Home Page         Papers by Peirce         Peirce-Related Papers


The Philosophy of the Future:
Can Pragmatism Dispense with Peircean Categories?

Ramon Vilà Vernis

At a series of lectures designed to make clear the meaning of pragmatism,<1> Rorty came more or less to define it as the philosophy of the future —literally “the apotheosis of the future.” With that he didn't mean simply that philosophy departments will be full of pragmatists in the future, but that the idea of future is embedded in the very core of the movement; pragmatism distinguishes itself among other philosophical movements by making a stand that we should take the future as our only “reason and justification.” As classically interpreted by John Dewey, considered by Rorty as the “paradigmatic” pragmatist, this idea reveals it as a typically American philosophy, and especially as a philosophy most friendly to democracy.

According to this, pragmatism was born as a reaction against what should be seen as the philosophy of the past, typically identified with European philosophy. This time again, the philosophy of the past is not called so merely on account of its historic prominence until our day, but rather because it took the past as its only “reason and justification.” Metaphysics would have always been a reactionary force, created by the privileged and directed at keeping them in that position: that's why it always tried to model everything as if it was a past thing, to send it back to the unimpeachable “that's it” that puts it out of reach to every human action. The result was a congealed world where everything was given in advance, though in metaphysical parlance that congealment turned into necessity, even into essence. The first step to know something was to renounce any aim at changing it: knowledge should be disinterested; its supreme goal, certainty. Philosophy worked as the most efficient of censorships.

Despite his taking Dewey as a model, Rorty didn't feel so comfortable with these kind of explanations: the very pragmatic aim of breaking up with essences suggests the loosening of any too close link between a philosophical thesis and a political stance (one can be a pragmatist and a reactionary, a pragmatist and an antidemocrat, etc.), or between a philosophical thesis and a national idiosyncrasy (the work done by many European philosophers overlaps with that of many American philosophers). Thus searching among Deweyan explanations of what thinking in terms of the future could mean, Rorty eventually chose one that is considerably more abstract and obscure: pragmatism would be an antidualist philosophy. We'll devote the rest of the paper to examining what can it mean to say that a philosophy is dualist or antidualist, and what could any of this have to do with time, but before that we would like to call into question the Rortyan selection of the “paradigmatic” pragmatist.

Years before Dewey could do so, Peirce had already suggested the opposition between past and future as the fundamental dilemma in Western philosophy (CP 8.7 ff.): all metaphysical oppositions could be reduced to an opposition between those who conceive reality as the original source from which all thought comes, and those who conceive it as the final destiny to which all thought leads to. Peirce sides with the latter, and with him all those who call themselves pragmatists; but if looking towards the future is the common bond among every pragmatist, there exist crucial differences among them. The future that Peirce is calling for as the “reason and justification” of philosophy takes the shape of a destiny. And here is where Rorty and many others have a problem: to think of the future in terms of a destiny doesn't seem too different to thinking it in terms of a past. If the future has something distinctive about it, it is its plural nature, and that's why the overcoming of dualism, in the singular mathematics required to the occasion, will take the shape of a pluralism; at any rate, that's the most common standing among pragmatists after Peirce, starting with William James.

However, Peirce had been first too in favouring these kind of pseudo-mathematical explanations, and took them much further than any of his heirs. So what would be his answer to their challenge? Probably to point out that pluralism, properly understood, is the purest expression of dualism, and consequently that it is not he who is thinking the future in terms of the past, but rather his critics. But it should be quite clear by now that no part of this discussion will make any sense for us unless we try to establish what philosophers mean when they speak of dualism. To that end we will start with a mental experiment suggested by Peirce long before such kind of exercises became a trend in philosophy. The experiment tries to take a closer look, precisely, at what is implied in a relation of two:
Imagine yourself making a strong muscular effort, say that of pressing with all your might against a half-open door. Obviously, there is a sense of resistance. There could not be effort without an equal effort that it resists. (CP 5.45)
A dual or dyadic relation can be any relation between two, whatever its nature. But there are relations that don't tell us much about either of its parties: two books with green covers, two scientists of the same height, two head scores at the same soccer league, etc. In this sense, such relations can be of barely any interest for a philosopher, let alone anyone with a less speculative disposition. But there are other relations of this kind that do tell us something about those that take part in it; Peirce calls these relations more “genuine.” The example he is suggesting here is one of the most genuine dyadic relations that one can imagine: each party is defined exclusively on account of his relation to the other, regardless of any other consideration. There can only be effort if there is a resistance, and only in the measure that there is such resistance; the same could be said the other way around. We could easily find more examples of this sort, but in the measure that we stuck to this definition of what a dyadic relation should be, all would take a similar air of “opposition” or “struggle,” both understood of course in the widest of senses.

