Version classiqueVersion mobile
OpenEdition Books

Metaphysical Knowledge

 | 
Claudine Tiercelin

Metaphysical Knowledge

Inaugural lecture delivered on Thursday 5 May 2011

Claudine Tiercelin
Traduction de Liz Libbrecht

Texte intégral

1Mr Administrator,
Dear Colleagues,
Dear Friends,

  • 1 Martial Gueroult, Philosophie de l’histoire de la philosophie, Paris, Aubier-Montaigne, 1979, p. 13

2“There was a time when it might have seemed presumptuous talking about metaphysics and of ‘system’”, wrote Martial Gueroult, one of my illustrious predecessors in this institution. And he added: “But whereas metaphysics and the spirit of system had already lost much of their worth by the late 18th century, at a time when Kant himself hesitated to pronounce the ‘name so decried of metaphysics’, they have now fallen to such a low level of discredit that whoever wished to defend them would risk being accused not of presumptuousness but of ignorance, naivety or stupidity1.”

3I am therefore fully aware of the boldness and perhaps the temerity or recklessness that it took for Pierre Corvol, Jacques Bouveresse and all of you, dear colleagues, to entrust me with the honour of holding a Chair which, for the first time in the history of the Collège de France, not only has the word “metaphysics” in its name but also associates it with the term “philosophy of knowledge” —so remarkably illustrated by the masters that Jules Vuillemin and especially Jacques Bouveresse have been for me.

It may indeed seem preposterous or audacious to propose an association of this nature, for in the minds of many, doubt is no longer permitted. Today the fate of what used to be “the queen of the sciences” is sealed, for we entered the age of post-metaphysical “thinking” long ago. Senseless dogmatism, verbose rambling of minds yielding to logical or onto-theological idols, or to fantasies of possible access to Reality, Truth, Knowledge: that is usually how we imagine metaphysics and ironize about its partisans, when there are any left. Disgruntled or supposedly lucid minds consider that certain “turns” (Kantian, Heideggerian, positivist, linguistic, cognitivist) have been taken, once and for all. Who would dare go beyond phenomena? Reality is inaccessible to us, caught as we also are, in any case, in the snares of language. Unless it was presented as research aimed at some celestial transcendence, or simply the reverse side of which theology is the right side, an approach that is still claimed, in the pure tradition of metaphysica generalis, to be investigating the essence of things or being as being would indeed seem ignorant, naïve or plain stupid. For the optimists or simply for those who, following their good sense, find that, in spite of it all, many things still remain knowable, and that it is conceivable, even possible, to hold them as true, it generally stands to reason that it is up to the numerous sciences to deal with it —something which, let’s admit, they do, and very well at that. And if, strictly speaking, there are things left “to think”, “to express” or “to live”, rather than “to know”, at least according to some, then literature and art take care of them, and even better, without any doubt, than metaphysics. Assuming that the latter still had meaning, what bone would be left for it to chew?

  • 2 Émile Meyerson, “De l’analyse des produits de la pensée”, Revue philosophique, Vol. CXVIII,no. 9-10 (...)
  • 3 Martial Gueroult noted that, incidentally, “the definition of philosophy as ancilla scientiae is eq (...)

4It is customary to be very pleased about the creation of a new Chair. I could be tempted to say that, in the present instance, this is not such good news. We may well wonder whether the reason for which it was necessary, for the first time, to introduce into the title of a Chair the term “metaphysics” —which all the philosophers who have made the history of the Collège de France, irrespective of their tradition, have naturally resorted to— might not actually be that, for a while now, we have completely forgotten what Émile Meyerson emphasized, that “man practices metaphysics just as he breathes”2. I will therefore not endeavour to define it and even less so to defend it here3.

  • 4 The Cambridge Edition of the Works of Immanuel Kant, Theoretical Philosophy, 1755-1770, Cambridge U (...)

5Paradoxically, one of the unquestionable characteristics of metaphysics is indeed the impression that one has, when practising it, of being able to “simultaneously achieve the greatest certainty possible”, while being incapable of giving or constructing a definition of its object. Kant noted that in metaphysics, contrary to mathematics, where I have absolutely no concept of my object, albeit confusedly, and I have to seek its distinct notion, “I have a concept which is already given to me, although it is a confused one. My task is to search for the distinct, complete, and determinate concept”4.

  • 5 This is how E. J. Lowe starts the book The Possibility of Metaphysics. Substance, Identity and Time(...)

6I would therefore like to use this lecture to refresh our memories and seek to understand why and how there is actually nothing inconceivable, and even less so improper, in choosing to associate metaphysics with the philosophy of knowledge. On the contrary, it is still possible today to envisage an authentic metaphysical knowledge, with distinctive methods and criteria of validation, and even to see in it a form of rational inquiry that is both legitimate and indispensable5.

1. Metaphysics “as one breathes”

  • 6 Claudine Tiercelin, “La métaphysique”, in: Notions de philosophie, Paris, Gallimard, 1995, p. 391 s (...)
  • 7 Étienne Gilson, L’Être et l’Essence, Paris, Vrin, 1st ed. 1948; 1972 ed., p. 55.
  • 8Dicemus igitur quod ens et res et necesse, talia sunt quae statim imprimuntur in anima prima impre (...)
  • 9 Étienne Gilson, op. cit., p. 56.
  • 10 Kant defined existence as the “absolute position of a thing”. Kant is one of those who insisted the (...)
  • 11 Étienne Gilson, op. cit., p. 14.
  • 12 Aristotle, The Complete Works of Aristotle, 4th ed., Barnes, Jonathan (ed.), Princeton, Princeton U (...)
  • 13 Cf. Pierre Aubenque, Le Problème de l’être chez Aristote, Paris, Presses Universitaires de France, (...)

7Since we have forgotten everything, or almost, we need to go back to basics6. Metaphysics is the science that deals with being as being. As Étienne Gilson pointed out, it is probably the only point on which metaphysicians agree7. If Avicenna is to be believed, no knowledge precedes being. The idea of being is imprinted in thought from the outset; it is the best known of all, the first and the most general notion accessible to us8. Difficulties start when we want to determine what a “being” is, and its specific properties, precisely in so far as it “is”9. For I can consider that being a being amounts to being (esse in the sense of existing), and I am content to position the thing absolutely10, in its act of being, for instance when I affirm that “God is or exists”, or that being a being means being “a being” (an ens11), that is, something that has the property of possessing being. But I can also use “being” to attribute a property or predicate to a subject. And what could be simpler: if I say that “the rose is white”, I have a linking verb, a neutral copula that enables me to predicate a quality (whiteness) of a thing (the rose). But this is where the problems start. For if, by making a statement such as “the rose is white”, I want to be able to say something that is neither nonsense nor pure tautology, I first have to ensure that I am predicating something of something (some being therefore, and not nothing). I then have to determine what the nature of these “somethings” to which the subject “rose” and the predicate “white” refer, and whether this nature is of the same order in both cases. Any analysis of being therefore has to provide the rules for the functioning of a meaningful discourse, be consistent with a language, and adhere to a certain logic. But it also has to specify the nature, that is, the condition of that which is in view of its very being, in short, the essence that makes the predication possible. As Aristotle so rightly reminded us: “It is not because we think that you are white, that you are white, but because you are white that we who say this have the truth12”. Hence, there can never be a radical break between logic and ontology. Compliance with the principle of non-contradiction is the first principle of the science of being. If being true is not part of the inquiry on being, it is because it has, in a sense, the same extension as being13.

