Natural Categories and Human Kinds - and Islamic Philosophy

Interview by Richard Marshall

'I think that one can be a realist about natural kinds without being an essentialist. And one can be a social constructionist (in some sense) about some social kinds while still being a realist.'

'We can’t just specify a priori the conditions that natural kinds must meet and then look to the universe to see what (if anything!) satisfies our wishlist. We have to start by looking at the world and at our scientific investigations to see what plays a central role in our theorizing and then go on to reflect on what general features they have.'

'What I was trying to argue was that there’s a route from epistemology to metaphysics. In other words, we start by looking at which categories are doing epistemic work for us and then we use them to infer which kinds exist.'

'... scientific realism can be understood in terms of the claim that our theories and categories aim at capturing the causal structure of the universe, not as a commitment to the existence of a mind-independent reality.'

'When I arrived at Columbia in 1985, the director of graduate studies, James Walsh, worked in medieval philosophy and he tried to encourage me to specialize in classical Islamic philosophy -- but he was an exception, being a medievalist. At the time, I politely declined. In part, I didn’t want to reinforce the notion that “every fish is an ichthyologist” (as a friend of mine once put it), that is, if you have a certain cultural background or identity, that’s what you should focus on in your academic work. (The converse, which is a very simplistic version of identity politics, is even more problematic.)' 

Muhammad Ali Khalidi is a specialist in general philosophy of science, philosophy of mind, philosophy of cognitive science and metaphysics. Here he discusses the debate between essentialists and social constructionists, natural categories and human kinds, naturalism, the difference between metaphysical Realism and Essentialism, epistemic kinds, natural kinds outside of science, mind-independent reality, the revival of Islamic philosophy, Orientalism, Al Farabi’s paradoxical remarks on the democratic city,  Zaki Najīb Maḥmūd and his logical empiricism, and whether his arguments link with twentieth century Arab thought.

3:16:  What made you become a philosopher?

Muhammad Ali Khalidi: Like many people who are drawn to philosophy, I think I was interested in a lot of different things by the time I got to university: politics, probability theory, cosmology, theater, evolutionary biology, poetry, developmental psychology, calligraphy, and so on. I thought (naively but not entirely mistakenly) that philosophy might allow me to maintain an interest in at least some of those things simultaneously. I knew something about philosophy because I was privileged to grow up in an academic family with lots of books on the shelves of our apartment in Beirut, Lebanon. So before going to university I had read a few philosophy books, including Russell’s Introduction to Mathematical Philosophy and Ayer’s Language, Truth, and Logic. By the time I went to the American University of Beirut (AUB), I was a budding logical empiricist -- and partly for that reason, I decided to major in physics! The other reason I chose physics over philosophy was that the philosophy major had been suspended due to budget cuts during the ongoing Lebanese civil war. But despite the turmoil in Lebanon, the department was an active one with a number of excellent teachers, so I took several philosophy classes. I was lucky enough to take two logic courses with the Australian logician David Makinson, a course on the history of analytic philosophy with British-Indian philosopher Salim Kemal, courses in epistemology and philosophy of science with Canadian philosopher Ken Ferguson, and a course on Plato with the American philosopher Richard Scott. It was a philosophical all-star team from all over the Anglophone world. They were all inspiring in their different ways. Pitched battles were often raging around the campus but we would sit around a seminar table discussing the Meno one day and Frege the next. Maybe philosophy was a way of escaping into an abstract, rarefied realm, away from the realities of civil war. Also, growing up as a Palestinian in Lebanon it was hard not to be interested in questions of justice, self-determination, refugee rights, the ethics of war, and related philosophical issues. After graduate school and a few short-term positions, when I didn’t manage to get a tenure-track position in the US, I was offered a position at AUB. Most of my former professors had left by then but it was a thrill to go back and teach in the same classrooms and lecture halls where I’d sat as a student.

3:16: You’ve two big philosophical interests – how science categorises nature and Islamic philosophy. Let’s start with the first topic. It’s common to have the debate between essentialists and social constructionists - ‘a dialogue of the deaf’ you call it – in philosophy. So to map out the territory for us, can you scope out the usual debate so we can see why yours is a distinct position against both sides?