In short, an interesting dyadic relation is that which tells us something about those that take part in it, and is more interesting –or more “genuine”– the more it tells about them. What should then be the meaning of dualism? Dualism should be the philosophical thesis according to which those are the relations that best help us understand the world, and that one way or another lurk behind our vision of most things. According to a quite extended idea, that would be a typically Western way of seeing things.

But let us return to our mental experiment. I'm pushing against a half open door, and I sense a resistance on the other side; I can only push inasmuch I find a resistance, and a resistance that is equal to my effort. In a setting like this, I could fall into a somewhat distressing doubt, let's say a “Cartesian” doubt: How do I know that I am the one who pushes, and the other the one who resists? Couldn't it be the other way around for whoever is on the other side? At this point it is important that we take the experiment seriously, and avoid as much as possible any consideration that goes beyond the purely dyadic relation between my effort and the resistance I find, as for example asking who is inside and who is outside, which are the intentions of the parties, etc. And if we do limit our perspective this way, the first idea that will come to mind will probably be to ask: Who started? Who got there before? Indeed, there doesn't seem to be any other way of answering the doubt of the dualist, though maybe someone more fastidious would say that such a time-related answer is still too removed from the rigorously dual staging of the experiment; after a certain amount of reflection, we might rephrase the question to say: Who was first?

Sooner or later every dualist will come to this question, regarding any issue he might be considering, and he will pose it in terms more or less analogous to ours: Which is the origin, the foundation, the cause of…? In all those cases the question points towards the past, according to a tendency already detected by Dewey in this philosophical tradition, or more strictly speaking to what we have called a first: in other words, we have to admit that one of the poles of the dyad cannot be entirely defined in terms of the other, as our previous definition would have it, but also in terms of itself. Thus if we want that our dualist be something more than a sceptic, we should extend a bit more his worldview to admit not only one but two kinds of fundamental relations, namely dyadic and monadic ones. In this revised sense, dualism seems indeed to be the door through which essences break into our world, as Dewey had also warned us.

However, our original and more stringent definition of dualism comes back with a vengeance, and prevents us from answering the very questions it seemed to suggest. That would have been, ultimately, the great revolution brought about by Kant. As Rorty has put it: “Kant produced a decisive change in the history of Western philosophy by offering a reductio ad absurdum of the attempt of distinguishing the role of the subject from the role of the object in the constitution of knowledge.” We have already seen that there is only one proper way of distinguishing roles in a dyad, and there can be no doubt that knowledge is understood in dyadic terms here. What would then be the reductio ad absurdum offered by Kant? In the simplest terms, that there's no answer to the question of who was there before (or first), so far as we stick to the dyadic perspective that lead to the question in the first place: there can be no effort before there being a resistance, and no resistance before there being an effort. We'll never know who pushes and who resists in our experiment, nor what's the role of the subject and that of the object in the constitution of knowledge. That's also what Heidegger had in mind when he spoke of the “abyss of imagination” in Kant.

However, the deep theoretical abyss to which dualism has brought us doesn't match with the easiness of our dealings with these kind of dilemmas in daily matters. If the answer doesn't come from asking who was before, then, how do we do it? According to Peirce:
If you find that the door is pushed open in spite of you, you will say that it was the person on the other side that acted and you that resisted, while if you succeed in pushing the door to, you will say that it was you who acted and the other person that resisted. (CP 5.45)
The question is not so much what went before, but what will come after: only by looking to the future, or to the result, we get clear about the matter of who pushes and who resists. That's how we give a role to each thing in our everyday experience, and not by looking to the past (or to the essence) as the dualist would have it. Of course, there's a stricter way too of putting the matter here, and that would be to say that we need a third party to determine what the first and the second would be. In other words, that would be to say that in a relevant sense our perspective on things is not dyadic, but triadic.