8Assuming I said: “The present King of France is bald”. To all appearances, I am attributing something (baldness) to something, or rather to someone, the present King of France. But is that so sure? In order for this being that the present King of France is, to truly be an essence, is he not precisely lacking being? It is difficult to imagine how someone could derive their existence from their essence alone! And if there is no king of France, to whom exactly are we attributing baldness? To a being which is a lesser being because it does not exist now (even though it does after all remain possible, who knows, with the help of a little nostalgia)? To nothing more than a non-being, because it is a contradiction in terms? Or to a sort of half-being, to something that does have being but only in our imagination? Suppose that this being is not simply nothingness: it is after all something, a being of our minds, like a unicorn or Father Christmas. What reality can we grant it? For we will have difficulty agreeing that there is the same quantity of being in the present king of France or in a unicorn as there is in a triangle, even though the latter does not either have any existence other than in our minds. Why? Probably because, as much of a being of reason as it may be, and even though it is not a physical entity, a triangle does have certain, very real effects. We can, with the help of such a real or formal being, perform a demonstration, produce mathematical results which, in many respects, defy the idiosyncrasy of our thinking and therefore have a very specific form of reality, of independence. Any mathematician knows that. But which form exactly? What is the reality of a mathematical entity, a figure or a number? Let us take this further. We state that this king is bald. But, in so doing, what exactly are we attributing to this king? A property that is essential to him? Probably not, for it is surely not baldness that makes a king of a king. If we wish to find the essence of the king or, as Thomas Aquinas pointed out, that which enables us to say, with some reason, that Socrates resembles Plato more than he does a donkey, a cabbage or a line, we will probably have to look elsewhere for the foundations of his intelligibility. So would it be an accidental property? Let us agree on that, and even agree that this pure “accident” of baldness has an essence. Of what would it consist? Would it be the absence of hair? But who does not remember the sorites argument, of the bald man, the pile, and other vague predicates: from how many hairs is there or is there not baldness? From how many grains of sand will we have a pile?

9Let us turn back to our apparently more simple statement: “The rose is white”. There is no more risk of confusion here, linked to the existential dimension of being. We are merely predicating a quality of a substance, more precisely a general quality (whiteness) of a particular substance (the rose); in short a universal of an individual. But if the whiteness is in this rose here, how can it be in that rose there? In a sense, it has to be there really (otherwise, where would the difference be between a white rose and a red one?) and, in another sense, not completely; for there are other white roses, and not all roses are white, and finally, there are not only roses that are white. In what form, then, does whiteness in general participate in the white rose in particular, and in the rose as such? Since Antiquity, this is how one of the major problems of metaphysics appeared: that of the participation of One in the Multiple —a problem that Medieval philosophers examined in the form of what was to be called the problem of universals. They then investigated, even more closely, the nature of the distinctions needed to think the relationship between general and particular (is this a real distinction, based on reason, even formal?) and the nature (here again real, conceptual or nominal?) of universals, whether they were in favour of realism, conceptualism or nominalism. The Middle Ages were and still remain —this can never be over-emphasized— a particularly important time for philosophy, owing to the richness and variety of answers that philosophers such as Pierre Abélard, William of Ockham or John Duns Scotus, to mention but a few, endeavoured to propose to this fundamental triangular difficulty. We do —and how lucky we are!— indeed have the words, the signs; but how can we be sure that they are not pure flatus vocis, that they do effectively relate to concepts, to intellections, to products of a judgment or of a thought; and, finally, how can we be sure that these concepts themselves do indeed talk to us about things, the world, that which is? Words-concepts-things: the golden triangle.

10As we can see, being can be said in multiple ways, and that is where everything starts. But it is also where the ongoing risk of equivocation, contradiction or pure nonsense comes from. It is understandable that Parmenides preferred to maintain a strict position on this: being is, non-being is not. To progress on that which is, we have to make being talk. As Aristotle emphasized, metaphysics starts only once there is language, ontology, discourse on being and on its significations. That is also why metaphysicians have often been accused of over-indulging in words. We understand the temptation to bypass it by preferring intuition. But that would mean judging that these “words”, so closely bound to our pool of common knowledge, actually have little importance, or are at least guilty of spawning, all on their own, false problems or scholastic quibbling that should not be taken literally. Yet is that possible or even desirable?

  • 14 These two sentences are from “How to make our ideas clear”, Popular Science Monthly 12 (January 187 (...)
  • 15 For an excellent perspective on these problems in contemporary philosophy, see Denis Zaslawsky, Ana (...)

11One of the contributions to philosophy of the contemporary reflection of authors like Peirce, Carnap, Russell, Wittgenstein, Quine, Putnam or Davidson, has unquestionably been the tremendous revival of this theme. They have taken into account the fact that philosophy starts where there are “significations”, as Gilles Granger, a fervent reader of Aristotle, Peirce and Wittgenstein, pointed out: “to mistake a mere difference in the grammatical construction of two words for a distinction between the ideas they express”, and “to mistake the sensation produced by our own unclearness of thought for a character of the object we are thinking14”. These are also the two mistakes that Peirce said required particular attention, with the same energy as Wittgenstein who subsequently pointed it out. And since we cannot ignore our statements, we have to adopt the appropriate logical and linguistic tools for this therapy15. Indispensable organon.

  • 16 As realistic and hostile to positivism as he may have been, Meyerson considered that our quest for (...)

12It is therefore important to heed —and I do so myself— the recommendations of Jules Vuillemin and Gilles Granger: not only logic but also (as Jacques Bouveresse has stressed) the philosophy of language lend irreplaceable support to philosophy. The difficulty is then to determine its precise function and scope. If we grant such an important role to language, it is of course because what counts is not language itself but reality. And, unlike Vuillemin and Granger, who were more drawn towards a nominalism of axiomatics or structures, Bouveresse supports a realism that is firmly grounded in the philosophy of perception. But he has constantly reminded us that the risk —and, probably in his eyes, the most serious mistake in philosophy, especially if one is attached, as one should be, to defending truth and the demands of rationalism is probably of venturing to say more than we actually know. I therefore fear that he may judge my audacity guilty of just that, and may consequently be more than reluctant to follow the second part of the programme that logician, mathematician, chemist and metaphysician Charles Sanders Peirce proposed, unlike Wittgenstein in particular. Peirce argued that, once the false problems of traditional metaphysics had been eliminated, the method of “conceptual clarification” (borrowed more from logic or semiotics than from ordinary language) had to lay the foundations of a new theory of signification and knowledge, but in such a way that it placed itself at the service of a metaphysics which, once purified, should be able to be scientific and realistic. Now, preferring “silk” to “cotton16” —unlike the realist Meyerson—, it is this ambitious and in many respects extravagant programme that I have essentially adopted. Therefore I must, naturally, attempt to explain myself.