MK: First, I have to apologize for that offensive and obnoxious phrase, which I regret using in the book. Having said that, it’s true that there’s something of a disconnect in the philosophical dialogue, but I don’t think essentialism and social constructionism are the only options when it comes to natural kinds. One can be a realist without being an essentialist. And one can also be a social constructionist about at least some kinds while still being a realist, as I’ll try to explain.

I think that essentialism is a misstep in the recent history of philosophy. To the extent that the thesis has any empirical content it’s either demonstrably false or unmotivated. I also think it’s well established now that you can’t derive essentialism from philosophy of language without smuggling in essentialist premises – and I don’t think that we have good grounds for accepting those premises. As for social constructionism, a number of philosophers (Ian Hacking, Sally Haslanger, Ron Mallon, to name a few) have explicated the various meanings of social constructionism very capably, and without going into detail, there are clearly certain ways in which the kinds of the social domain are socially constructed and the kinds of other domains generally are not. I think that one can be a realist about natural kinds without being an essentialist. And one can be a social constructionist (in some sense) about some social kinds while still being a realist.

3:16: You call your position naturalist and you title your book natural categories and human kinds. We might have expected natural kinds and human categories. Why the switch around – and what is the claim you make when you call yourself a naturalist?

MK: Part of the reason behind the switch was that I was trying to undermine the usual dichotomy set up between the (natural) kinds that we discover in the world and the (human) categories that we invent to help us get by. It seems to me that when we try to understand the world, we aim to capture its causal structure, and to the extent that we do, we manage to identify natural kinds. Another reason for the switch in the title was that I wanted to emphasize the continuity between the “natural” and the “human,” or the similarities between the domains of the natural sciences and those of the human or social sciences.

As for naturalism, I think it means different things to different people. Maybe it’s been so over-used that it’s become an empty honorific (or a gratuitous slur). I think that one of the best in-depth philosophical explanations of what it means to be a naturalist is in an article by Philip Kitcher (“The Naturalists Return,” Philosophical Review 1992). But the easiest way I can think of to illustrate what I mean by naturalism in this context is by way of an anecdote. A few years ago I was at a workshop on natural kinds and one of the participants raised his hand during one of the discussions and said something like: “Forgive me for being an old-fashioned metaphysician but I always thought that philosophy should spell out the conditions that natural kinds must satisfy, and then we can turn to science to see what satisfies those conditions.” To me, that summarizes what’s wrong with non-naturalist approaches to metaphysics and to natural kinds in particular. As a friend of mine said to me when I recounted that story: “What if we come up empty-handed?” We can’t just specify a priori the conditions that natural kinds must meet and then look to the universe to see what (if anything!) satisfies our wishlist. We have to start by looking at the world and at our scientific investigations to see what plays a central role in our theorizing and then go on to reflect on what general features they have.

To some people, this may sound like an abdication of philosophy’s distinctive role or mission. It gives rise to what might be called the “naturalist’s dilemma”: naturalism seems to entail that philosophers musteither settle for pronouncing empty tautologies from the armchair or attempt to make contentful statements by getting their hands dirty alongside empirical researchers. So if naturalist philosophers want to avoid the first horn of the dilemma, are they just doomed to being junior partners in the scientific enterprise? Do we just turn into glorified lab assistants? It seems to me that there are plenty of theoretical questions to be addressed by philosophers who have one eye on science and take heed of empirical work without necessarily engaging in it themselves. Of course, there are also some very fruitful collaborations between scientists and philosophers. But even in the absence of such joint efforts, there is plenty of room for philosophers to make theoretical contributions to human knowledge without necessarily becoming scientists. Even philosophers working in ethics can’t afford to be unaware of research in some branches of psychology and sociology.

3:16: What’s the difference between metaphysical Realism and essentialism, and do either have a role to play in your approach to understanding categorization?

