some attempt to find questions from non-analytic traditions, but it was difficult to find candidates that enough of our target group would be familiar with.” To begin with, if your reason for excluding questions about marginalized traditions is, essentially, that they’re marginalized, the apology rings a little hollow. But the key issue, to my mind, is that this way of asking questions creates a skewed picture of what philosophy is. You won’t fix that issue by asking different questions.

Is there anything so wrong about asking a bunch of questions, though? The authors of the survey, after all, aren’t trying to tell anyone how they should do philosophy. And while PhilPapers is a key node in the philosophical profession, it would certainly be hard to prove that this survey actively shapes research agendas or hiring and promotion decisions. As the authors emphasize, they’re making a contribution to the sociology of philosophy. No substantive philosophical claims are intended or implied.

In fact, I don’t even think the authors mean to be characterizing philosophical practice in a certain way. But that’s precisely what’s interesting, and troubling, about the survey. The questions prod respondents to think about philosophy in such a way that many people—the authors presumably included—so take for granted that they don’t even notice that there’s prodding going on.

That prodding aligns with other forces acting on philosophy as a profession within the modern university. In order to find work and keep working, philosophers are strongly pressured to have a “brand.” They should be able to articulate clearly what subfields they work in and what approaches to what questions in those subfields they’re most engaged by. Some very good philosophers can do this comfortably so I don’t mean to suggest that there’s something wrong in itself with this orientation to your own work. But I doubt that the sort of self-presentation that makes you easily readable to hiring committees and grant committees correlates in any strong way with being a good philosopher.

The survey calls to mind a quotation attributed to the social scientist Daniel Yankelovich:

The first step is to measure whatever can be easily measured. This is okay as far as it goes. The second step is to disregard that which can’t be measured or to give it an arbitrary quantitative value. This is artificial and misleading. The third step is to presume that what can’t be measured easily isn’t very important. This is blindness. The fourth step is to say that what can’t be easily measured really doesn’t exist. This is suicide.

Bourget and Chalmers don’t mean to disregard, presume or dismiss anything. But the survey encapsulates a kind of snap-to-grid thinking in philosophy that encourages practitioners to prioritize what’s tractable over what’s valuable. The survey shines a light on the state of the discipline. But much of the important work lies in the poorly lit areas that don’t lend themselves to statistical measurement.

All of this risks sounding like sour grapes coming from someone who’s struggled to find a firm footing in the profession. And I suppose to some extent it is. But looking at the survey gives me the same queasy feeling I get when I look at the prompts on online dating apps. There’s this tremendously complex thing that matters a lot to me—and whose mattering is intimately tied to its complexity—and I see it reduced to a few neat data points.