:: Article

without concepts

Edouard Machery interviewed by Richard Marshall.

Edouard Machery is a killer cool philosopher working on the cutting edge of interfaces between analytic philosophy, psychology, xphi and cognitive science. He’s a continental doing analytic philosophy who thinks philosophy without science is blind. He’s always investigating social phenomena like racism and the ‘integration challenge’, alongside the nature of concepts and whether they are the same as perceptual representation. This month he’ll be going head to head with the chillin’ blue-haired philosopher Jesse Prinz in Latvia on this very issue. He thinks concepts aren’t a natural kind and kind of thinks that studying them is like studying a science of Tuesdays. He’s also brooding on what the folk think and whether experts have judgements that can be trusted, suggesting that philosophy needs to be humble. Everything he does goes to the heart of how we think about ourselves and all in all is one hell of a badass groove. Shakin’.

3:AM: How did you become a philosopher? Were you always thinking about philosophical stuff or were there experiences that brought you to the subject?

Edouard Machery: I am French, I grew up in France, and I did most of my studies in France (at the École Normale Supérieure—where Sartre, Merleay-Ponty, and Foucault studied and where Althusser taught—and at the Sorbonne). In France, philosophy is taught during the last year of high school. In high school, I was very much into math, and I wanted to become a mathematician. When I encountered philosophy, I discovered a form of rigor that was distinct from mathematical rigor, but that was nonetheless genuinely rigorous. It was also harder than math, more precisely than the kind of math we were taught in high school, and it had a broader significance. In brief, I fell for it, and I specialized in philosophy at the École Normale Supérieure.

I became somewhat disappointed with philosophy as it was taught in France, which was way too historical for my taste. American philosophers sometimes think French students are taught a huge amount of so-called continental philosophy, but in fact Derrida, Deleuze, and others are not central to the teaching of philosophy in French departments of philosophy. Fortunately, I think. Phenomenology and history of philosophy are the bread and butter of a French philosophical education. In any case, philosophy felt stale, and I was longing for the energy that was animating the French philosophy of the first decades of the 20th century or the debates between Sartre and Merleau-Ponty.

By accident, I discovered analytic philosophy, which rekindled my philosophical desire. The rigor as well as the commitment to clarity and argumentation that attracted me to philosophy in the first place are central to much (if not all) of analytic philosophy. I was lucky to get involved with the Institut Jean Nicod in Paris, where some of the best French analytic philosophers were already working, and I was also lucky to meet Steve Stich, who invited me to visit Rutgers and who exerted an extraordinary influence on my thinking and my approach to philosophy.

3:AM: I think one way that we can immediately see the importance of your approach to philosophy and cognitive science is by discussing your work on racism. Racism has traditionally been thought of as either a question of nature – roughly, the thought that we’re born to think in racial terms– or nurture – roughly, our culture, upbringing, environment constructs races, and that they don’t exist in nature. You took the two research traditions, the nature tradition and the nurture tradition, and combined them. Can you say something about why you thought this combined approach was important at the time and what difference such an approach has made on research into this? Has it been an approach that has been well received by those in the previously opposing camps?

EM: Many social phenomena, such as racism, have been studied by, one the one hand, cultural anthropologists, sociologists, and historians, and on the other hand, by biologists and by evolutionary-minded behavioral scientists (anthropologists and psychologists). Sadly, these two traditions have failed to engage with one another, and, as a result, our understanding of many social phenomena remains incomplete. In my opinion, it is uncontroversial that social and psychological phenomena like racism or morality result from evolved cognitive structures, whose understanding requires an evolutionary perspective, but that many of their properties are the product of contingent historical trajectories. Integrating the two explanatory traditions is what I called the “integration challenge.” In my view, the theory of cultural evolution provides a framework for this integration.


[Photo of Gil-White from his home page.]

