:: Article

truth, success and frank ramsey

Pascal Engel interviewed by Richard Marshall.

[Photo: Claire Poinsignon]

Pascal Engel is a funky French philosopher who sees analytic philosophy as a European product, thinks that analytic philosophy is unified by a commitment to a set of epistemic norms, that these normative commitments are not trivial, that his interest is epistemology not french epistemology, that much french historical epistemology is wrong, that knowledge is neither elusive nor empty, that Williamson’s ideas about ‘insensitive invariantism’ are largely right, that Williamson is wrong about luminosity and one or two other things, that Frank Ramsey was unique, that the Ramsey’s principle is important, that Ramsey was a pragmatist closer to Peirce than Dewey, that Ramsey was no straightforward anti-realist and instrumentalist about truth, that functionalism is a Ramseyan legacy in the philosophy of psychology, that French philosophy isn’t interested in deflationism, that deflationism is false, and that French philosophy has its own resources. Chillin’ stuff, like finding invincible summer in the depths of winter…

3:AM: What made you become a philosopher?

Pascal Engel: Pataphysics. I was a teenage pataphysician. Towards the age of fifteen I read Alfred Jarry ‘s Exploits and Opinions of Dr Faustroll, Pataphysician. There was a chapter called “Elements of pataphysics”. It started with the sentence: “An epiphenomenon is what adds up to a phenomenon”. So very early on, I knew what “epiphenomenalism” means. Pataphysics was defined as “the science of imaginary solutions which attributes symbolically to lineaments the properties of objects defined by their virtuality”. I suppose that a lot of schoolboys are attracted to philosophy through such bizarre and deep sounding “definitions”. In my case, the fact that they were so pompously absurd played a great role in my outlook. These strange words led me to read philosophical dictionaries, and to get acquainted the post-Kantian German philosopher doctrine of Hans Vaihinger in his Philosophie des Als Ob, according to whom most of our views about the world are fictions, a kind of generalized fictionalism. I learnt to like abstractions, without taking them too seriously.

I was also very lucky to benefit from an early teaching of philosophy. In France, philosophy is taught at the high school. We had eight hours of philosophy per week. My teacher, Claude Rebours, was remarkable. Within one year we were able to study in full three classical works – Plato’s Republic, Descartes Meditations, and Nietzsche’s Genealogy of Morals – and to cover the whole program of the philosophy class, which at that time was huge and very demanding. I was also a teenage Deleuzian. I read Deleuze’s Nietzsche et la philosophie, some Kant and Hegel, and started reading Foucault and Derrida. So by the age of 17, I was ready to follow the philosophical path.

Philosophy was then very popular in France, but not in the present day sense of people going to philosophical cafés or festivals, and of the success of lousy books on happiness and wisdom. Philosophy was a political subject, which made a difference to our lives. The intellectual atmosphere was electric, and just after 1968 we had a sense that something was happening, although nobody knew what it was. Sartre was still very active in leftist politics, as well as Deleuze and Foucault. They were our stars, and everyone was discussing Freud, Lacan, Marx, Mao. We dreamt – but only dreamt – of being street fighting men. Philosophy was at the center of everything, we had the feeling that ideas had immediate applications in politics, although we did not see exactly which ones. At the Ecole Normale Supérieure, my teachers where Louis Athusser and Jacques Derrida. They taught us to be wary about any philosophical view whatsoever, either because it was politically suspect or because it had to be deconstructed. I actually was not very impressed by them, and preferred to attend Gilles Deleuze’s classes in the “experimental” university of Vincennes. I also followed every course by Michel Foucault at the Collège de France. I probably understood only 20% of what they were saying, but that was enough to breed my enthusiasm.

Some years later, I was to reject their views, but I would probably not have gone into philosophy if my teachers had been – say – A.J. Ayer, Peter Strawson or W.V.O. Quine – whom I would have found boring and useless if I had read them at the time. I also found Husserl and Merleau-Ponty boring. At the Sorbonne I had a class by Levinas, who was a very charming man, but his style –he was commenting Husserl as if it were the Talmud – did not appeal to me. At some point, however, I realized that if I had to go on doing philosophy in the way Deleuze and Foucault where doing it, it would lead me to nowhere. Their views were only critical – we were told actually that philosophy, as a serious theoretical subject, was dead – and it was hard to figure out what the successor discipline would be: some version of psycho-analysis, some kind of historical or sociological inquiry, or some kind of literary writing, some sort of Borgesian metaphysics , or perhaps some improbable mixture of all these? In spite of the nihilistic atmosphere which surrounded my early philosophical education, I felt that in some sense philosophy had to deal with truth, knowledge, and reason. I started reading Russell, Frege, Carnap and their commentators. It was at that time (the early seventies) very unfashionable to do so in France, but we had also some great teachers: Jules Vuillemin at the Collège de France, Jacques Bouveresse at the Sorbonne, Gilles Granger in Aix en Provence. They were not analytic philosophers. Their actual style was still historical. They explained to us the classics of analytic philosophy, but rarely engaged in a direct confrontation with the authors that they studied.