MK: They’re two positions that often go together in metaphysics, but I think they’re orthogonal. One has to do with the ancient debate about universals and particulars. Are there transcendent or immanent universals in the world, or does the world just consist of particulars? That’s a debate that I find difficult to adjudicate and I don’t have anything original to say about it.

As for essentialism, regardless of whether you think kinds are universals or just collections of particulars, you can say that members of a kind must satisfy a set of necessary and sufficient conditions to belong to the kind. You can also say that membership in a kind amounts to having one or more intrinsic properties. And you can further claim that kind membership pertains to members in every possible world in which they exist. These are all distinctive essentialist theses, though perhaps not all essentialists hold all of them. But I think each of these theses is problematic. I think we’ve discovered that paradigmatic kinds violate all of them, and that paradigmatic non-kinds satisfy some of them. Perhaps more importantly, I think that none of them is sufficiently motivated. At least, that’s what I argue in my book. As I said, I think essentialism represents a wrong turn in contemporary analytic philosophy, and it’s not compatible with naturalism. It may even be a reflection of some deep-seated cognitive biases, as some philosophers and psychologists have argued (see e.g. Sarah-Jane Leslie, “Essence and natural kinds: When science meets preschooler intuition,” Oxford studies in epistemology 2013).

3:16: You say that natural kinds are epistemic kinds don’t you? So what’s an epistemic kind?

MK: In using the expression “epistemic kind,” I was following the lead of philosophers of science like Paul Griffiths and Ingo Brigandt, who have used the expression “investigative kind.” In hindsight, the expression “epistemic kind” is a bit misleading because the notion of natural kind pertains primarily to metaphysics. What I was trying to argue was that there’s a route from epistemology to metaphysics. In other words, we start by looking at which categories are doing epistemic work for us and then we use them to infer which kinds exist. The inference is not automatic and needs to be qualified in various ways (for instance, some categories may be redundant or doing dispensable epistemic work) but in ascertaining the natural kinds, I think that there’s no substitute for starting from our investigative categories. This is one respect in which I think some contemporary philosophizing is misleading. I have in mind the common saying: “Don’t mix your epistemology with your metaphysics.” I think that, at least in this instance, one can’t do metaphysics without doing epistemology, and in doing so, some mixing is inevitable, or at least some zigzagging between the two. That’s at least part of what it means to naturalize metaphysics.

3:16:  Some will argue that your position is either too restrictive, others that it’s too liberal. What are the best versions of these responses and how do you counter them?

MK: It’s true, I get it from both sides. In response to the too-liberal challenge, I think it’s pretty clear that there is no principled reason to select some subset of kinds, like elementary particles, chemical elements, chemical compounds, and (perhaps) biological species, and say that they’re the true kinds in nature and everything else is just fictitious or arbitrary. There are no relevant discontinuities between these kinds and many others across the domains of the geological, biological, cognitive, and social sciences, and there’s no way that one can cordon them off from the others. All the supposed shortcomings of special science kinds can be found in these cases too (for example, lack of necessary and sufficient conditions, lack of sharp boundaries, and so on), or so I tried to argue in my book. And conversely, many of the virtues of the kinds in physics and chemistry (for example, their robust causal profiles) can be found in these other kinds too.

On the too-restrictive side, some philosophers would include such things as kinds of beers or baked goods, which I would resist, because I don’t think these kinds correspond to clusters of properties that have a causal relationship to other properties. The fact that doughnuts are glazed but muffins aren’t or that muffins often contain walnuts but doughnuts don’t is a matter of convention not causality. It’s true that there are some domains where causality and convention tend to get entangled with one another, especially in the social world. Still, even there, I think one can discern the causal connections that characterize the structure of the domain in question and distinguish them from purely conventional links between properties.

3:16:  Do your natural kinds occur outside of the natural sciences? 