2. The Integration Challenge: Metaphysics and philosophy of knowledge

13If something can be said about being, it is that we can know it. Metaphysical realism implies cognitive realism. This might perhaps surprise a mathematician, spontaneously or, at least, one often driven to Platonism. After all, in mathematics, as in metaphysics, we are dealing mainly with abstract entities that cannot be located in space and time. They necessarily exist and are immutable, irrespective of the contingent details of the creation of the world. The knowledge of these entities is certain, a priori, and differs from the fallible knowledge of the other sciences. While the truth of mathematical statements depends on the properties of these entities, it does not in any way depend on our capacity (or incapacity) to know or to verify them. To hell with epistemology: ontology can take care of itself and of mathematics. As good Platonists, mathematicians will often consider these entities to be real in proportion to their inaccessibility. And if, by any chance, someone dares to ask them how such knowledge is possible, it will suffice them to answer, as Gödel would, by invoking the famous, mysterious and calamitous “intuition”. Actually, this type of answer bothered Plato already. In the Theaetetus, he pointed out that it would be difficult to say of someone that they knew, if they were not able, apart from showing that their knowledge or opinion were true, of attaching it to reason, in short, of justifying it. And try as we may, it is easier to say of a child that she knows Pythagoras’ theorem if she is able, not only to recite it but also, on request, to prove it and thus to explain why and how the theorem is true. Furthermore, how are we to understand the fact that mathematical knowledge is so successfully applied in the time-space domain of the physical things of the world in which we live and where we assume that any knowledge concerning them implies a minimum of causal interaction with them? As Quine and Putnam would say, we would like it to be “indispensable” to postulate mathematical entities, just as it is indispensable for the physicist to postulate electrons or other entities which are basically unobservable; but we do nevertheless expect a semblance of justification. Otherwise, what are we supposed, quite simply, to refer to? —especially since the dichotomies that the positivist logics upheld so strongly, between the a priori and the a posteriori, the analytical and the synthetic, formal sciences and natural sciences, have largely lived on. To be sure, we have for a long time associated, on the one hand, the a priori, analytical and necessary, and, on the other, the a posteriori, contingent and synthetic. Since Kant, we have known that certain judgments can be synthetic a priori. Quine and Putnam have taught us that the clear-cut distinction between analytical and synthetic should be revised, and that logic and mathematics are also empirical sciences, more so than we think. Discussions in the seventies around Kripke’s work reinforced our doubts as to the validity of some of our modal “intuitions” on identity, necessity or essentialism. It seems that some necessary truths are themselves knowable only a posteriori. Thus, we can conceive that water is not H2O, yet it is impossible for it not to be; it is necessarily so, even though this is a necessary truth that we discover a posteriori; but its necessity is as unquestionable as that of such analytical and a priori statements as “water is water” or “H2O is H2O”. Likewise, there can be contingent a priori truths such as “the standard measure at Sèvres measures one metre”.

  • 17 Charles Sanders Peirce, The New Elements of Mathematics, C. Eisele (ed.), La Hague, Mouton, 1976, v (...)

14There is no certainty that the only answer to the dissatisfactions of Platonist realism consists in sinking into a particular form of antirealism, as in the case of mathematics, where intuitionist, formalist or finitist reactions were quick to occur. Mathematical entities, it was said, are not abstract entities; one should rather see them as mental constructs or as physical entities or symbols of a language, and adopt one form or another of nominalism. Nor is there any certainty that these answers suffice to explain, for example, the trivial yet mysterious fact that “although mathematics deals with ideas and not with the world of sensible experience, its discoveries are not arbitrary dreams but something to which our minds are forced and which were unforeseen17”. It is, in reality, perfectly conceivable, even if we consider it questionable, not to opt for antirealism and, on the contrary, inspired notably by the extraordinary subtlety of the solutions proposed by the Medievals, and without reverting to Platonism, to now defend a substantial realism that is perfectly adapted to the requirements of modern science and, more generally, those aimed at knowledge.

  • 18 Christopher Peacocke, Being Known, Oxford, Clarendon Press, 1999, p. 1-2.
  • 19 Ibid., p. 1.
  • 20 Christopher Peacocke, The Realm of Reason, Oxford, Clarendon Press, 2004, p. 267. On this point see (...)

15On the other hand, it seems to me that the lesson that can be learned from the misadventures of Platonism is indisputable. One of the questions that any cognitive undertaking —and not only mathematical knowledge— faces acutely is the one that Christopher Peacocke called the “Integration Challenge”: in other words, “the general task of providing, for a given area, a simultaneously acceptable metaphysics and epistemology, and showing them to be so18. How can one “reconcile a plausible account of what is involved in the truth of statements of a given kind with a credible account of how we can know those statements, when we do know them19”? Peacocke, rightly so, considers that this binds metaphysics and epistemology together and that “[f]ailures to achieve the integration, in a given domain, of our metaphysics with our epistemology characteristically show up as manifestly defective accounts of knowledge of what it is for a content concerning that domain to be true”. Being able to understand that “would be to have a key not only to the epistemology and metaphysics of a domain but to the nature of ourselves as rational thinkers20”.

  • 21 Jacques Bouveresse, ibid., p. 399. [our translation]
  • 22 Ibid.
  • 23 Charles Renouvier, Les Dilemmes de la métaphysique pure [1901], Paris, Presses universitaires de Fr (...)

16This is also a point that Jacques Bouveresse has stressed, and that he considers to be crucial if we want to be able to satisfactorily answer the question of what something like “philosophical knowledge” consists or consisted of: “The concept of philosophical truth, as explained by those who use it, should not make incomprehensible and completely mysterious or unlikely, with regard to our idea of what knowledge in general is, the way in which we can understand the truth in question with the means that we are supposed to have at our disposal. And conversely, there should not be complete discordance nor even too wide a gap between what we know about the processes through which beliefs and philosophical certainties are formed, and our idea of the kind of truths that we are supposed to know owing to them. ”21 I am not so sure that I am as pessimistic as Bouveresse as regards the answer that he thinks has to be given. “It is probable that in reality” he writes, “we have perfectly clear ideas neither on the meaning that can be considered as truth in the case of a discipline like philosophy, nor on the means of knowledge that afford us access to this kind of truth”22.  But if there is one point on which I am sure I agree with him, it is this: the philosopher (although, as we have seen, he is unlikely to be alone…) does not actually have any alternative but to face this challenge at some stage, at least if he sets himself —which seems desirable— a programme that is deliberately engaged in a “rationalist” direction. And if that applies to philosophical knowledge, it clearly applies a fortiori to its most speculative and abstract part, in appearance at least, which is metaphysics. We are repeatedly told that the knowledge of metaphysics surpasses our means —for in this field we are in the presence of truths of a different nature to those encountered in science: transcendent, unverifiable, inaccessible truths, all of which defy, in their integrity, sentient experience and the space-time domain, and which therefore could be neither hypothetical nor a “matter of belief”, as Renouvier noted in Les Dilemmes de la métaphysique pure23.

17Perhaps, in spite of everything, we give up too quickly, unless we consider from the outset that the extraordinary vitality currently experienced by research in metaphysics is an illusion. But if we are reluctant to take seriously the very idea of the possibility of metaphysical knowledge, it is perhaps also because we are not yet fully aware of the evolution that has taken place in our very concepts of science and knowledge, or of the considerable yet recent growth in the field of the philosophy of knowledge. Before going any further, I therefore need to say a few words on the subject.

3. The Evolution of the concepts of science and knowledge, and the upheavals in contemporary philosophy of knowledge

  • 24 Ernest Renan, Essais de morale et de critique, Paris, Michel Lévy Frères, 3rd Edition, 1861, p. vii (...)
  • 25 Claudine Tiercelin, Le Doute en question. Parades pragmatistes au défi sceptique, Paris, Éditions d (...)
  • 26 Claudine Tiercelin, “The Fixation of Knowledge and the Question-Answer Process of Inquiry”, in F. L (...)
  • 27 Johann Friedrich Herbart, 1806-1808, Hauspunkte der Metaphysik, Sämtliche Werke (19 vol.), Langensa (...)

18We shudder at the idea that “metaphysical knowledge” implies a return to the system, understood, at the very least, in a dogmatic sense. In the preface to Moral and Critical Essays, Ernest Renan noted that defending a dogma means proving that we believe in it and therefore that we need it24. It is a sign of weakness. But for a long time now, dogma has no longer prevailed, neither in metaphysics nor in this model of knowledge that, for many and rightly so, remains scientific knowledge. Very few metaphysicians are still obsessed by the search for eternal, universal, overarching truths. On the contrary, the vast majority seek to understand our relationship with the real —something that we can do only by starting off from where we are, and not from “nowhere”. But in my opinion, it is also on this model of inquiry, and no longer on that of justified true belief or of the epistemology of virtues, that we must consider knowledge in general25, including metaphysical knowledge. However, the aim of an inquiry —the Socratic system of questions and answers, doubts and beliefs— is to fix the latter and not to provide an absolute and definitive truth26. Hence, for whosoever wants to contribute to elucidating the cement binding things together, or attempt to reconcile in some way the “philosophy of nature” and the “philosophy of intellect”, it is less a matter of proposing a system than of highlighting —as J. F. Herbart would put it— the “main points”27 of that which should constitute a realist scientific metaphysics worthy of the name.