Their work was very intelligent, often deep, but I had the impression that the argumentative work and formulation of theses was always kept at a distance. There was rarely actual discussion in class. When I discovered the way philosophy was done in England and in the U.S, where people jumped onto what you say at once and express clearly their disagreements, I was struck by the contrast, and soon preferred the more naïve and open-minded Anglo-American style. When I went to Berkeley in 1978, I was very surprised to see philosophers talking about philosophy and exchanging arguments in the corridors of Moses Hall. I had never seen that before in France!

In a sense, when I started working in analytic philosophy I had to do a second philosophical education. I rejected the views and the attitudes of my poststructuralist teachers, but I kept an interest in the history of philosophy, which I have always taught, and I also rejected the kind of narrow specialization which analytic philosophy more or less imposed. I have always found that there are important analogies and bridges between theoretical and practical philosophy, between the philosophy of language and the philosophy of mind, between epistemology and ethics, and I always feel that when a philosopher is not able to trace some of these connections and to have at least a rough idea of the history of his subject, there is something missing in his or her outlook. In this sense, I am a synthetic philosopher, and- I am sorry to say – bear the marks of my continental education. I wish I had succeeded more in transforming myself than when I tried to transform others.

3:AM: You count yourself as analytic and work in Europe. So what philosophical distinction do you think exists between Analytical and Continental philosophy, if anything?

PE: Geographically and historically speaking, analytic philosophy is a European product. It was invented by Austrian and German philosophers in the nineteenth century, later by English, Viennese and Polish philosophers. It reached the US only by the nineteen forties. It is true that it then became a mostly Anglo-American phenomenon. I was very tempted, after my dissertation on Kripke in the early eighties, to stay in the US where analytic philosophy was in its heyday, and to do a second dissertation with the philosopher whose work fascinated me most, Donald Davidson. But apart from the practical difficulties and in spite of my lasting attraction for Anglo-American analytic philosophy, I have always considered myself, culturally speaking, as a European. I had also, at that time, a pioneering spirit. I hoped to establish firmly the analytical style of work in France.

In my generation, a number of European philosophers have studied in the UK and in the US, and have come back to their respective countries, in the hope of changing the intellectual and institutional landscape. I cannot say that they succeeded – the resistance to the analytic way of practicing philosophy is still very strong today in Europe and the university system is still very different from the American one– but they have at least managed to create a common space of cooperation. So it is, if one compares with the situation when I was a student or a young academic, much easier to do analytic philosophy in Europe than thirty years ago. Concerning my own attempts, however, I still find very relevant a remark that was once put to me by Mark Sainsbury: “You try to force the French to eat fish and chips”. Indeed I could never force them into such culinary revolutions.

Ideologically speaking, it is clear that from the eighties onwards analytic philosophy has lost its unity. On the one hand, logical positivism lost its grip, as well as the methodological primacy of logic and the philosophy of language and others fields like philosophy of mind, metaphysics and epistemology tended to occupy center stage. On the other hand, a number of philosophers trained in the analytic tradition distanced themselves more and more from its main views and methods, sometimes to the point that they found inspiration within the continental tradition. What is there today in common between the disciples of David Lewis, David Armstrong or Timothy Williamson on the one hand, and those of Wittgenstein, Brandom or Putnam, although they still share a common heritage? Between cognitive science and empirically minded naturalists on one side and hard liners metaphysicians or epistemologists on the other? Do they even read each other’s writings or publish in the same journals?

There are basically three attitudes which one can take. (i) A deflationist stance, illustrated e.g by H.J. Glock in his book What is Analytic philosophy?: there is no doctrinal nor methodological unity, only a set of family resemblances between analytic philosophers. (ii) An eliminativist stance, according which to analytic philosophy does not exist anymore, and has to be replaced by – depending on the perspective adopted – experimental philosophy, Wittgensteinian quietism, or some post-analytic successor subject , most often a kind of ecumenical and user friendly mixture of the two traditions. (iii) A conservative stance: analytic philosophy has still to be characterized, in spite of all its diversity, by a set of core subjects and by a common method, based on substantial theoretical commitments. I have long argued (for instance in my book La dispute) that analytic philosophy rests centrally upon : a) a theoretical, and not only a purely critical or therapeutic attitude, b) a strong commitment to take logic and formal methods seriously , b) acceptance that the subject still has a common core composed of metaphysical and epistemological theses, and c) an interest for description, not only reconstruction, formal or metaphysical. Indeed a number of analytic philosophers do not find either one these commitments exhaustive or exclusive, but even when they tend to minimize one of them, they at least recognize their legitimacy and fruitfulness.

Thus Wittgensteinians who disparage the use of formal methods will keep a strong link with other analytic philosophers if they consider that the main issues of the philosophy of logic and mathematics for instance are not purely illusory. Moral and political philosophers tend to move away from the center of the field when they consider that the issues of meta-ethics are not worthy of attention, and cognitive science inspired naturalists or experimental philosophers who reject any kind of conceptual analysis or metaphysical inquiry whatsoever leave the boat. It would be absurd, however, to tie analytic philosophy to very specific doctrines about metaphysics or logic (such as those that prevailed at the time of positivism).