MK: Yes, and that’s partly why I’ve become increasingly dissatisfied with the expression, “natural kinds,” since it regrettably suggests a close connection with the natural sciences. Ian Hacking has pointed out that the first philosopher to use the term “natural kind” was not John Stuart Mill but John Venn. But what Hacking doesn’t stress is that Venn seems to have thought he was using Mill’s expression. At one point, Venn writes: “he [i.e. Mill] introduced the technical term of ‘natural kinds’ to express such classes as these.” This leads me to suspect that we owe the expression to a slip of the pen on Venn’s part! Mill sometimes spoke of “Natural Classification”, but never of “natural kinds,” just plain “Kinds,” or (occasionally) “real Kinds.” (He often capitalizes nouns like “Classification” and “Kind,” as was the common practice for many nouns in 19thcentury English.) In some recent work, I’ve been using the term “real kind” instead of “natural kind,” which is an expression also used by Ruth Millikan, though she has a distinct view of what real kinds are, which I don’t entirely share. Another reason for resisting the expression “natural kind” is that it suggests that artificial kinds cannot be natural kinds, which is also unfortunate since synthetic chemical compounds, genetically engineered plants, artificially selected animals, and many artifacts are all good candidates for kinds. So I think we would avoid a great deal of confusion if we jettisoned “natural kind,” but I don’t know if “real kind” will ever catch on.

Incidentally, Mill certainly thought it was at least possible for there to be kinds outside the purview of the natural sciences, in the social realm, as did Venn. So this is not a novel position. Nineteenth century philosophers like Whewell, Mill, and Venn posited natural (or rather, real) kinds in the social domain. In my view, the reason that there can be real kinds in the social domain is simple. There are clusters of causal properties in the social domain, as there are in the domains of physics, chemistry, and biology, and these properties give rise to other properties in a regular fashion, so there are kinds there too. People (myself included) sometimes talk of “social kinds,” but since this might suggest a contrast with natural kinds, perhaps it would be better just to say: real kinds in the social domain. Of course, causality is usually much messier in the social domain, but causal patterns are discernible there as they are in many other domains that are characterized by what Bill Wimsatt calls “causal thickets.”

3:16: Is your naturalist approach to natural kinds compatible with realism about them – this is scientific realism here not the metaphysical kind you discussed earlier? How does your account involve properties and causality – and why don’t you think scientific realism is about positing a world independent of the human mind?

MK: It’s very common in philosophy nowadays to encounter the phrase “mind-independent reality,” and to see realism defined in those terms, as in: “realism is committed to the existence of a mind-independent reality,” or something to that effect. Now some philosophers have recognized that this characterization is not quite right as it stands, since if we are realists about minds and their products (including, not just such things as thoughts, emotions, and perceptions, but also social institutions, processes, and events), we clearly cannot characterize them as “mind-independent,” so our understanding of realism needs to be modified. We could just say that realism means something different when it comes to different domains, but that seems ad hoc. So some philosophers have tried to distinguish an innocuous sense of mind-dependence from a problematic one. 

But I can’t seem to find an acceptable way of making this distinction. For instance, if we say that we can be realists about entities that are causally mind-dependent but we shouldn’t be realists about entities that are constitutively mind-dependent, that doesn’t do the trick. After all, minds themselves and many of their products are constitutively mind-dependent, not just causally. One might try to say that real entities are independent not of minds but of the classifying activities of human minds. But Hacking has shown that many real kinds in the social domain are interactive in the sense that they are shaped in part by our classifying activities, and I have argued that this also applies to artificially selected animals and plants and other real kinds outside the social domain. Indeed, the process of sexual selection can be said to give rise to mind-dependent kinds, as when peacock tails are fashioned by the desires of peahens. Maybe there’s some other way of specifying the supposedly problematic sense of mind-dependence. But I think mind-dependence is a red herring in this context. When it comes to natural kinds, we should be realists about clusters of causal properties. So scientific realism can be understood in terms of the claim that our theories and categories aim at capturing the causal structure of the universe, not as a commitment to the existence of a mind-independent reality.

3:16: So why isn’t your position a social constructivist one? 