  • 28 Gilles-Gaston Granger, “L’argumentation du Tractatus: Systèmes philosophiques et métastructures”, i (...)

19Not that it is necessary, as some suggest, to limit metaphysics to the analysis of our usual ways of talking about that which, in a non-reflexive mode, we see as features of the world in which we live; nor to a mere “meta-structure”, the object of which is the meaning, and which would constitute a substitute for the absent criterion of truth, as Granger preferred to envisage it28. We need to accept the idea that metaphysics is effectively the study of the most fundamental structure of reality (and therefore coextensive with ontology), that its goal is truth, and that, if it is still meaningful to ask the Kantian question of its possibility as a science, it is because it corresponds, as a rational inquiry, to certain characteristics that are sufficiently distinctive to be able to constitute the heart of an independent discipline, and whose intellectual justifications are worth being explored.

20Indeed, the idea is not only to refute the neo-Kantian claim for an impossible metaphysical “knowledge”, relayed by both positivist and Heideggerian analyses, whether it be the outcome of our cognitive limitations, human finitude, or of the particular nature of the objects of metaphysics or, simply, of its lack of object. But nor is it to reduce this “knowledge” to the model, inspired by science, of a metaphysics that is in some way scienticized or “naturalized”. For, irrespective of how closely they are connected, the possibility of a “metaphysical knowledge” or a “scientific metaphysics” largely transcends the question of the turbulent relationship between metaphysics and science.

  • 29 Patrick Suppes, Probabilistic Metaphysics, Oxford, Blackwell, 1984.

21Nevertheless, and to begin with, we need to know what is classified under the concept “science”. For only laziness or ignorance or, in certain cases, intellectual dishonesty can explain why some still grant credit to a positivist and scientist vision of scientific knowledge. No individual in good faith, and especially not a philosopher of science, could agree today that the concept of “science” or of “knowledge” which prevails, is the “imperial” one, still found in Kant’s work, of a complete, apodictic system, or one endowed with the virtues of necessity and universality. As Patrick Suppes noted some time ago already in Probabilistic Metaphysics29, the term has to be conceived of more in a form that forces us: to rethink not only certain “absolute” concepts (space, time, being), but also the probabilistic nature of laws, matter, and causality; to note the almost impossible access to certain knowledge because it tends rather to be approximate, temporary, conjectural, fallible; and to give more than its due to the complexity of phenomena, doubts, the idea of research proceeding by trial and error, rectifications, and falsification as much as confirmation or verification. It is therefore not a sign of hostility to science —on the contrary— to note that it is characterized not by scientism and dogmatism but rather, as all normal scientific activity attests, by uncertainties, trial and error and incompleteness, which together make the risk of scepticism particularly high (as I will discuss in the first series of lectures that I will be giving this year). And it is not surprising, in this respect, to see the extent to which this risk is present, perhaps even more frequently among scientists than among philosophers.

  • 30 I embarked on an analysis of this question in 2005, in Le Doute en question, op. cit.

22Hence, there is nothing surprising about the examination of the sceptical and idealistic risks being at the heart of work in the philosophy of knowledge over the past fifty years; or about the fact that volumes devoted to this inquiry are almost as numerous as those that, in their quest for convincing answers to the sceptical challenge, have proposed various strategies to improve the very definition of the concept, this time of knowledge, by showing the extent to which epistemology largely transcends metaphysics and ethics, so true it is that the struggle against scepticism has to be waged on all three fronts simultaneously30.

23Thus, how can we be sure of the reality of the external world if, as the Cartesian “sceptical scenario” shows, we lack the means to distinguish dreams from wakefulness or to show that we are not prey to an evil genius or, in its Putnamian variant, not merely brains in a vat? How can we be sure of the reality of the past, the existence of other minds, our (inductive) beliefs in the future, the reality of our perceptions? (Metaphysical challenge.) Are we therefore not dealing only with appearances, so that, all things being equal, we can agree to anything and everything? How then can we deal with the risk of neutralizing ethical choices through indifference? Must we give up believing in the existence of certain moral realities, in the possibility of proceeding in ethics as in science to all theorisation, evaluation, justification? And even if, in ethics, we give up on realism, can we avoid relativism, and even the cynicism that often stems from it? Do we not talk of epistemic responsibility or intellectual virtues just as we talk of ethical responsibility or moral virtues? Is it absurd to envisage an ethics of belief or of the norms of rationality? (Ethical challenge) How can we escape Agrippa’s classical challenge —taken up by Sextus Empiricus in the Outlines of Pyrrhonism and sometimes called the “Münchhausen trilemma” or “Fries’ trilemma”— which states that any definition of knowledge must either lead to an infinite regress in justification, or stop randomly in this regress and lead to dogmatism, or else lead to a circular definition? (Epistemological challenge) How, finally, can we improve the definition of knowledge proposed by Plato, but which he himself maintained was aporetic, by avoiding the obstacles that all, or almost all, the contemporary theories proposed over the past three decades have stumbled against?

  • 31 Themes highlighted mainly by such authors as Peirce and Wittgenstein; see Claudine Tiercelin, Le Do (...)

24In this young continent that the philosophy of knowledge is, we can now, very schematically, distinguish at least four main streams which already have their classical texts and reference authors: first, those that continue to situate themselves against the background of the Platonist analysis of knowledge which I have already mentioned, essentially as true belief endowed with reason or justification; second, those who follow an Aristotelian model, in what has come to be called “virtue epistemology”, to highlight the importance of the intellectual agent and the role of educating the agent’s aptitudes or dispositions in epistemic virtues; third, in a more Lockean mode this time, those who see knowledge as a primitive mental state; and, fourthly, those situated against a more “pragmaticist” or “realist” background, where knowledge is understood essentially as a dynamic process of investigation (inquiry) involving forms of practical knowledge (knowing how) just as much, or even more than forms of propositional knowledge (knowing that), together with the perception of values, and where —at least as much as an analysis of the conditions of possibility of knowledge— an analysis of the conditions of possibility of doubt itself31 and of normative rationality is indeed required.

25All these recent developments in the philosophy of knowledge are huge steps forward and indications of the intense vitality of the discipline today (some of the main themes of which I will address by focusing more particularly on an examination of the value of knowledge).

26But they also imply obligations for any philosopher concerned about better defining the relations between the knowledge he may pride herself on and that which characterizes the other fields of knowledge —and, in particular, the obligation to respond appropriately, as I have said, to the Integration Challenge. Hence, anyone seeking to determine the constitution of metaphysical knowledge cannot— neither more nor less than anyone else —exempt themselves from this. It is therefore also necessary to question the types of beliefs, “truths” and justifications that we may be dealing with in metaphysics. Are they beliefs whose truth stems from common sense? Scientifically established truths and therefore necessarily contrary to the “manifest image” reflected to us by the world? Or beliefs that are really of a very different nature? Either way, what reasons, what justifications do we have for maintaining these beliefs, for favouring one or another conception of truth, for judging the knowledge of the things that it reveals to us as merely conceivable, or possible, or even necessary?

27By following these recommendations and by favouring the fourth model mentioned above, that of rational inquiry, I now wish to indicate, at least very broadly, how and why metaphysical knowledge seems to me, not only legitimate but in many respects indispensable for whosoever wants to get rid of idealism, relativism, irrationalism and, even more so, of the sweeping cynicism which often goes hand-in-hand with them today.

4. Metaphysical knowledge: the four steps

28Four equally important steps are necessary here.

  • 32 For details on this step, the reader is referred to my book Le Ciment des choses, Paris, éditions d (...)

29Any metaphysical inquiry requires that one start from the formal aprioristic framework of analysis, which alone allows one to obtain the conditions of possibility —in terms of conditions of truth and meaning— of the concepts that we use, and to be careful, in particular, to make the necessary modal distinctions32.