Nevertheless, it seems to me that analytic philosophy is, and has to be, unified by a commitment to a set of very general epistemic norms. These are: that truth and knowledge are the aim of belief and of philosophical inquiry, that there is such a thing as an autonomous “space of reasons”, that there are objective epistemic norms and values that there is such a thing as being right or wrong in philosophy, that there are standards of argument, and that logic, both as an organon and as a canon has a strong normative role to play in philosophy. I call these commitments substantive, because a realist and cognitivist view of norms and reason seems to me to be the one which makes the more sense.

Contrary to what many people say, these commitments are not trivial. Their exact nature is controversial. They are actually rejected most clearly by the post-Nietzschean and post-modern philosophers, and by some naturalists today, and by the so-called “pluralists” about normativity. I admit, though, that these criteria have been shared by philosophers formed in the phenomenological tradition and by some (mostly Peircian) pragmatists. I find, however, that in general analytic philosophers have been those who are the most self-conscious about these norms. It is actually a very important task of contemporary philosophy to inquire into the nature of epistemic norms and values, and this inquiry is quite relevant to the metaphilosophical question of the nature of analytic philosophy. Those who reject these cognitive norms outright are most of the time critical of analytic philosophy. It is certainly absurd to draw any sharp line, and to divide the profession into card holding analytic and card holding continentals. But it does not follow that there are no dividing lines.

3:AM: Is French epistemology changing thanks to you? Didn’t it used to be bound up with Bachelard, Canguilhem, Foucault, Popper and Kuhn but you’ve been part of a movement to broaden the scene to include the likes of Lewis and Williamson. Have the arguments about truth and scepticism that you have introduced been a wake up call to the contemporary French philosophical scene?

PE: Besides the fact that it’s not for me to say, I have no such ambition – my ambition is to be an epistemologist, and I am a French epistemologist by accident. Would you ask an American epistemologist whether he or she is changing American epistemology?

What is called “epistemology” in France actually does not designate the theory of knowledge. It designates a sort of historical philosophy of science. It is a tradition which goes back from Comte’s positivism and which includes writers like Cournot, Duhem, Brunschvicg, Meyerson, Bachelard, Cavaillès, Canguilhem, and, to a certain extent, Michel Foucault. It typically does not address the basic issues of what is called epistemology in the English-speaking world: the problem of external world skepticism, the foundation of induction and of scientific inference, and the definition of knowledge. These issues have not been completely absent from the tradition in France, but the historians of science have always tended to consider that an abstract analysis about the principles of knowledge was useless and illusory. They aim at analyzing scientific reasoning from within scientific theories, by investigating the historical changes which they undergo. For them epistemology does not exist apart from the actual practice of scientists, which can be only described by historians. Any kind of normative analysis about what knowledge is or ought to be seems to them completely misguided.

I have no objection against doing history of science in the piecemeal, erudite and most of time very original way in which it is done in France. It has produced some great works. But it seems to me false that it is done in a purely descriptive manner, as if the best a philosopher could do would be to give an intelligent paraphrase of what the scientists are doing. Bachelard, Canguilhem or Foucault in fact make substantive claims about the nature of scientific knowledge, about rationality, truth and realism, although these are most of the time buried within their historical narratives about science and its boarders. Their claims often amount to various forms of rampant relativism. When Foucault talked about historically situated “games of truth” and “knowledge formations” or epistèmai”, he actually radicalized Bachelard’s historical rationalism and Canguilhem’ positivism. I actually oppose much of contemporary French historical epistemology in thinking that scientific practice, as well as our ordinary natural knowledge, involves an irreducible normative dimension. There are norms of knowledge. This is actually a view which was defended by some French philosophers at the beginning of the last century (e.g André Lalande, a now forgotten figure, in his book La raison et les normes), but which has fallen into disrepute, mostly thanks to Canguilhem and Foucault.

Much of my own work in epistemology has been about the nature of epistemic norms: Are there any such norms? How do they differ from values? From reasons? Are they involved in the nature of knowledge? What is their ontological status, i.e are they real and objective? How do we have access to them and what is their epistemology? How do they guide our beliefs? In some sort of prescriptive way, like imperatives or obligations? Or like permissions to believe? These questions, which belong to what we might call meta-epistemology, have close relationships with classical issues in epistemology, such as internalism vs externalism about the nature of justification, the nature of belief, assertion and judgment. They have obvious parallels in meta-ethics. What are the relationships between epistemic and ethical norms and values? Can they be expressed in terms of our reasons to believe and to act? What is the nature of these reasons?