MK: Given what I just said it would be easy to accuse me of lacking a commitment to the existence of a mind-independent reality! But a better way of putting it is to say that I don’t think realism can be captured in terms of a commitment to the existence of a mind-independent reality. So

I’m a realist who thinks that the joints of nature are causal relations and that science aims to discern those joints. Perhaps science doesn’t succeed in doing so as well as we might like and there may be many ways in which our current categories do not capture the causal structure of the world due, at least in part, to certain social biases. But I don’t think those biases are inevitable or that we’re incapable of overcoming them in the long run. We once thought there was such a thing as hysteria; now we know better.

But there’s a complication here that isn’t always brought to the fore. In some cases, merely believing in a certain kind and acting in certain ways tends to give it causal efficacy. I think that race has no more biological basis than hysteria, yet by virtue of the fact that generations of people have believed in it and acted accordingly, a kind has been created in the world. This real kind has significant effects not just in the social world (e.g. rates of incarceration, income levels) but also in the biological realm (e.g. rates of type-2 diabetes and hypertension, life expectancy, susceptibility to the coronavirus). These effects might cease to obtain if we all stopped discriminating on the basis of race and we redressed past wrongs. So it’s not inconsistent to be both a realist and a social constructivist about some kinds.

This also shows, incidentally, some ways in which the social and biological can become entangled, and how social causes can have biological effects. Indeed, it sometimes gets tricky to make a sharp distinction between the biological and social -- just as it is often tricky to disentangle the biological and psychological, the psychological and social, and so on. And that doesn’t just apply to the special sciences. We’re taught in elementary school about the difference between physical changes and chemical changes, and this is a useful rough-and-ready distinction, but there are many causal processes that straddle the divide.

3:16:  Turning to Islamic philosophy, there seems to be a revival of it in the philosophical canon and guys like Ibn Sina (Aviccena) and Ibn Rushd (Averroes) are being returned or put in alongside more familiar medievals like Aquinas  - although some are still a bit critical of there being too much focus on getting the texts right than in analyzing the arguments. You’ve also identified Orientalism as being very much part of how Islamic philosophy has been approached. Can you sketch for us what you take orientalism to mean in this context and are things improving? 

MK: Yes, those guys are getting more attention! Certainly more so than when I started grad school. When I arrived at Columbia in 1985, the director of graduate studies, James Walsh, worked in medieval philosophy and he tried to encourage me to specialize in classical Islamic philosophy -- but he was an exception, being a medievalist. At the time, I politely declined. In part, I didn’t want to reinforce the notion that “every fish is an ichthyologist” (as a friend of mine once put it), that is, if you have a certain cultural background or identity, that’s what you should focus on in your academic work. (The converse, which is a very simplistic version of identity politics, is even more problematic.) 

I didn’t take a serious interest in the Arabic-Islamic philosophical tradition until after graduate school and once I did, I found a great deal of fascinating material to work on. I can’t claim to be a real expert, since I don’t have the depth or breadth of knowledge, but I’m certainly a passionate dabbler. But to get back to your question, I agree that there’s been renewed interest in these texts, which I think you’re right to characterize as a “revival.” The Arabic-Islamic philosophical tradition occupies an unusual position vis a vis what we now think of as “western philosophy,” since it was once very much a part of that tradition but it’s been actively edited out of it for centuries. As far as Aquinas was concerned, Avicenna and Averroes were required reading and popped up all over the place in his writings. For Bacon, they were to be expunged -- along with Aquinas and other medievals! In the New Organon, he writes that the “Arabians and the Schoolmen… crushed the sciences” rather than improved them. But soon after that, they become practically invisible in the tradition (much more so than the Latin medievals), even though I would argue that they remained a covert and mostly unacknowledged influence. (For an excellent account of this influence, see Catherine Wilson’s contribution to the Routledge History of Islamic Philosophy, edited by S.H. Nasr and O. Leaman.) 