  • 33 Cf. ibid., Chap. 2, “Les pièges du scientisme”, p. 97-186.

30It is then necessary to see whether, and in what sense, the categories of our thinking and our language are not simply “functions of judgment”, as Kant believed, but well and truly the mirror of the categories of reality. Thus, metaphysical inquiry implies an a posteriori investigation and a comparison with the empirical sciences33.

  • 34 Cf. ibid., Chap. 3, “Le réalisme et ses mirages”, p. 187-246.

31It is then necessary —and this is the strictly epistemological part of the Integration Challenge— to reflect on the way in which, in turn, our categories and judgments constitute knowledge a posteriori, by comparison with the sciences which are already constituted, but also, by seeking to determine whether and for what reasons, for example, scientific theories are true, can be labelled justified knowledge, and demand or not a commitment to scientific and epistemological realism. This is the third step34.

32Finally, it is also important, in the fourth step, to examine the type of reality that these theories are talking about, the nature of the actual properties (especially categorical or dispositional) constituting it, the causal and nomical relations between them, and that which is necessary to guarantee the unity, in short, to make “the cement of things”. And then comes the time for truly metaphysical commitment.

33I will be content here, in the time left to me, to give some brief indications on the importance of the first two steps and on the way of envisaging them.

34The first step is thus the therapeutic one. We analyse our concepts, not to describe their grammatical or logical structure, but to refine them and, if need be, revise them. If, unlike Kant, we agree that metaphysical knowledge can go beyond phenomena and do more than inform us on certain necessary characteristics of what we think about it —in other words, that it can, at a given point in time, account for what reality per se consists of—, then conceptual analysis will have to be set within a project of revision and not of mere description.

35In metaphysics, we seek less to say what there is than to determine the particular mode of being of such and such a thing. It is clearly properties, and not mere words, that we are looking for. Conceptual analysis will therefore not be reducible to mere linguistic or logical analysis. Nor do we inquire about just any kind of properties. When we ask the question “What is a person?”, for example, we do not need to hear ourselves say that every person has a profound desire to be loved, even if that is the case. Of a property, we wonder what makes it have an impact on what an object is, in short, on what its essence is and what justifies the distinctions that we make between essential properties (or, let us say, definitional ones) and accidental properties. We therefore first have to elucidate what governs our classificatory practice, and clearly locate and count things by introducing no more entities than are necessary.

  • 35 John Duns Scotus, Ordinatio, I, 3, § 81. Opera omnia (Civitas Vaticana: Typis Polyglottis Vaticanis (...)
  • 36 François-Xavier Putallaz, “Introduction” to: Jean Duns Scot, Traité du premier principe, traduction (...)
  • 37 John Duns Scotus, A Treatise on God as First Principle, Kessinger Publishing, § 26.
  • 38 By focusing on the quidditive being, Duns Scotus did not have in mind an “essentialisation of metap (...)
  • 39 See for example: A Treatise on God as First Principle, op. cit., § 25.

36But we have also known, since Duns Scotus, that the autonomy of metaphysics has been allowed by the fact that it is about, not so much what is, as what might be, in short, about the possible. If metaphysics can become autonomous in relation to the other sciences (logic, physics, but also theology35), it is because, for the Subtle Doctor, it has its own object, the ens commune, i.e., being taken in its total indeterminacy, which is not reducible to either the quiddity of the sentient thing (in which it nevertheless has to contract), nor to logical predictability alone (which is nevertheless the only thing able to confer universality on it). It is thanks to this that it is able, over and above the opposition between being and the possible, to ensure the “real-possible”, that is, the very reality of the possibility of being of things which exist and, therefore, of the condition of possibility of the sciences as sciences36. Hence, scientific rigour demands that we start from the possible, the only place that can cover the domain of the contingent existent and that of the necessary, or of metaphysical quiddity37. And, in order to do so, there is no other method than that consisting in reasoning through the possible, which does not mean (contrary to what posterity has often accused Duns Scotus of) deducing the first principle by analysis at the end of a conception developed on essences38, but trying to reveal the internal structure of the real possible, in so far as it is true that it is rooted in the concrete real, in such a way that we can infer the former from the latter39.

37It is probably experience that will be able to teach us, afterwards, which of several metaphysical possibilities has the most chance of being true in reality. But experience alone cannot determine what is real in the absence of a metaphysical delimitation of the possible. If we were not able to draw a line between possibilities, independently of the way in which the world presents itself, we would simply be unable to say what it looked like. When we describe reality, we spend our time recognizing and excluding possibilities. We spontaneously experience it, for example by reading a map, to determine where the eye of a river is, or the closest town, or by making a statement as simple as “the train leaves at six o’clock”. In this respect, the possible is as natural to us as the actual. It is an unavoidable determinant of reality.

  • 40 We can thus agree that we can methodologically use the possibilia without presupposing, at least at (...)
  • 41 Tamar Szabo Gendler and John Hawthorne (eds), Conceivability and Possibility, Oxford, Clarendon Pre (...)

38Contrary to a widespread idea, modal logic and epistemology are far less abstract than they are believed to be. In fact, it is on them that we should rely, and on the models and thought experiments that they propose, in particular by applying our intuition —not intellectual intuition but intuitions about possible cases40— in order to dispel the modal illusions of which we constantly fall victim. Indeed, we know that, depending on the conception that we have of the possible, for example, we will more or less readily consider that which is conceivable (at least understood in the sense of something not of the imagination but of understanding alone) as a reliable guide or not as to what is possible. But something can be a priori conceivable and yet impossible. That which is conceivable is therefore not enough, at first sight, to ensure that something is epistemologically possible or impossible. If I know that the cat is on the mat, even if I can perfectly conceive of a situation in which it is not, it is not epistemically possible for me to say so. And that which is logically possible (or not contradictory) is not necessarily physically or metaphysically possible (taken in the absolute sense where God, for example, might have wanted things to have been different). Hence, it is not possible, in any of these senses, for something to be red and not red; it is logically but not metaphysically possible for something to be red and not extended; it is metaphysically but not physically possible for something to travel faster than the speed of light; and it is possible, in all three senses of the word, for something to travel faster than a space shuttle41. We can also deem logically possible or necessary, in the broad sense, that which is true in any logically possible world, that is, in any possible world. Logically necessary will then be coextensive to that which is metaphysically necessary. Some will object that the proposition “Water is H2O” can be untrue in all logically possible worlds. Others will however reply that it is not only the laws of logic, associated to concepts or definitions of water and H2O that make “Water is H2O” logically necessary: it is the nature of water, whose composition we now know. We understand why in fine, logical necessity will, in this case, tend to be confused with metaphysical necessity.

39Conceptual analysis is therefore valuable because, while it does have an a posteriori and empirical part, it is also able to obtain a priori results, by broadening the usual function of our concepts through the means of the convenient possibilistic models of modal logic.

40Thus, “modal knowledge” teaches us, for example, that the extension of a term in an actual world and in a counterfactual world is not the same; and therefore that in order to know the C or Counterfactual-extension of a term, we need to know something about the actual world. When the A or Actual-extension and the C-extension of a term differ for certain worlds, there is between them a crucial difference of epistemic status. Even if, for example, we understood the term “water” before 1750 (and Dalton’s chemistry), we did not know its C-extension to a world, for some world other than the actual world. By that I do not necessarily mean that we were unaware of its essence, but that, to identify water in a counterfactual world, we need to know something about the relations that exist between the counterfactual and the actual worlds. This was possible only after the discovery that, in the actual world, H2O was the watery stuff. On the other hand, we knew the A-extension of “water” for any world, as it did not depend on the nature of the actual world. Our ignorance of the actual world was therefore of no import as regards the knowledge of the A-extensions of words, since we did not know the A-extension of “water” in the actual world before 1750, and were nevertheless capable of identifying the A-extension of “water” in this world and, in fact, in all worlds.