These issues might seem to be very alien to the French context. Ironically, however, they are not so far from some of Foucault’s concerns: wasn’t he after all always discussing the nature of knowledge , or norms and their relationships with power ? Indeed he had a very different take and did not even accept the idea that there are objective epistemic norms. More generally, the French epistemologists would have no patience with Gettier’s problem, if they knew about it. Who cares whether Tom Grabit’s twin brother might have stolen the book from the library, thus defeating your belief that Grabit did it? Or whether Henry sees real or fake barns facades? Indeed many analytic philosophers have themselves mocked Getteriology.

But it is false that the Gettier problem is insignificant: it serves as a springboard for philosophers to test their views about the conditions of knowledge. Moreover it’s not true that issues in normative epistemology cannot be put into contact with issues about the history of science, or about issues in social epistemology. Most French epistemologists typically assume that knowledge is a social phenomenon, and suspect that the individualist style of much analytic epistemology is beside the point. They accept implicitly a form of revisionism about social epistemology, to the effect that issues such as the nature of knowledge, justification or reliability do not arise when knowledge becomes “social”. This is wrong, as we can see, for instance, from the problems that arise about testimonial knowledge or about the nature of practical knowledge. Social epistemology is but an extension of individual epistemology. Nobody is going to ask French philosophers to abandon their tradition and to start doing Gettierology or the kind of epistemology which is done in the US. But if French historical epistemologists saw better that the issues which form the basis of the analytic philosophy of knowledge are central for their own pursuits, they probably would have more interest in these. Unless, indeed, they consider epistemology useless and dangerous, as a number of people in science studies, such as the sociologist Bruno Latour who once said: “I consider epistemology like asbestos. It’s a perfect product which has been flocked in all buildings to avoid fires and now we realize that it provokes professional illnesses.”

3:AM: As an epistemologist do you think knowledge is elusive because the term is empty, and would that be an approach developed out of your work on Ramsey’s principle, (which I’ll ask about in a minute)?

PE: I do not take knowledge to be elusive or empty. On the contrary, it seems to me to designate a bona fide natural kind, although not one which is easy to pin down. Unlike contextualists, I take expressions such as “knowledge” or “knows” to be invariant across contexts (although I do not deny a certain amount of context sensitivity in our epistemic terms). In the current jargon I am an insensitive invariantist. Unlike eliminativists about knowledge, among whom I count a number of experimental philosophers, I do not think it is an empty term. I do not think, however, that knowledge can be defined through a set of necessary and sufficient conditions. This I take to be one of the lessons of Gettierology.

This does not mean that there is nothing to say about knowledge and that epistemologists have to pack up and leave. Although knowledge cannot be defined in the strict sense of this term, we can still characterize it functionally, through its relationships with other notions, such as those of justification, evidence, reliability, or safety, and we can try to give explanations and theories about these notions. Thus it makes sense to ask whether internalism or externalism about knowledge are correct, whether reliabilism or evidentialism are correct, and to work out the best versions of these.

I also take it that , although the analysis of knowledge is basically a conceptual and a priori matter, we can learn a lot (although not everything) from cognitive science, ethology and especially from developmental psychology about what knowledge is. These issues cannot be dealt with only at the level of an account of knowledge in general, but have to be dealt with about particular kinds of knowledge depending on its sources (perceptual, inferential, testimonial, a priori) and of its domain (natural, scientific, moral, aesthetic). Perhaps there is no single account which works fully for all domains, but I take it that they have a number of traits in common (in this respect the shape of the issues is pretty much like for truth, which can vary across domains, but keeps a functionalist core)

Ramsey’s ideas about knowledge, which were in many ways pioneering, are perfectly consistent with this functionalist account. Ramsey defined knowledge as based on a reliable process. For him it was not a definition, but a functional characterization. He held that belief is a disposition to act, and that the content of a belief is constituted by the difference it makes to the success of actions based on it, given the desires of the agent. This is Ramsey’s principle, and the basis of “success semantics”. It does not mean that truth is defined by utility, nor that a belief’s content consists its success conditions, independently of holistic conditions and of the interpretation of desires ( I myself think of this as a revision of Davidson’s method of interpretation) . The same thing holds for knowledge. Knowledge is the kind of state that, together with desires, leads to the success of actions. This is not a reductive definition either, but a functional one. It states that knowledge has a certain role in practical reasoning. So I agree with those philosophers who take knowledge to make a difference to action and to be a necessary premise in practical reasoning. Nevertheless, I disagree with those, among whom many pragmatists, who take this to entail that knowledge is to be defined as what makes a difference to the success of our actions. Ramsey’s principle does not entail pragmatism as a reductive view about knowledge and truth.

Knowledge, belief, truth and action stand together in close relationships, although they cannot be defined independently from each other. They are not elusive for that.

3:AM: You support Tim Williamson’s ideas about ‘insensitive invariantism’- that knowledge is constant and context free. You also commit like Williamson to epistemological externalism and knowledge as a form of safe belief rather than a belief counterfactually sensitive to truth. Can you say something about your positions here.