Racism is a case in point. As I have argued, following in part Gil-White’s groundbreaking work, we have evolved a sensitivity to “ethnic markers” (roughly, to markers such as clothes, accent, etc., that indicate what cultural group one belongs to) and a motivation to interact preferentially with members of own our cultural group. Racism is a by-product of this evolved sensitivity and motivation, and it emerges when skin color and other physical properties trigger our sensitivity to ethnic markers.

This hypothesis is useful to understand the unity of a large range of social and psychological phenomena, whose fundamental identity has often been ignored, or even denied, by historians and cultural anthropologists. On the other hand, research in history and cultural anthropology is needed to understand the peculiarities of racism in different historical contexts.

This hypothesis and this integrative approach have been overall well received, and some anthropologists such as Ray Scupin have been looking for evidence of the universality of our sensitivity to cultural markers.

3:AM: So this willingness to merge different research programmes is very much part of your approach to philosophy. You are a well known member of what has been labeled xphi. Were you always looking to do interdisciplinary work, blending psychology and cognitive science with philosophy, or was it something else that got you into it, maybe the feeling that sticking to one discipline was too limiting for the questions you were raising?

EM: I got into a naturalistic approach to philosophy when I was writing my dissertation on concepts. I felt that philosophers of mind such as Peacocke and Fodor had misunderstood the point of the psychology of concepts, and had failed to say interesting things about it. I also thought, and I still think, that the debate about concept individuation in philosophy had reached an impasse, and I found in the psychology of concepts a way to broaden the range of issues of interest.

In addition, as I mentioned earlier, my philosophical orientation was very much influenced by Steve Stich. I can’t express my views on this matter better than him: “There are no proprietary philosophical questions that are worth answering, nor is there any productive philosophical method that does not engage the sciences. But there are lots of deeply important (and fascinating and frustrating) questions about minds, morals, language, culture and more. To make progress on them we need to use anything that science can tell us, and any method that works.”

3:AM: Now the big idea that you’re thinking about is the concept of ‘concept.’ It’s a particularly important area of research because concepts are usually thought of as the things or vehicles of our thought. The idea is that we can’t think without concepts, conceived in some way. Now you are not happy with that, but before we get to your ideas, I think it’ll be useful if the topography of the domain is mapped out for us. You wrote a seminal piece, ‘Two Dogmas of Neo-Empiricism’ that actually does that. So could you summarise the general options that you discuss there so that non-specialists get what the various positions are between different conceptions of concepts?

One of the most interesting questions about concepts is whether concepts and perceptual representations are of the same kind, as traditional empiricists such as Hume and neo-empiricists such as psychologist Larry Barsalou and Jesse Prinz would have it, or whether they form two fundamentally distinct kinds, as Fodor has argued. The first option has become influential in philosophy and in psychology, but I have expressed skepticism about it in a few places. Incidentally, this is one of the questions that will be debated in a conference that Prinz and I are organizing in Latvia in May.

3:AM: I guess that most of the sassy readers at 3ammagazine will have read Pinker but the big names on the divide between the neo-empiricists and those that oppose it are Jerry Fodor and Jesse Prinz. They both think concepts are required, but disagree in fundamental ways. Could you flesh out the schema you gave in the last question by saying what the big divide is between these. Can you give examples as to how differently they’d answer the same question? I ask this because when we look at your approach these two figures are interesting points of contrast.

EM: We can illustrate the contrast between neo-empiricists and proponents of amodal cognition such as Fodor by focusing on categorization. Suppose that you see a dog, and that you judge that it is a dog: you are then categorizing the object of your perception as a dog. According to amodal theorists, to do so you retrieve from memory a concept of dog, which is a representation that has nothing to do with the perceptual representation of the dog you are perceiving. It may be something like a word. According to Prinz and Barsalou, the concept of dog you are retrieving from memory is itself a set of perceptual representations of dogs, and you match these perceptual representations (consciously or unconsciously) to your current perceptual representation of a dog. So, for neo-empiricists, but not for amodal theorists, entertaining a concept is a form of imagining or simulating. Jesse Prinz and I discuss this issue in our Philosophy TV dialogue.