In Rorty's eyes, though, and in those of many pragmatists, that conclusion is entirely out of place: Why only a third and not a fourth, a fifth, a sixth, or whatever number it be? If it makes any sense at all to suggest that we look at the future rather than the past, it can only be this: that there's no given answer to the question of who's first, and that there can't be. Moreover, why should we have a special interest in the question of who's first? Why should we have an interest even in relations defined in such logical-mathematical terms? The mistake of the dualist is indeed to reduce our understanding of the world to a single privileged form, it doesn't really matter which; that's what makes him always come back to the question of the first, the essence or the foundation, and what ultimately causes his failure. In Rorty's terms, the question is to “change the world-images built with the help of those Greek oppositions [essence and accident, substance and property, appearance and reality] for an image of an ever-changing flux of relations, of relations without terms, of relations of relations.” Rorty labelled this way of seeing things “relationalism” —which does seem an improvement from the more conventional “pluralism”—, and thought that it captured the exact meaning of his proposed antidualism, and of his idea of taking the future as our only “reason and justification.”

But if Rorty's charge was that triadism offered nothing really new to dualism, Peirce's answer would have been, conversely, that it is relationalism that bears this most basic of faults. Rorty spoke of not focusing on one, nor two, but a whole open range of relations; but the point here is whether those relations are reducible or not to the notion of “being in relation to another” that defines a dyadic relation. And it is difficult to think that Rorty had in mind any other kind of relation, when all the features he produced to describe them —or what amounts to the same, to suggest that it is not possible to describe them— bear the most characteristic stamp of dyadic relations, as our previous experiment has shown them; moreover, that seems to be his idea of what a “relation” is. So rather than opening before us a whole range of relations, what Rorty really did was banish monadic relations —which he didn't consider relations at all, but some nonexistent “terms”— from his worldview, and send us back again to our first definition of what being a dualist could mean.

If we really want to escape the circle of dualism in its two aforementioned versions,<2> we'll have to show at least that there's another kind of relation that is not reducible to the dyad nor the monad. And that's exactly what Peirce wants to show us in the case of triads; the strategy to do so, of course, will be another of his mental experiments. On this occasion we are dealing with a hunting expedition where the hunter plans to shoot an eagle; to this poor end, the hunter will be able to choose whatever day he likes, put on a beige jacket or a khaki one, hide behind a rock or under a tree, etc. It doesn't seem easy to describe the intentions of the hunter without reference to this kind of triadic relations among the hunter, something else, and the prey. However, at a certain point of the expedition there seems to be a radical change: from the moment the hunter pulls the trigger, the cast of characters involved in the relations required to describe what happens is cut down dramatically to two. Thus we have a series of dyadic relations between the finger and the trigger, the trigger and the cartridge, etc.; there wouldn't be any problem to break up each of these dyads into any number of dyads, but there doesn't seem to be a need to resort to any other class of relations.

If we asked the hunter about this change in our way of describing the situation, we would find that it is also far from being indifferent to him; indeed, it lies at the very heart of all his art and all his worries. The reason for that, as Peirce goes on to explain, is that “should the eagle make a swoop in another direction, the bullet does not swerve in the least” (CP 1.212). If the eagle had swooped one moment earlier, instead, the hunter would have been able to change the position of his rifle, and the desired relation between the parties –at least by one of them– would have been saved. We discover thus a crucial difference between dyadic and triadic relations, which renders them irreducible both at a theoretical and a practical level. A triadic relation is that which leaves one of its parties (the third, or the means) partly undetermined, so that the other two can remain determined in the relevant sense, should a change occur in either of them; in this sense, we only understand it properly when we talk in terms of the future. A dyadic relation, on the other hand, leaves nothing at all to be determined; we could say that we only make justice to it when we talk in terms of the past.

So it seems we have found two different kinds of relations in the hunting expedition, both indispensable to describe one or another part of it. However, the mention of two temporal points of view on the matter should make us suspect that we may have been too fast to conclude that one part of the expedition requires a dyadic account, and another a triadic account. The dualist of course will have no problem to admit that the entire hunting expedition can be described in dyadic terms; what he won't see is the other possibility, that is, the possibility of describing the whole picture in triadic terms too. Even if he admitted that some parts of the story (in particular, the actions of the hunter) are much more cogently described in such terms, he will insist in seeing that as an oddity, and will seek some kind of explanation ad hoc. But that's exactly what Peirce would never admit: if he recognizes only one triadic relation in the whole picture, then he'll have to agree to describe it all in the same terms.