PE: When some time ago I wrote on Ramsey I used to think of myself as a Bayesian, and thought of knowledge, in a rather classical way, as a form of rational credibility. I have been convinced by Williamson that knowledge cannot be so analysed, and I have now a strong sympathy for what he calls “knowledge first” epistemology. However, if one does not spell out the principles of such an epistemology, it is not much more than a useful and exciting slogan. Knowledge first epistemology has still to be built up.

As I said above, I reject contextualism about knowledge attributions, and side with classical invariantists. My objections to contextualism are not very original. It would be too long to detail here, since there are distinct kinds of contextualism. Contextualism can be a thesis about knowledge attributions from the point of view of the ascriber or from the point of view of the subject to whom knowledge is ascribed. Contextualists also differ about the nature of the mechanisms which create the context sensitivity of knowledge attributions: the stakes which can be perceived or not by the subject, salience of certain kinds of questions. I do not deny the existence of these phenomena and their ubiquity, but I wonder whether the examples and cases that are pervasive in the literature entail the sometimes very bold and sweeping consequences which are drawn from them, for instance that knowledge is subject to strong pragmatic encroachment or systematically varies with the stakes. I wonder whether the linguistic intuitions cannot be explained on much more classical grounds, such as Gricean implicatures.

So I remain unconvinced by the semantic arguments, and I am not convinced either that the linguistic evidence – or the psychological evidence, since this has been widely studied from that angle too – is overwhelming. I am even less convinced by more radical views, such as neo-relativism about semantics. I have also doubts about how much we can conclude about the nature of knowledge on the basis of knowledge ascriptions. Moreover, contextualist solutions to skepticism seem to me to misrepresent the generality and the universality of the skeptic’s challenge.

Safety and sensitivity are the basic counterfactual conditions for knowledge. The first says that if one knows that P one’s belief could not have been false, the second says that if P were true, one would believe it . Both amount to the general idea that if one knows that P one could not easily have been wrong. Many people think that sensitivity is too strong, and that safety is too weak. There are various versions of both, but basically they need to be supplemented by other conditions, such as Prichard’s anti-luck conditions, Sosa and others’ reliabilist virtue epistemology, or other conditions about reliability. It would be nice if we could handle the notion of safety only through the semantics of counterfactual conditionals, some modal conditions, and probability. But it’s not clear that these conditions, which are necessary, are sufficient. Some emphasize the Nozickean notion of method, but this too has to be specified. In sum it seems to me that the jury is still out about which condition one must add to the bare modal theory of knowledge. I’m afraid that I can’t be more specific here.

3:AM: At some points however you seem to incorporate internalist views of knowledge, for instance when you talk about the phenomenology of perceptual experiences as having the ‘feeling of seeming to ascertain that a proposition is true.’ You fight this off though don’t you and argue that it’s not full-blooded internalism and so you are consistent. Can you explain this and how you deny that knowledge is luminous ie the KK principle? I guess the question is why you think your Moorean arguments don’t leave you disagreeing with Williamson over this?

PE: In my book Va savoir I have argued that pure externalism or reliabilism about knowledge, and a pure safety account, are not enough to account for some conditions of knowledge. Two main conditions emerge, and here I disagree with Williamson. The first one concerns self-knowledge. It is typically luminous and subject to the KK principle, and I do not see how one can get rid of the phenomenon. In particular self-knowledge seems to me to be based on the rationality of belief, which is manifest in the exclusion of Moorean thoughts such as “P but I do not believe that P”. To speak like Sydney Shoemaker, we cannot be “self-blind” if we are to have beliefs at all. The same, as well known, holds for knowledge. So unlike Williamson, I take exception to luminosity. The second phenomenon is skepticism. Like many others, I have always found the reliabilist answer to the skeptical challenge too easy. The skeptic is indeed wrong in thinking that the individual in the Bad Case has the same evidence as the one in the Good Case, but skepticism only makes sense if we grant the internalist premise that it seems to us that things would be the same in the Bad Case. But these exceptions to luminosity do not make me an internalist, and do not lead me to reformulate the conditions of knowledge in an internalist fashion, for instance in the way evidentialists do.

I once thought that I could incorporate the internalist element through the notion of entitlement or of prima facie justification defended by a number of authors. I was attracted by this notion because it is quasi externalist: we need not be aware that we are entitled to be such. I am less sure now that this is the correct form of internalist requirement. It is also important for me to keep some form of internalist requirement in epistemology. The reason is that I want to argue that the phenomenon of what is called the transparency of belief – namely the fact that when we ask ourselves whether we believe that P we typically ask ourselves whether P – accounts for the regulation of belief by a norm of truth ( actually a norm of knowledge). Nevertheless the Moorean argument that one knows that one has two hands is perfectly compatible with Williamson’s externalism. There is no inconsistency in attempting to integrate internalist elements within a broadly externalist view. Indeed it is what a number of epistemologist have been doing, for instance Ernest Sosa when he distinguishes first-order level and second-order level knowledge (although I do not combine internalism and externalism in the way Sosa does). Another, and not minor, point of disagreement that I have with Williamson is that I grant the existence of a priori knowledge. In brief, one can be a knowledge first theorist without subscribing to all the aspects of Knowledge First Oxfordianism.