3:AM: Now all that is by way of throat clearing for your own theory. You depart from both these approaches in a subtle way. You deny that concepts are a natural kind. Your book ‘Doing Without Concepts’ elaborates this idea. You argue that we should do away with talk about concepts, which is really very radical. Some might say you’re throwing out the baby with the bathwater and the bath as well! Can you explain your idea?

EM: It IS a fairly radical idea! And not one everybody is happy with!
In any case, psychologists and philosophers of psychology often assume that concepts share many scientifically important properties, and that the goal of a theory of concepts is to identify these properties. In philosophical jargon, concepts are supposed to form “a natural kind.” So, psychologists and philosophers of psychology have developed various theories of concepts, and have defended their pet theories by undermining the competing theories. The take-home message of Doing without Concepts is that this “natural kind” assumption is fundamentally misguided, and that as a result many debates between psychologists and philosophers about what the right theory of concepts is are empty.

To make that claim, I review in great detail and I assess the psychology of concepts of the last 40 years. Many good responses have been written in response to this idea and to the evidence I put forward, but I still find my views compelling.

3:AM: I asked about Fodor and Prinz because in a way these two are representative of different approaches to concepts that your position challenges. In a way you say Fodor is talking about something orthogonal to your concerns, but Prinz is being directly contradicted by your view. Is that right? Can you say something about this and the general significance for philosophy of mind that your approach brings to the table.

EM: This is correct. The question Fodor is asking, viz. “In virtue of what are we able to think about the objects of our thoughts as such (e.g., about dogs as such, about water as such),” is orthogonal to the question I am asking, which focuses on the bodies of information that determine the course of cognitive processes such as the processes underlying categorization or induction. In contrast, Prinz’s views are at odds with mine, since Prinz’s neo-empiricism amounts to an empirical hypothesis about the nature of these bodies of information. Prinz has a great discussion of the contrast between our views in Can Concept Empiricism Forestall Eliminativism?.

On my view, the mind turned out to be much more complicated and to have a much more baroque structure than philosophers of mind traditionally assume. We typically have many distinct ways, partly disconnected, to think about the same thing (dogs, water, presidents, visit to the dentist), and many words turn out to be polysemous, even if we are not aware of their polysemy. I have tried to provide some empirical evidence for this latter claim with Selja Seppälä in an essay Against hybrid theories of concepts, and Josh Knobe and Sandy Presada have a recent paper in Cognition Dual character concepts and the normative dimension of conceptual representation that fits with the gist of my views.

3:AM: Now when we talk of natural kinds we contrast them with non-natural kinds. So water is a natural kind and science can investigate it, but Tuesday isn’t. So we are used to thinking that there can’t be a science of Tuesdays. But you’re saying concepts are like Tuesday. So could there be a science of Tuesdays, just as there is of concepts? I guess what I’m wondering is whether your approach to concepts really does change how we think about how we think in pretty significant ways.

EM: Yes, that’s a nice way to put my views. A science of concepts would be like a science of Tuesdays. As you can imagine, not all psychologists are thrilled!

But this view has a silver lining for psychologists. If I am correct, there are a bunch of exciting empirical questions that have been ignored by psychologists, and that should be tackled urgently. These include, How are the concepts organized? Do some concepts have priority over others? How are resulting conflicts resolved? Are they triggered in different contexts? And what is the relevant mechanism? How are different types of concepts acquired?

3:AM: As with many of the xphi community you are interested in investigating folk theories of various phenomena. In a paper with Jonathan Livengood you give a couple of reasons for this interest: firstly, ‘folk metaphysics ought to be dragged out into the open and exposed to criticism,’ and secondly ‘folk metaphysics ought to be studied , because metaphysicians often assume without evidence that they know what the folk think, and these assumptions are sometimes wrong in important ways.’ In that paper you were looking at causation by absence. So can you give examples of some of the things you have done to find out what the folk are thinking and how it isn’t what we think they do?