41Now, what one can know independently of what one can know further about the world can be said a priori. The answers to the questions pertaining to C-extension generally depend on the nature of the actual world and are a posteriori; but the a priori part of conceptual analysis does not in any way depend on the question of knowing which world is in fact the actual world (just as the question of knowing what to do if the weather is good does not depend on the question of knowing whether the weather is good or not).

42Hence, conceptual analysis enables us to define and identify what we are talking about; to answer the question of whether the way in which things are, put in a particular vocabulary, makes an explanation in another vocabulary true. And, once again, this is not insignificant when one thinks about how important it is to have clear ideas in metaphysics, and —as Peirce and Wittgenstein in particular exhorted us— not to mistake a small difference in our words or in our ideas for a difference in things. It therefore seems, at least in the indispensable therapeutic step, that it is in our interests to adopt it in metaphysics.

  • 42 For a brilliant illustration of this way of proceeding, see the book by Frank Jackson, From Metaphy (...)

43But it is particularly valuable in so far as it does not in any way dispense with —and even goes hand-in-hand with— the method of tests or “energetic massaging” to which we subject our folk intuitions. Basically, the question here is deepening reflection in the style of that which Locke inaugurated, that of thought experiments, and then of testing various possible solutions based on the network of (often incompatible) intuitions that we have constituted, to see those that need to be preserved or eventually excluded. Hence, in some respects, the metaphysician is doing nothing other than what the developmental psychologist or the political scientist undertakes. Like the biologist, economist or physician, he gathers the elements of the folk (or popular) theory that are going to constitute a “network” of principles and, like them, sets out to test them and, if necessary, to revise them —by finding out, in particular, about the contradictory nature of these intuitions, for example on personal identity or change, whether one adopts a three-dimensional perspective (by representing an object existing at a time as totally present at that time, or “enduring”), or a four-dimensional perspective (the object existing at a time by having a temporal part at that time, in short, by “perduring”). The fact that folk intuition is not sacrosanct also means that, if such a concept has proven itself, this does not imply that it would be irrational to change it, in light of what we learn from reflection or empirical discoveries, or at least to make “reasonable adjustments42”. In short, and this is not insignificant, in metaphysics —a science formerly seen as universal, necessary and apodictic— we have to accept the principle of fallibilism.

44We have to maintain this link between metaphysics and psychology. And it is important because, whether we like it or not, our beliefs are indeed the product of evolution and are driven more by the practical demands of survival than by the theoretical demands of metaphysical truth. This also means that metaphysics, as engaged as it may be in a project of revision, has to be attentive to what the philosophy of mind, psychology and the cognitive sciences have to teach it —even if that does not imply that it will not have a critical attitude towards them. Here we may well envisage a kind of harmony between man’s cognitive faculties, on the one hand, and reality, on the other. At the very least, it is necessary to reflect upon the nature of the links that may exist between that which is natural in us and that which is normative, as Kant had moreover perceived, in the form of what he called a “system of preformation of pure reason”, the third way that he had already considered in metaphysics.

45As Duns Scotus had so clearly seen, proceeding by the logical possible is therefore not merely a methodological precaution: all unintelligibility, all logical impossibility is in fact usually —whether Descartes would have liked it or not— the sign of a real impossibility. While the real possible is not reducible to the logical possible, they are not either foreign to each other, otherwise our concepts would be nothing but words, and our science would be devoid of all objective content.

46On the other hand, we cannot confuse the logical possible with the real possible. And the constant pitfall that needs to be avoided is the very one that Duns Scotus is accused of falling into: constructing a metaphysics where non-contradiction would be enough to say the properties of being. The only fear that the immodest adoption of conceptual analysis might arouse would surely be this: pursuing too doggedly the path that some have treated as essentialist (going from Suarez to Wolff); wanting to analytically deduce the real based on essences conceived of as possibilities; forgetting that ordinary language (or modal logic) is not an adequate guarantee of the reality of classes in nature, for example; or justifying the logical procedures themselves by describing them in a more or less acknowledged —and, according to some, shameful and irretrievable— form of essentialism.

47Exaggerating the absence of a connection between the a priori aspect of the analysis and its a posteriori aspect would therefore be disastrous. And this is not only due to the risk of ethereal or vaporous metaphysics that it would entail, but also to the fact that logic and semantics themselves are always to some degree linked to our “empirical reasons” and therefore to our scientific discoveries.

48Second step. It is therefore advisable, in the next step, to determine the exact place that remains, in this context, for the a posteriori dimension of the metaphysical project itself —in other words, for the empirical sciences— in the determination of what is.

49It is probably not so easy: and our contemporary Averroists readily say that the sciences are in the best position and the most legitimate to tell us what the reality of things consists of. If there are fundamental questions to ask on the existence and the nature of space and time, it is up to cosmology and quantum physics to answer, and not to armchair speculative metaphysics or a metaphysician “in a redingote”.

50To conclude, I will therefore propose a few principles and rules of conduct to facilitate life together.

  • 43 Hans Reichenbach, The Rise of Scientific Philosophy, Berkeley, University of California Press, 1957 (...)
  • 44 As proposed today, for example by James Ladyman, Don Ross, David Spurrett and John Collier, Every T (...)

51First principle: Reject scientism and, as Hans Reichenbach wrote in 1957, refuse to consider outright that what a “scientific philosophy” needs is simply a “reorientation of its philosophical desires43”. Stop claiming, as certain scientist metaphysicians are tempted to do, that the exclusive motivation for metaphysics is the “attempts made to unify the hypotheses and the theories which are taken seriously by contemporary science44”.

  • 45 Moritz Schlick, “Le vécu, la connaissance, la métaphysique”, in: Antonia Soulez (ed.), Manifeste du (...)

52Second principle: Remember that explaining —and this remains an objective— means aiming for a communicable content, as Moritz Schlick45 emphasised, but it also means reducing. It would be excessive to reconsider the successes which have thus been made possible in our knowledge and our mastery of nature.

53Third principle: No serious metaphysician can get away from the explanatory principle of causal and nomological closure of physics (the laws of which are universal and fundamental), to which the special sciences are subjected, nor from that of an ontological (or even minimal epistemological) physicalism, if by that we mean the basically trivial idea that all things which exist in this world are pieces of matter and structures from the aggregate of those pieces of matter, all behaving according to the laws of physics.

54Fourth principle: Do not be too hasty to join the “non-reductionist consensus” to which certain dilemmas can lead, such as the physicalist’s dilemma of the status of mental properties. Endeavour rather to avoid fear of reductionism, by seeking new models that make it possible to distinguish reductions from pure eliminations or, as is increasingly the case, to accomplish procedures of reduction capable of highlighting inter-level, rather than intra-level, mechanisms between the different sciences,

55A few rules will then help the metaphysician and the scientist to function together.

  • 46 Wilfrid Sellars, “Empiricism and the Philosophy of Mind”, in: Wilfrid Sellars, Science, Perception (...)

561. Acknowledge, along with Wilfrid Sellars, that when it comes to describing and explaining the world, science is indeed “the measure of all things, of what is that it is, and of what is not that it is not” at the fundamental level of things46. On the basis of all the reasonable criteria of interpretation of scientific theories, this gives us, as a negative image, a first principle to adhere to: there is unquestionably a considerable number of philosophical positions that, in the current state of knowledge, we can no longer support, even if we also have to bear in mind that the unity of science remains an ideal, that the causal completeness of the physical has not been achieved or is closed only in principle, and that it can be considered as a “methodological norm” but not as an “analytical” principle.