3:AM: You call Frank Ramsey ‘one of the first self-conscious “technical” analytic philosophers in the contemporary sense.’ He had a wide range despite his short life but you look at one aspect of his work – ‘Ramsey’s principle’ which is that ‘the truth of an agent’s belief guarantees the success of his actions’. Can you say something about this, how he applied it and why it is so significant?

PE: Ramsey was unique in his combination of logical and mathematical skill, range and depth, and capacity of articulate a philosophical theory both formally and by way of argument and conceptual analysis.

As I said above, Ramsey’s principle says that true beliefs are those that lead to successful actions whatever are our underlying desires, and that a belief’s truth condition are those that guarantee the success of our actions, based on that belief and whatever are our underlying desires . It should not be understood as a definition of truth or as a definition of the content of our beliefs. In other words it does not entail that truth can be defined, in the pragmatist’s sense, as utility or success, or that the semantic content of our beliefs can be defined as the success of our actions, or as their teleofunction, in the manner of Ruth Millikan, Fred Dretske and David Papineau. In our book on Ramsey, Jerome Dokic and I do not understand « success semantics » in the reductivist sense. Ramsey’s principle just states a functional condition on truth and belief. It is significant because there is a strong, quasi transcendental, connexion between knowledge, belief, truth and action. This connexion can be understood in various ways.

It can be understood in a Bayesian way, by taking beliefs to be credences, and through understanding of actions as the products of such credences and desirabilities. But then knowledge is just high credibility, and beliefs are subjective probabilities, in Ramsey’s terms « partial beliefs », but not full beliefs. Ramsey’s principle is actually about full beliefs, and does not work for partial beliefs. For my part, I take full or categorical beliefs to have primacy over partial beliefs or credences. I reject full blown Bayesianism. But I’s quite hard to determine what are the relationships between categorical beliefs and credences.

3:AM: He’s a type of pragmatist isn’t he, linked perhaps to Peirce, but you show that he’s very different from several versions of pragmatism out there. Can you summarise what makes his pragmatism distinctive from what you call ‘smooth’ versions of pragmatism. Is Rorty a smooth pragmatist in your sense then?

PE: Indeed Ramsey was a pragmatist. He said he was, and Hugh Mellor and others have been right in interpreting him that way. But there are pragmatists and pragmatists (here most of what I know about these matters I owe to Claudine Tiercelin). Ramsey’s pragmatism is closer to Peirce’s. Peirce’s pragmatism is very different from James’s. It acknowledges the existence of real universals, refuses to define truth as utility or success, and accepts the existence of norms and ideals. James may have granted norms and ideals, but he was, overall, a nominalist and a partisan of truth as utility. Ramsey was a redundandist or deflationist about truth, a realist about knowledge. It is less clear whether he was a realist about universals like Peirce, but I believe that Mellor has interpreted him convincingly in this way.

Both Ramsey’s and Peirce’s pragmatism are very different from Dewey. Rorty is closer to Dewey. Rorty rejects any substantive theory of truth, considers that issues about realism and correspondence make no sense, assimilates truth to justification by consensus and takes epistemic norms to be empty. James himself distinguished between tender minded and tough minded philosophers. Pragmatism, under the hands of Rorty and a number of others, has become tender minded. It has become a sort of version of post-modernism.

3:AM: You tend to think of pragmatists as having an anti-realist, instrumentalist view of truth conditions but Ramsey’s case complicates this view doesn’t it?

PE: Yes indeed, although it’s true that here are strong instrumentalist overtones in Ramsey, for instance in his conception of scientific theories. Although his views have been traditionally understood as a version of anti-realism, it is not sure that it is. Nevertheless, a number of philosophers today, such as Robert Brandom, Huw Price and Simon Blackburn, interpret Ramsey as an anti-realist, or as a “quasi-realist”. I prefer a stronger beer.

In general I am bewildered by the sheer ignorance of the Peircean version of pragmatism by most later day pragmatists. It is true, that, in all terminological rigor, Peirce himself was not a “pragmatist”, and he is miles away from Dewey or, for that matter, Rorty or Price. This had led to quite a lot of misunderstandings, some of which he tried to dispel himself be renaming his view “pragmaticism”. But one century later, they are still with us.

3:AM: You also doubt that Ramsey’s principle was a sort of biological functionalism don’t you? This distances you from seeing Ramsey’s Principle in terms of the teleosemantics of Ruth Millikan, Fred Dretske and David Papineau where what we get is a naturalistic account of beliefs doesn’t it? So is your version anti-naturalistic?

PE: I think I have already answered this question more or less above. My main difference with teleosemantical versions of success semantics is that I do not expect success semantics to give a reductive account of intentionality and meaning. My version of success semantics, however is not anti-naturalistic, if by this you mean that it does not entail any kind of dualism or the acknowledgement of entities which fall out of the causal order. But here too, there is naturalism and naturalism. Almost everyone today is a naturalist. In so far as I insist very much on the normative element in belief and knowledge, and take a realist understanding of the normative domain, I both part company from the anti-realists and from the hard liners among naturalists. I am not a “pragmatic pluralist” in the sense defended recently by Huw Price, if you want.