To the extent that they are given some attention, they are mostly dismissed. In his history of western philosophy, Bertrand Russell writes: “Arabic philosophy is not important as original thought. Men like Avicenna and Averroes are essentially commentators… Mohammedan civilization in its great days was admirable in the arts and in many technical ways, but it showed no capacity for independent speculation in theoretical matters.” The Palestinian-American literary scholar Edward Said characterized Orientalism as the view that the civilizations of Asia (especially, Islamic civilization) are monolithic and essentially different from western civilization. So Russell’s is the kind of position that I would characterize as Orientalist and it recurs in much of the nineteenth and twentieth century. But in the past couple of decades, there has been a considerable effort to approach this tradition by philosophers (rather than philologists or those who read the texts as historical artifacts) and this has resulted in a rich body of work.

3:16: You’ve examined Al Farabi’s paradoxical remarks on the democratic city, comparing and contrasting his views with those of Plato. What do they say about the democratic city and what does his departure from Plato reveal about the differences between these two philosophers? Is he more a democrat?

MK: It’s hard not to be more of a democrat than Plato! I wouldn’t say that Farabi is a democrat but he pointedly describes the democratic city in more favorable terms than Plato, despite his reverence for him. Farabi never explicitly signals his disagreement with Plato on this issue, but his description of the democratic city departs from Plato’s in various significant ways and at one point he considers the democratic city to be the best of the non-ideal cities. That’s clearly inconsistent with Plato’s views, since he ranks it second-to-last in Book VIII of the Republic

So why does Farabi take this position? I’ve argued that there are two main reasons. The first is that he ranks systems of government based on the ease of transforming them into the ideal system. By that measure, he seems to think that the democratic city is the one that is most liable to be converted into a virtuous city. That’s because the central value in the democratic city is freedom, so citizens are free to pursue their own conception of the good life. (Perhaps it should be considered a liberal democracy rather than a democracy full stop.) That means that there will be virtuous citizens and even virtuous communities within the democratic city, who can constitute something like a vanguard that would make the whole city virtuous. The second reason that Farabi regards a democratic form of government as the best of the non-ideal forms is that he’s influenced by his own milieu. Farabi was writing in 10thcentury Baghdad, an Islamic city with a decidedly multi-cultural, multi-ethnic makeup, and he seems to appreciate that diversity and find it intrinsically valuable. That appears to be partly why he describes the (liberal) democratic city in decidedly more favorable terms. Moreover, the description is very vivid and seems to derive from his own experience. He says for instance that “people of every race multiply in it,” that there will be “all kinds of copulations and marriages” (whatever that means!), and that strangers cannot be distinguished from residents. Farabi’s Baghdad wasn’t a democracy but it was a cosmopolitan and diverse society.

3:16: And what’s the apparent contradiction in Farabi’s verdict concerning the second best city?

MK: He seems to follow Plato at one point in saying that the democratic form of government is second-to-last, but at another point in the same text he declares it second-best. So I try to resolve the apparent contradiction by saying that it’s second-best in terms of ease of transformation into the virtuous city, though not perhaps in terms of its intrinsic resemblance to the ideal form of government.

3:16: You’ve written about Zaki Najīb Maḥmūd and his logical empiricism. What is this position and how does he defend it? Is he taking to task a certain kind of metaphysical approach to philosophy and defending one which sees philosophy as a science?

MK: I should explain that Mahmud (1905-1993) is a philosopher from Egypt who studied for a doctorate in philosophy in London in the 1940s. While he was there, he read Language, Truth, and Logic in preparation for A. J. Ayer’s inaugural lecture as professor of philosophy at University College London in 1946. Immediately, he underwent a lighting conversion to logical empiricism. When he returned to Egypt, he spent much of his career trying to explicate and disseminate the gospel of logical empiricism to the educated public in the Arab world. One of his first books was titled, The Myth of Metaphysics(1953), which scandalized many Arabic-speaking intellectuals. His positions and arguments weren’t particularly original, and were strongly influenced by the writings of Carnap, Reichenbach, Neurath, and others, including Russell, Moore, and (of course) Ayer. I was aware of some of his work ever since I was an undergraduate but had never read him in any depth until I was asked to contribute an essay on Mahmud for the Oxford Handbook of Islamic Philosophy. That handbook was an interesting project because the editors wanted to bring it all the way up to the present rather than leave the impression that Islamic philosophy was a thing of the past, or that it went extinct after Ibn Rushd (Averroes) at the end of the twelfth century, which is often the way the tradition is represented. I’m sympathetic to that idea but I’m not sure that one can say that there’s really one continuous tradition from the ninth century to the twentieth. After all, it’s hard to maintain that a logical empiricist like Mahmud is part of the very same tradition as philosophers like Farabi and Ghazali.