572. But, and it is philosophy of science itself that teaches us this, there is never a royal way leading from a physical theory, a mathematical theorem or experimental results, directly to fixed and necessary metaphysical consequences. This is what enables us to put forward a second principle, positive this time. As experience alone does not have the logical strength to determine scientific theory, or even the interpretation of a given scientific theory, we should, through the evaluation of the ontological consequences of the various positions, be able, from within the empirical situation, to propose certain metaphysical consequences, and then to evaluate their cost and/or relevance.

  • 47 Wilfrid Sellars, “Philosophy and the Scientific Image of Man”, in: Wilfrid Sellars, Science, Percep (...)

583. It is therefore not only for aesthetic reasons of coherence that there is room for metaphysics, nor because we aspire to systematic ambitions or dream of unity or of an absolute position. We need it, in the strong sense of the word, and above all, to interpret scientific theories themselves. There is autonomy of metaphysics and irreducibility, in certain respects, to what is done in science, including to the metaphysical problems peculiar to science itself, concerning for instance: the choice that has to be made between scientific realism or instrumentalism; the conception to adopt on the laws of nature, natural kinds, etc.; or the question of whether, while science informs us on surprising things and no longer models the world in terms of objects, it invalidates or not the daily status of objects such as tables and chairs; or, if there is still sufficient cohesion on our scale, in certain aspects of the world, for the group of cognitive systems that we are, driven by practical interests, to be able to know our way around in things and to categorise them for everyday life. Hence, the utility and even the indispensability of a “stereoscopic” point of view, “where two different perspectives on a landscape are fused into one coherent experience”. So true it is that “[w]hatever the ultimate constituents of conceptual thinking, the process itself as it occurs in the individual mind must echo, more or less adequately, the intelligible structure of the world47”.

594. Let us add a fourth rule. As I have mentioned, science is an undertaking of categorisation. But neither the phenomena observable on the macroscopic scale, nor the scientific information concerning the internal constitution of things, can enable this categorization to succeed without a specifically metaphysical delimitation of things.

  • 48 E.J. Lowe, A Survey of Metaphysics, Oxford, Oxford University Press, 2002, p. 174 sq.

60Hence, we should already have an ontological categorization of substances and the conditions in which, for example, a living organism can metaphysically remain identical through a certain change (the tadpole that turns into a frog), to be in a position to decide, irrespective of the number of empirical trials that we have, whether this change has to be classified as a transition phase or as a substantial change48.

61Here we see just how important it is to distinguish, in questions of this nature, that which is strictly-speaking the a priori contribution of metaphysics and the a posteriori contribution of the empirical sciences.

62Finally, being complete is one thing, being fundamental is another. Now, from this point of view, and despite our unquestionable ontological physicalist intuitions, the principle of causal closure is ontologically neutral as to what does or does not lie at the fundamental level of things; in other words, as to what essentially defines physics and, in particular, as to whether the levels of reduction that we can agree to with regard to method are or are not on an ontological level. What is the world made of? Of several layers, strata, levels of reality, or only one? Of several levels of properties (categorical and dispositional, the latter being reducible to the former) or only of one of them? As we have seen, it is neither conceptual analysis alone, nor science that will enable us to answer these questions, which is why we cannot fully talk of “metaphysical knowledge” until at least the other two steps in the process set out above have been completed.

5. To conclude

63Metaphysics is a discipline that is too arid to be able to be practised alone. Even Peirce, who spent most of his life as a hermit, had founded a Metaphysical Club. Although French Academia still has a long way to go, especially in philosophy, to return to the paths of an authentic academic community, I would not like to end this lecture without mentioning my indebtedness to university teaching and especially to my students, without whom I would not be here before you today. By separating teaching from research institutions, as we still do all too often in France and unfortunately increasingly often, we are doing no favours to either of them. In this respect I have always agreed with my professors, Jules Vuillemin, Gilles Granger and Jacques Bouveresse, and I am aware of how much I owe them, both intellectually and for having kept the ideals of the University, as an institution, alive. I also owe a great deal to the university departments where I was fortunate to work, both in France and in the United States. But I also owe much to the welcome I received, at a time when my doubts on the University were not in question in a team such as that of the Jean Nicod Institute, and to all my friends who turned it into a unique space, especially Joëlle Proust. Finally, I hope it will be acknowledged that Pascal Engel’s work counted as much as it counted for me.

  • 49 Ernest Renan, Essais de morale et de critique, op. cit., preface, p. iii.

64The Collège de France is certainly not a club. It is an institution —and what an institution— but it is also a company that spontaneously envelopes in its generosity all those who have the extraordinary privilege of belonging to it. Whoever has this honour has, above all, duties, and the first of those duties is never to proceed behind a mask. Ernest Renan noted that this was an obligation inherent in any right to the freedom of thought: “For truth has its rights just like the good, and we gain nothing from those timid lies that fool no one and result in nothing but hypocrisy49”. Like the philosopher from Tréguier, that will henceforth be my only requirement.

65Even though, or perhaps because, he was profoundly pious, Peirce considered that the “puny, rickety and scrofulous” state in which metaphysics was in his day, stemmed from the fact that it had remained for too long in the hands of those whom he called “professional theologians” or “seminarian philosophers”. And he urged us to remember that metaphysics is and must remain above all “a science of observation” to which it is nevertheless necessary to apply mathematical exactitude. These are simply observations that are so pervasive that, notwithstanding their importance, we no longer pay any attention to them.

66The last century held our breath for too long. Why don’t we, simply, start “breathing” again?

Notes

1 Martial Gueroult, Philosophie de l’histoire de la philosophie, Paris, Aubier-Montaigne, 1979, p. 13.

2 Émile Meyerson, “De l’analyse des produits de la pensée”, Revue philosophique, Vol. CXVIII, no. 9-10, September-October 1934; quoted in Émile Meyerson, Essais, Revised edition by B. Bensaude-Vincent, Paris, Corpus des Œuvres de philosophie en langue française, 2008, p. 156.

3 Martial Gueroult noted that, incidentally, “the definition of philosophy as ancilla scientiae is equally hazardous for it as the definition of philosophy as ancilla theologiae. […] Some defend Thomas Aquinas, some defend Einstein, but who will defend philosophy? […] If ancilla scientiae is to be defended in a sort of scientific spirit, and ancilla theologiae in a sort of theological spirit, it is by philosophy that philosophy has to be defended, and by metaphysics that metaphysics has to be defended” (op. cit. p. 18-20). [our translation]

4 The Cambridge Edition of the Works of Immanuel Kant, Theoretical Philosophy, 1755-1770, Cambridge University Press, 1992, p. 256.

5 This is how E. J. Lowe starts the book The Possibility of Metaphysics. Substance, Identity and Time Oxford, Clarendon Press, 1998, p. 2.

6 Claudine Tiercelin, “La métaphysique”, in: Notions de philosophie, Paris, Gallimard, 1995, p. 391 sq.

7 Étienne Gilson, L’Être et l’Essence, Paris, Vrin, 1st ed. 1948; 1972 ed., p. 55.

8Dicemus igitur quod ens et res et necesse, talia sunt quae statim imprimuntur in anima prima impressione, quae non acquiritur ex aliis notioribus se […].” (Avicenna, Metaph. Tract. I, cap. 5; in Avicenna latinus Liber de philosophia prima sive scientia divina, I-IV, critique of the medieval Latin translation of S. Van Riet, Louvain, E. Peeters-Leiden, E.J. Brill, 1977, p. 32-33.

9 Étienne Gilson, op. cit., p. 56.

10 Kant defined existence as the “absolute position of a thing”. Kant is one of those who insisted the most on this duality of being: “Being is evidently not a real predicate, that is, a conception of something which is added to the conception of some other thing” (p.410); and “By whatever and by whatever number of predicates —even to the complete determination of it— I may cogitate a thing, I do not in the least augment the object of my conception by the addition of the statement: This thing exists.” (p. 411) Kant, I. The Critique of Pure Reason, translated by J. M. D. Meiklejohn, A Penn State Electronic Classics Series Publication, 2010.