3:AM: Do you approach your investigations into, for example, the philosophy of psychology applying a similar approach to Ramsey’s Principle? Common sense worldviews and scientific ones tend not to converge. One only has to think about the weirdness of modern physics compared to what seems to populate ontologies of common sense to see this. So how should we understand the difference between scientific knowledge about minds and our own pre-scientific beliefs about them?

PE: There is a Ramseyan legacy in the philosophy of psychology too. It is, for a large part, the legacy of functionalism in the philosophy of mind. Functionalism gives an elegant and forceful answer to the problem of the compatibility between the “scientific” and the “manifest” image of psychological phenomena. It characterizes these phenomena at the higher level of their common sense interactions between each other and with actions, and allows for the reduction of these functionally characterized states to physical states. Indeed, things cannot be as neat as this, as many philosophers have argued. Some point out that functionalism does not account for the external character of mental content, others for their phenomenal nature. Physicalists aren’t happy either for they do not see how one can preserve the macro level when one moves onto the micro level of neuroscience.

For my part, my main problem with functionalism, especially for the central case of functionalism about belief, is that it does not account for the normative nature of our propositional attitudes. But I do not reject functionalism nor the project of cognitive psychology, broadly speaking, of accounting for psychological laws at an intermediate level between the common sense network of mental states and the neurophysiological level. In my view even at that intermediary level, there is a normative dimension in cognitive states.

3:AM: You’ve argued against deflationist views about truth. In the French context in particular I’d have thought that this was somewhat scandalous what with Foucault and Latour hovering about. When you argued against deflationsim were you conscious of being iconoclastic? Were there partisan kickbacks? And how do you respond to those who say that you’re dragging an exciting French philosophical tradition into the pointless and boring ‘quagmire’ of Anglophone analytic philosophy, as Rorty once contended?

PE: I’m afraid that nobody in France ever had the slightest idea of what deflationism about truth means, and about the difference between it and other kinds of theories of truth. Actually the philosophy of truth , under the form that it takes in the analytic tradition since at least the 1920s– roughly the debate between classical theories such as correspondence, coherence, verificationism, pragmatist vs various version of deflationism and minimalism – has been absent from he French context. Pragmatism was discussed at the time of Bergson, but apart from Julien Benda, a stern critic of Bergsonism – nobody really discussed the pragmatist theory of truth.

Nietzsche is an interesting case. Everywhere he denounces philosophers’ reverence for truth, says that the concept is empty, coming down to a set of metaphors, the product of the remnants of Platonism, a dangerous ideal at the service of the forces of resentment, etc. But elsewhere he says that he himself is a seeker after truth, that no ideal is more noble than truth, etc. Is he a nihilist about truth or not? Many Nietzschean passages, however, can be understood as saying this : there is no problem with our ordinary concept of truth, with its associated trivialities ( to assert something is to assert it as true, that to say something true is to say that things conform to what we say, etc. ), but truth is a deep illusion ( an “error”) as soon as one takes it having some kind or essence, as designating some sort of transcendent domain of the Platonic kind, and as making it a Goddess. In this sense, Nietzsche is close to deflationism about truth.

One finds a similar duality in Foucault: on the one hand, he tells us that the “history of truth” is the history of the ways in which power imposes itself within “knowledge formations”; on the other hand, he says that he uses the ordinary, humdrum conception of truth and sees nothing wrong with it. Therefore, in a sense, many poststructuralists were deflationists in spirit if not according to the letter of the doctrine. You find in Rorty exactly the same kind of move. Bernard Williams has called Rorty and others “the negationists” (in his book on Truth and Truthfulness), but he excepts Nietzsche from this group. The deflationist move seems to be disarmingly simple and innocuous. But it is not. The deflationist about truth is pretty much like the man who robs you of your wallet just in your face, and says “So what ?” with a smile before you can protest.

3:AM: So what’s wrong with the deflationist position? You say there are things that appeal but wrong inferences are drawn from the things that appeal – for example when you argue that ‘minimalism about truth does not imply a minimalism about truth aptness.’ So how do you end up being a realist about truth?

PE: What’s wrong with deflationism is ( basically and among other objections) that i) it does not account for all the ordinary uses of “true” (in spite of all the efforts of pro-sententialists, deflationists, disquotationalists and redundantists , I am still in doubt that these can account for generalizations of the form “What he says is true”, ii) it has a problem with the semantic paradoxes, (iii) it does not account for the normative dimension of assertion, iv) it does not account for the difference between truth and justification. Indeed I do not ignore that there are now very sophisticated versions of deflationism, including answers to (ii) ( see the work of J.C Beal), that Paul Horwich, one of the main advocates of the view has now refined and consolidated it through a lot of forceful arguments. But (iii) and (iv) still seem to me very strong objections. They were pressed by Crispin Wright and later Huw Price, who nevertheless defend versions of what they call minimalism about truth, a sort of “inflated version” of deflationism.