Let us turn back to the experiment again: if the hunter changes the position of his rifle depending on the movements of the eagle, it is because he has an expectation of what will happen when he pulls the trigger. And what is that expectation? Of course, that the bullet will change its trajectory depending on the position of the rifle. In other words, the hunter can only aim at his prey so far as the bullet also “aims” at an end of its own, definitely not the eagle but the trajectory that every bullet would follow under the same circumstances. The same analogy –debasing for some, mystifying for others– with the actions of the hunter will have to be hidden in everything that is around him, or else it makes no sense that he runs after his prey, that he tries to aim at it, or generally that anyone tells a tale about a hunter and his prey.

If we consider the issue from this perspective, however, we'll see that most relations we had been regarding as dyadic until now are in fact triadic relations, perhaps less “genuinely” triadic than those where the hunter is directly involved, but triadic nonetheless; moreover, we'll see that in a strict sense we can't describe anything in dyadic terms, we can only experience it.<3> We have already said that it is only possible to distinguish a first from a second in a dyad, and thus understand it in some measure, if we bring some kind of third to the setting; now we see this is only a restatement of the old wisdom from Aristotle, namely, that we can't even conceive of the purely determined. And that's precisely the wisdom that gets lost, incidentally, among those who rebuff Peirce for talking of future as a destiny. According to what we have seen, there will always have to exist a determined side to the future, or an end, for there to be an undetermined side to it, or a means. But this end will only be intelligible insofar it is not only an end, but also a means, or otherwise there would be nothing at all to say about it. It would be pure secondness, pure determination —death, if we prefer to use anthropological notions. In this sense, Peirce talked of something quite similar to the “being unto death” of Heidegger, though he often referred to it in terms that have caused a certain confusion, such as the “final opinion.”

However, it is likely that we'll never get over our confusion, nor really understand how something can be a second in some sense, and a third in some other sense, if we don't understand what it is in itself, that is, if we don't eventually answer the question of what is a first. The mental experiment proposed by Peirce in this case will sound much more familiar to a philosopher used to present-day discussions in the field: “If a man possessed no other colour-sensation but that excited by this sealing wax... he might devote his life to thinking about it, but he never would discover that there were those three respects: luminousness, chroma, and hue. They are not seen in the colour taken by itself but only in the colour as it appears in comparison with others.”<4> The colour red taken by itself refers to nothing but itself, as opposed to what happened in previous experiments; moreover, that seems to be the most “genuine” way of conceiving it, so far as any attempt at defining it by reference to something else seems to lose sight of what's most peculiar about red. There's a famous version of this same experiment by a later philosopher,<5> which appears an almost complete inversion of it: the subject of this experiment, Mary, could give us a perfect description of red at every conceivable level, but since she's been locked all her life in a black-and-white room, she knows nothing of what most of us would call the very “redness” of red. The experiment points out again at the fact that red cannot be properly captured by any description different from itself, but as we shall see, there are quite different aims behind each version of the experiment.

To most contemporary philosophers, Mary's experiment shows –or purports to show– that there is a “subjective” side to experience that no amount of “objective” data could capture in a proper sense; ultimately, this would be a proof that there is something like a subjective experience in the traditional sense. But the point of the experiment for Peirce couldn't be more removed from an attempt of discerning an inside and an outside to our experience of things; indeed, that's just the kind of dyadic perspective that he is trying to dispel with the experiment. That's why his version carefully erases from the mind of the subject not only the experience of red, but any other experience that could bear a relation to it, as those of black and white; and that minor difference is not without consequence, for if the problem with Mary was that nobody could explain to her what red is properly like, the subject of the Peircean version could not even see it if they showed it to him! For how could anyone see a colour, if it is not by contrast with another colour? Peirce's answer to this objection confirms even more that the difference between what's inside and what's outside the mind is of no consequence here:
If everything in the world… were precisely of this sealing-wax red, though we should not be distinctly aware of it, I suppose that it would tinge our disposition, and so be, in some sense, in the mind. If it would not, this would be merely a psychological fact: it would have nothing to do with the quality red in itself.<6>
The attempt at imagining how the quality red could be in itself, apart from any intimation or reference to something else, be it inside or outside the mind, has brought us past not only any conceivable description, but also any conceivable experience. We're left only with redness as a pure possibility… and that's exactly what the result of the experiment should be. To find out what something is in itself we need only retreat from what it is in relation to any other, though it is difficult to find a more elusive path, since everything is presented to us as a second. Maybe that is why we always tend to mistake the essence of a thing for something given or determined in some measure, or in Peircean terms, what makes us mistake a first for a second —or still in Heideggerian terms, what makes us forget the “ontological difference.” According to the temporal hint we have been following all along, what something is in itself can only be understood in the most rigorous present, that is, as a pure possibility. Ultimately, the point is to come back to the core of the Platonic doctrine of ideas, or to what this very doctrine already started to mistake.