11 Étienne Gilson, op. cit., p. 14.

12 Aristotle, The Complete Works of Aristotle, 4th ed., Barnes, Jonathan (ed.), Princeton, Princeton University Press, 1991, p.1051 b6–8.

13 Cf. Pierre Aubenque, Le Problème de l’être chez Aristote, Paris, Presses Universitaires de France, 1972, p. 167. In this sense, both Heidegger’s and Brentano’s interpretations of truth in Aristotle’s work are equally unreliable. Truth is neither a matter of revealing nor simply of attributing logic. Talking of the truth of things simply means that the truth of human discourse is always prefigured, or rather pre-given in things, even if it is revealed only in our discourse on it.

14 These two sentences are from “How to make our ideas clear”, Popular Science Monthly 12 (January 1878), 286-302.

15 For an excellent perspective on these problems in contemporary philosophy, see Denis Zaslawsky, Analyse de l’être, Paris, Minuit, 1982.

16 As realistic and hostile to positivism as he may have been, Meyerson considered that our quest for the profound essence of things —especially for the mysterious causal power, the “glue”, the “cement” that binds effect to cause— is “vain” because actually “nothing like it exists”. He therefore stopped midway. Refusing to follow through to the end the metaphysical routes of the “philosophy of nature”, he acknowledged his preference for “cotton” rather than “silk”, and chose the routes (nurtured by the history of philosophy and of science) of the “philosophy of the intellect”. In this way he illuminated the “pathways of Thinking” and —too partially, in my opinion, despite its revival— the link between nature and intellect. See “Philosophie de la nature et philosophie de l’intellect”, Revue de métaphysique, t. XLI, no 2, 1934; cited in Émile Meyerson, Essais, op. cit., p. 121.

17 Charles Sanders Peirce, The New Elements of Mathematics, C. Eisele (ed.), La Hague, Mouton, 1976, vol. 2, p. 346.

18 Christopher Peacocke, Being Known, Oxford, Clarendon Press, 1999, p. 1-2.

19 Ibid., p. 1.

20 Christopher Peacocke, The Realm of Reason, Oxford, Clarendon Press, 2004, p. 267. On this point see also Jacques Bouveresse, Résumé des cours du Collège de France 2007-2008, p. 400.

21 Jacques Bouveresse, ibid., p. 399. [our translation]

22 Ibid.

23 Charles Renouvier, Les Dilemmes de la métaphysique pure [1901], Paris, Presses universitaires de France, 1927, p. 7.

24 Ernest Renan, Essais de morale et de critique, Paris, Michel Lévy Frères, 3rd Edition, 1861, p. viii. And he added: “Everyone makes their own faith to suit themselves”. [our translation]

25 Claudine Tiercelin, Le Doute en question. Parades pragmatistes au défi sceptique, Paris, Éditions de l’Éclat, 2005, conclusion.

26 Claudine Tiercelin, “The Fixation of Knowledge and the Question-Answer Process of Inquiry”, in F. Lihoreau (ed.), Knowledge and Questions, Grazer Philosophische Studien, no 77, 2008, p. 23-44.

27 Johann Friedrich Herbart, 1806-1808, Hauspunkte der Metaphysik, Sämtliche Werke (19 vol.), Langensalza, 1887-1912; re-ed. Aalen, Scientia Verlag, 1964, vol. II, p. 175-226;

28 Gilles-Gaston Granger, “L’argumentation du Tractatus: Systèmes philosophiques et métastructures”, in Études sur l’histoire de la philosophie en hommage à M. Gueroult, Éditions Fischbacher, Paris, 1964, p. 139-154.

29 Patrick Suppes, Probabilistic Metaphysics, Oxford, Blackwell, 1984.

30 I embarked on an analysis of this question in 2005, in Le Doute en question, op. cit.

31 Themes highlighted mainly by such authors as Peirce and Wittgenstein; see Claudine Tiercelin, Le Doute en question, op. cit., chap. III.

32 For details on this step, the reader is referred to my book Le Ciment des choses, Paris, éditions d’Ithaque, 20, Chap. 1, “Comment contourner les illusions modales et tirer le meilleur parti de nos intuitions”, p. 39-95.

33 Cf. ibid., Chap. 2, “Les pièges du scientisme”, p. 97-186.

34 Cf. ibid., Chap. 3, “Le réalisme et ses mirages”, p. 187-246.

35 John Duns Scotus, Ordinatio, I, 3, § 81. Opera omnia (Civitas Vaticana: Typis Polyglottis Vaticanis, 1950–).

36 François-Xavier Putallaz, “Introduction” to: Jean Duns Scot, Traité du premier principe, traduction de Rudi Imbach, Paris, Vrin, 2001, p. 41.

37 John Duns Scotus, A Treatise on God as First Principle, Kessinger Publishing, § 26.

38 By focusing on the quidditive being, Duns Scotus did not have in mind an “essentialisation of metaphysics”, since essence was not, in his eyes, the ultimate point of what is. Existence (as with Kant, as well) transcends, in a sense, the possible by its richness. “I say that […] logical possibility is a mode of composition formed by the intellect whose terms do not involve contradiction, and so this proposition is possible: ‘God exists’ [...] but real possibility is what is received from some real power”, Ordinatio, 1, d.2, p. 2, q. 4, n 262 (Text available from www.aristotelophile.com/Books/Translations/Scotus%20Ordinatio%20I%20dd.1-2.pdf). In other words, it is existing reality that is at the root of possibility, and not the reverse.

39 See for example: A Treatise on God as First Principle, op. cit., § 25.

40 We can thus agree that we can methodologically use the possibilia without presupposing, at least at this stage, a strong ontological thesis of some kind. Perhaps the possible worlds are of the same ontological kind as ours; perhaps, with the exception of our world, they are abstract entities; or perhaps structural universals; or certain types of collections of interpreted sentences. Or perhaps they are nothing at all. But we can make them comprehensible by approaching them in terms of combinations of properties and relations. We are then not obliged —contrary to what modal realism maintains— to subscribe to a strong ontological thesis aimed, for example, at arguing that the possible (like the impossible) are objective properties of things, independent of our capacities to conceive and imagine.

41 Tamar Szabo Gendler and John Hawthorne (eds), Conceivability and Possibility, Oxford, Clarendon Press, 2002, p. 5.

42 For a brilliant illustration of this way of proceeding, see the book by Frank Jackson, From Metaphysics to Ethics: A Defence of Conceptual Analysis, Oxford, Clarendon Press, 1998, Chapters 2 and 3, in particular.

43 Hans Reichenbach, The Rise of Scientific Philosophy, Berkeley, University of California Press, 1957, p. 305.

44 As proposed today, for example by James Ladyman, Don Ross, David Spurrett and John Collier, Every Thing Must Go: Metaphysics Naturalized, Oxford, Oxford University Press, 2007, p. 1.

45 Moritz Schlick, “Le vécu, la connaissance, la métaphysique”, in: Antonia Soulez (ed.), Manifeste du Cercle de Vienne et autres écrits, Paris, Presses Universitaires de France, 1985, p. 183-197.

46 Wilfrid Sellars, “Empiricism and the Philosophy of Mind”, in: Wilfrid Sellars, Science, Perception and Reality, Routledge & Kegan Paul, 1963, p. 173.

47 Wilfrid Sellars, “Philosophy and the Scientific Image of Man”, in: Wilfrid Sellars, Science, Perception and Reality, op. cit., p. 4 and p. 15.

48 E.J. Lowe, A Survey of Metaphysics, Oxford, Oxford University Press, 2002, p. 174 sq.

49 Ernest Renan, Essais de morale et de critique, op. cit., preface, p. iii.

Auteur

Professor at the Collège de France
Liz Libbrecht (Traducteur)

© Collège de France, 2013

Conditions d’utilisation : http://www.openedition.org/6540