3:16:  How do his arguments link with twentieth century Arab thought, in particular, the possibility of reconciling tradition and modernity, the compatibility of Islamic religious thought with liberalism and democracy, and general questions of progress, secularism, and the state of Arab and Egyptian society?

MK: The short answer is that they don’t! And I think that’s partly why his ideas didn’t have much traction. Most philosophers and intellectuals in the Arab world were occupied throughout the 20thcentury with these other questions that you mentioned, so the status of metaphysics and the synthetic a priori were not of much interest. Mahmud’s logical empiricism had a negligible influence in the Arab world. Later in life, he turned to these other, more topical questions, but I have to say that I think his contribution is not particularly valuable. In my view, part of the problem is that he doesn’t make enough of an effort to engage with the Arabic-Islamic philosophical tradition, or relate logical empiricism to broader philosophical concerns.

In the Arab world as in many other parts of the world, questions of decolonization, imperialism, war, military occupation, inequality, religious extremism, and other pressing issues tend to dominate intellectual discussions. So metaphysics, epistemology, and philosophy of science are often regarded as luxuries, and the relationship between these pressing issues and abstract philosophical questions is not always clear. Moreover, in a part of the world where there is little funding for scientific research, the philosophical questions that arise from modern science are not always given priority.

At the same time, I think that the way many modern Arab thinkers engage with the Arabic-Islamic philosophical tradition is problematic. Some intellectuals in the Arab world today mine the tradition for certain nuggets or enlist it to support their own causes, by trying to show that the Arab-Islamic philosophical corpus is really materialist or secular or liberal or Marxist, or whatever. I think it would be healthier to interpret it, critique it, respond to it, and to engage in a constructive dialogue with it, rather than try to weaponize it to fight certain battles.

3:16: And finally for the curious readers here at 3:16, are there five books you can recommend that will take us further into your philosophical world?

MK: To narrow it down, I’ve chosen some books that I find myself returning to repeatedly. And to narrow it down further, I’ve also restricted myself to works written in dialogue form.

Plato, Meno

Rewards regular re-reading -- I’ve tried to use its approach to innate ideas as a model for understanding innateness in contemporary cognitive science.

Ibn Rushd (Averroes), The Incoherence of the Incoherence (translated by Simon Van Den Bergh)

In my opinion, this work represents the acme of the Arabic-Islamic philosophical tradition. It consists essentially of a debate between two philosophers, Averroes and Ghazali, since it’s the former’s detailed reply to the latter’s work, the Incoherence of the Philosophers, which he quotes almost in its entirety. It’s not a real debate, since Averroes was writing decades after Ghazali died, so he always has the last word, but it still reads like a philosophical dialogue between two formidable dialecticians. They debate such questions as the eternity of the universe and the nature of the connection between cause and effect (I’ve translated the latter debate in my anthology of Islamic philosophy.)

Galileo, Dialogue Concerning the Two Chief World Systems

A startlingly fresh debate between an Aristotelian and a natural(ist) philosopher.

Berkeley, Three Dialogues between Hylas and Philonous


One of the books that got me hooked on philosophy in the first place -- and converted me briefly to idealism.

Lakatos, Proofs and Refutations

A highly under-appreciated classic of twentieth-century philosophy. To my mind, Lakatos shows better than anyone else the untenability of the analytic-synthetic distinction, even in mathematics. “Monster-barring” deserves to be a central concept in philosophy.


Richard Marshall is biding his time.

Buy his second book here or his first book here to keep him biding!

End Times Series: the index of interviewees

End Time seriesthe themes