EM: For a large part of philosophy (e.g., for the kind of questions about concepts we just discussed or for whether human nature exists—one of my research interests), lay opinions or folk theories do not matter at all. But for other parts of philosophy, they do, in part because philosophers appeal to them.

Examining folk theories or judgments empirically may turn out to be surprising in more than one way. Sometimes, we find that philosophers just don’t have that good a grasp of what lay people think. This is what the paper with Jonathan Livengood tried to establish. We show that some claims made by David Lewis and by Helen Beebee about lay people’s understanding of causation by absence are mistaken (but see Dunaway, Edmonds, & Manley, The Folk Probably Do Think What you Think They Think for an empirical criticism of our work).

In other cases, philosophers assume that judgments about philosophical cases or thought experiments (what is often called, misleadingly, “intuitions”) are likely to be true or reliable. A huge amount of work in experimental philosophy casts doubt on this view, by showing that these judgments are influenced by irrelevant variables such as culture, age, order of presentation of cases, and so on. My 2004 paper in Cognition, “Semantics Cross-Cultural Style” (with Ron Mallon, Shaun Nichols, and Steve Stich), illustrates this approach. We presented participants in the USA and in Hong-Kong with cases inspired by Kripke’s famous Gödel case, and, as we had predicted, we found that Americans tend to have Kripkean intuitions (“Gödel” refers to the man originally called “Gödel”), while Chinese tend to have descriptivist intuitions (“Gödel” refers to the man originally called “Schmidt”)! This finding has now been replicated several times, including in Japan (by Jonathan Livengood and Justin Sytsma), My collaborators and I have also used various formulations of the cases to address some concerns with my original work. To give another example, David Colaço (a grad student at Pitt, HPS), Was Buckwalter, Steve Stich and I have recently shown that some important epistemological judgments (for the aficionado: judgments about fake barn cases) vary with age: Older people are much less likely to ascribe knowledge to an agent when she could easily have formed a false belief instead of the true belief she did form.

It is common to respond that findings on the vagaries of lay people’s judgments say little about experts’ judgments such as philosophers’ judgments (a response known as “the Expertise Defense”). This response has now been challenged experimentally. I have shown that intuitions about reference of linguists and philosophers are influenced by their theoretical commitments (Michael Devitt and I have been discussing this issue in a recent exchange in Theoria): Semanticists and philosophers of language, who are likely to have read Kripke’s Naming and Necessity, are more likely to have Kripkean intuitions than linguists who pay attention to the descriptions associated with words such as (sociolinguists, terminologists, anthropological linguists, etc.). Eric Schwitzgebel and Fiery Cushman (Expertise in Moral Reasoning? Order Effects on Moral Judgment in Professional Philosophers and Non‐Philosophers as well as Kevin Tobia, Wes Buckwalter, and Steve Stich in Moral intuitions: Are philosophers experts? have shown that philosophers’ judgments about trolley case and other moral cases are influenced by various biases. In brief, philosophers’ and other experts’ judgments too are influenced by irrelevant variables.

To be honest, we do not yet know how often judgments about cases of philosophical interest are influenced by irrelevant variables. To address this question, Steve Stich and I have recently been awarded a large grant by the Fuller Theological Seminary / Thrive Center in concert with the John Templeton Foundation for a project on “Intellectual Humility and Cultural Diversity in Philosophy: An examination of the extent and implications of cultural diversity in philosophical intuition.” The goal of this grant is to run the first large-scale, systematic empirical study of the diversity of intuitions and to assess the implications of this diversity for the practice of philosophy.

3:AM: So can you say something about the general significance of this work. For instance, do you think that the new ways of understanding who we are is leading to a pretty radical new view of what people are. Is our self image changing because of this philosophical work?