Do I end up a realist about truth? Yes. In my book Truth and elsewhere, I have tried to argue that what I call minimal realism can accept all the platitudes about truth that deflationists and minimalists accept, without concluding that truth is a “thin” or purely logical property. Truth is a substantive because the concept of truth registers a distinctive norm, applicable to all the discourses to which it is apt to apply, which is that a proposition is true because there is something in the world in virtue of which it is true. This parallels the fact that the norm of assertion and of belief is knowledge. To assert that P is to represent oneself as being in position to know that P. That does not make truth depend on our knowledge of it. Our best theories might well be false. Hence I am a realist about truth. Now the precise form of this realism still has to be formulated. I am, however, skeptical about the view that statements are made true by truth-makers. But I agree that one needs some notion of fact, and a substantive one (not one which only makes facts, as David Armstrong says, the mere “tautological accusatives” of propositions. But that’s a longer story, which cannot be given here.

3:AM: Do you think there are other areas of French intellectual endeavour that are in need of more Anglophone influences or was philosophy an exception? Is French science, economics, history parochial?

PE: Although I may sometimes have given this impression in the past, I do not consider myself especially as attempting to bring Anglophone influence to the French intellectual life, because here are many sectors of Anglophone research or culture which I do not particularly think as worthy of being influential. I am not really sure that French philosophy is, or has been in the past, in need of Anglophone influence. The paradox is that much of Anglophone philosophy and culture has been influenced by some trends in French philosophy which I would have preferred not to be so influential, especially when they come back to France as adorned by the prestige of having been successful in the English speaking world!

I got interested as a student in English speaking analytic philosophy because I tried to escape the atmosphere of irrationalism and misology which was reigning at that time within French culture, and I found within Anglo-american analytic philosophy the resources that I needed at the time. It is true that for the kind of topics which interested me at the time –philosophy of logic and language – we did not have in French the proper resources. I translated a number of books and wrote introductions because I needed them for my teaching. I just tried to survive within an often hostile world. What I did not know then is that I actually could I have found within the French tradition the rationalist antidotes that I needed, for instance in the writings of Julien Benda, a great French rationalist on whom I have recently written a book. When one actually looks at the kind of irrationalist writings of French origin that the English-speaking world seems ready to absorb and to praise–such as, for instance the self-proclaimed “speculative realism” –which bounce back to us with the label of “new French philosophy”, or when one reads a certain kinds of post-analytic philosophy or of free spirited “pragmatism” coming from the US., and even when one reads a certain kind of analytic philosophy, one feels that one does not need that kind of Anglophone influence.

A number of people today are unhappy about the way analytic philosophy goes. I have myself said, in the valedictory editorial comments that I wrote for the journal dialectica, that there is much bad analytic philosophy. I do not know exactly the reasons for this phenomenon. The fact that analytic philosophy is now everywhere? The pressure to publish? The fact that in today’s cyber-sphere, everyone can become famous for 15 minutes? The fact that people read less and read only within their limited area of research, and that they read less and less the classics of the history of philosophy? A certain kind of aggressivity which reigns in today’s academia? These phenomena are widespread, and do not pertain only to the English speaking world.

With respect to French philosophy, it contains, within its own tradition, the resources necessary for a defense of rationalism. I am thinking of a line of thinkers like Cournot, Renouvier, Ribot, Benda, Nicod, Couturat, Cavaillès, Vuillemin, Granger, Bouveresse and others. Most of these names are probably unknown within the English speaking world. It is sad that this tradition is often foreshadowed by intellectual products which bring just the opposite. But is also a reason for hope, if one can revive and reinvent this tradition. If we do, it will not be distinctively French, but, hopefully, international.

3:AM: And finally, for the intrigued Francophile readers here at 3:AM, are there five books other than your own that we should go out and read to discover more of your philosophical world?

PE: I suppose that your Francophile readers expect me to list a series of books in French and by French philosophers. Some of them have been translated into English. They seem to me of interest to your non Francophile readers as well.

-Julien Benda La trahison des clercs, Paris, Grasset, 1927, Engl tr. The Betrayal of the Intellectuals, Engl. Tr. Norton 1955 and Transaction Publishers 2006

Gilles Gaston Granger, Essai d’une philosophie du style, Paris A. Colin 1969

Jules Vuillemin, Nécessité et contingence, l’aporie de Diodore et les systèmes philosophiques, Paris, Minuit, 1985, Engl tr. CSLI, Stanford,1992

Jean-Louis Gardies, L’erreur de Hume (Paris, PUF, 1987

Claudine Tiercelin, Le ciment des choses, Paris, Ithaque 2012

Richard Marshall is still biding his time.

Buy the book here to keep him biding!

First published in 3:AM Magazine: Friday, July 25th, 2014.