EM: Some of this work is surely changing our self-image, or at least it should! Many philosophers of mind such as Dave Chalmers and Ned Block are convinced that phenomenal consciousness is a real phenomenon in part (perhaps in large part) because its reality strikes them as introspectively obvious. Well, if phenomenal consciousness is really obvious, as Justin Sytsma and I have argued in Two conceptions of subjective experience, then lay people should have a concept of phenomenal consciousness: They should draw a sharp distinction between mental states that have a phenomenal character and those that do not. But we have shown experimentally that they don’t, which suggests that they do not have a concept of phenomenal consciousness. We take this to suggest that phenomenal consciousness is not obvious at all, and, if it is not, why would we believe it is a real phenomenon at all? If our argument is convincing (and there are many responses one could make—we discuss them in our paper), then we need to revise our view of the mind dramatically! There may be no qualia, there may be no phenomenal consciousness.

In addition, the project “Intellectual Humility and Cultural Diversity in Philosophy” may lead us to advocate for a greater humility in philosophy. The methods commonly used in contemporary philosophy may be challenged by the diversity we expect to find. So, for example, how should we investigate moral permissibility if judgments about permissibility vary dramatically across cultures? Why should we care about what philosophers call “knowledge” if billions of people in other cultures don’t value this particular epistemological condition? That’s the type of question that philosophers would have to face if our project is successful.

3:AM: Are there some things that from this new perspective that we should be thinking about more seriously than perhaps we have done. I’m thinking about the tv show ‘Fringe’ which imagines that research on the fringe of science (or pseudo-science!) into things like ESP, life after death and stuff like that has borne fruit! This stuff is often characterized as being nutty, because it is in conflict with what proper science tells us, but given that this is probably just a folk intuition, isn’t there a possibility that the belief that nutty pseudo science is nutty is just another bias of our cognition and so should be revisited?

EM: Mmmh. There must be room for mavericks and heretics in science, a point highlighted by recent research on the social structure of science (for instance in Ryan Muldoon and Michael Weisberg’s work). After all, Darwin’s views were once marginal. On the other hand, the difference between Darwin and ESP advocates is that the former made a compelling case for evolution, while decades of research done by the latter have failed to produce even vaguely suggestive evidence.

3:AM: Do you find yourself having difficulty with people now that you know that so much mental life is kind of strange? And doesn’t it make you worry that much of what we are is mysterious given that you argue that much of the important conceptual work is not open to introspection?

EM: One of the things psychology has taught me is that we are sometimes poor at predicting how we would behave, and I often express skepticism at my friends’ and acquaintances’ assured predictions about how they would behave in such and such situations. When I predict my behavior, I tend to assume that I would behave like most people, and I often try to determine what psychological research predicts about the relevant type of behavior. This sometimes leads me to predict that I will behave in particular way, while I feel, with great confidence, that *I*, in contrast to other people, will behave differently. Weird mind split.

3:AM: Looking forward, where are your investigations going next? You’re working in a very exciting field with some very smart people. If you were to predict the state of play in say a decade, what do you think we’ll know then that we don’t know now? And who are the people to watch out for (alongside yourself of course!)

EM: I am working on several projects at the same time in addition to the project on philosophical humility I mentioned earlier.

For a few years now, I have been involved in a very lively debate about how human nature should be reconceptualized in light of progress in biology, genetics, and psychology. Steve Downes and I have just published a reader on this question, Arguing about Human Nature , and we hope that an increasing number of philosophers will turn their attention to this question. I have further ideas about this question, which I hope to write in the coming years.

I am slowly writing a new book about the foundations of statistics and about methodology in psychology and the behavioral sciences. I am defending classical statistics (the type of statistics developed by R.A. Fisher and Neyman and Pearson) against various criticisms, arguing for what I take to be the only consistent interpretation of classical statistics,a nd proposing various inferential norms for scientists. The book mixes fairly abstract arguments and concrete proposals about particular norms of inference. I hope it will be of interest to philosophers and psychologists alike.

Somewhat related, I am also working on the methodology of cognitive neuroscience and neuropsychology, and I have recently published several papers on the topic.

In the philosophy of psychology, I have developed fairly heterodox views about what is known as “implicit bias,” and I am turning the talk I have been giving into an essay for a fantastic volume edited by Jenny Saul and Michael Brownstein.

In moral psychology, I have interest about whether the concept of morality is a universal or whether it is, as I suspect, a cultural invention. I am doing some empirical work to try to get at this question, and I hope to be able to tell a broadly Nietzschean story about the invention of morality. I have already argued in detail with Ron Mallon that, while our sense of norms evolved, morality itself as a distinct kind of norms is probably a cultural invention (see the chapter on the evolution of morality in John Doris’s fantastic Moral Psychology Handbook with OUP).

As for the people that you should keep an eye on, I’d like to put a plug for my former and current graduate students. Justin Sytsma is one of the most interesting experimental philosophers, and his work on consciousness is extremely influential. Jonah Schupbach has done some groundbreaking work on explanatory power and on inference to the best explanation, and he brings together formal and experimental methods in the philosophy of science. Elizabeth O’Neill, who is finishing her PhD with me, is doing some really exciting work in moral epistemology, Greg Gandenberger, another of my PhD students, is working on incredibly difficult issues on the foundations of statistics, and has developed a new proof of the likelihood principle, and Joe McCaffrey, who is just starting his dissertation, is scrutinizing the project of localizing cognitive functions in the brain, which is at the core of contemporary cognitive neuroscience. All these research projects are incredibly exciting.

3:AM: And finally, are there any books you’ve read whilst engaged in this fascinating work that have been illuminating for you?

EM: The Fragmentation of Reason and Deconstructing the Mind by Steve Stich have been very influential for my intellectual formation. I am also a huge fan of Fodor’s early books, The Language of Thought and Representations, and of Dennett’s Brainstorms. I re-read these books regularly.

My work on concepts was very much influenced by Paul Griffiths’s book on emotions, What Emotions Really Are. This book illustrated how philosophy of psychology should be done. The same is true of Shaun Nichols’s, John Doris’s, and Jesse Prinz’s books.

Boyd and Richerson’s Culture and the Evolutionary Process as well as their collection of articles (The Origin and Evolution of Cultures) played a large role in my attempt to meet what I called earlier the integration challenge.

Nisbett’s work on cultural diversity, The Geography of Thought, woke me from my dogmatic slumber. I used to ignore the role of culture in shaping our thoughts, and Nisbett and his colleagues’ research led me to revise my views (although I am not entirely convinced by Nisbett’ historical explanation of the differences his book reviews).

Finally, E.O. Wilson’s On Human Nature is a fantastic, though often erroneous, read. It has convinced me of the legitimacy of viewing human beings as one would view any other species, by identifying the modal, stable patterns of behavior or of thought. This remains a very provocative read.

3:AM: And finally, finally, the smart set at 3:AM always like a reading list. So can you give us your Top 5 books that we should be reading to understand better your field?

EM: I recommend the books I cited earlier in this article, but here are a few additional suggestions (more than five, sorry!):

Stich’s collections of articles (Collected Papers I and II) is a great read.

Knobe and Nichols’s anthology, Experimental Philosophy, is a must read if one is interested in experimental philosophy.

Culture of Honor: The Psychology of Violence in the south by Nisbett and Cohen also illustrates the importance of culture.

Doing Bayesian Data Analysis by Kruschke is the best introduction to Bayesian methods in statistics. Although I am skeptical of the call for statistical reform in psychology, psychologists should be better acquainted with Bayesian methods.

Finally, two fantastic handbooks: The Moral Psychology Handbook by John Doris and the Moral Psychology Research Group and The Oxford Handbook of Philosophy of Cognitive Science by Eric Margolis, Richard Samuels, and Steve Stich.


ABOUT THE INTERVIEWER
Richard Marshall is still biding his time.

First published in 3:AM Magazine: Friday, May 10th, 2013.