In fact, the insistence of this last experiment in ignoring the difference between what's inside and what's outside the mind goes only one step further in a direction that was already implicit in our previous experiments —and we can see this also because Peirce never called them “mental” experiments. Rather, he called them “phenomenological” experiments, a kind of investigation intended to establish what things are truly “in themselves,” and that purports to get to them by ignoring any idea of an inside or an outside, or more generally any previous assumption about the means or the results of an investigation. Of course, an investigation of this sort is properly unfeasible to any investigator of more than some months of age, but its formal definition –and its practical approximation– is the last and most rigorous version of that “empirical” reformulation of Kantism that Peirce envisaged from his first philosophical paper.<7> The point is to get over the dyadic setting in which Kant continued to think, precisely to give a proper development to his basic intuition, which we may now sum up as the impossibility of finding something that is in itself and not in relation to another.

In the field of philosophical methodology, the point is to change the transcendental method for an experimental method. Peirce will continue to call “categories” the results of his investigation, and will continue to define them as the most general aspects of experience, as should be the results of any proper philosophical investigation. But the sense of this formal definition cannot be more transformed: they are no more the previous conditions from which every knowledge stems, but the final conclusions to which every knowledge leads. Similarly, his categories cannot be proved by any particular reasoning or “deduction” –and that has also misled some commentators of Peirce<8>– but by a process that can only be defined as unlimited in every conceivable sense, or what Peirce formally described as phenomenology. In this sense, the criticism that this discipline generally bears on account of its absolute lack of guarantees reveals a misunderstanding not only of its role in Peirce's own philosophy, but more generally of the intrinsic nature of the experimental method, whose proper essence –though persistently distorted both in theory and in practice– is to completely renounce the idea of a guarantee of knowledge. That's why, incidentally, we find such problems in providing a justification for a method that elicits so little real doubt.

We thus see that the way to bring forward the program suggested both by Dewey and Rorty, the overcoming of dualism, consists as they also suggested of thinking in terms of the future (or more generally, in terms of time, as was suggested too by Heidegger). But it is the very greatness of the overcoming that they envisage what explains the meagreness of the instruments they have to bring it about: in fact, they have nothing to answer to the questions posed by the dualist –generally, questions about the foundation of what they say–, since those very questions have lost to them a good deal of their sense. And it's the confusion caused by these questions, and the urgency to answer them, that explains a tendency among pragmatists to lose sight of the meaning of what they themselves say. That is why Rorty, for example, did never recognize his own “relationalism” in Peirce's categories, and insisted in pondering why those three categories and not more, or why not some other, when those very questions are just what he is trying to disband the rest of the time. In fact, Peirce did respond that those categories are not intended to block our recognizing any other, indeed, that they are intended to work just the other way around; but in a sense, to say that is already beside the point.

In those lectures where Rorty tried to deliver some key insights on pragmatism, he pointed out that the best way to see things as a pragmatist should is to treat them as if they were numbers; the reason for that is that a number is nothing else than the relations that constitute it, which leads us more than halfway intro freeing things from their “essentialist” congealment. But then maybe we would like to ask, if using Peirce's categories is the best way of understanding pragmatism, why would Rorty want to drop them? The answer seems to be here that as important as it was for Rorty to talk of things as if they were numbers, it was also to insist that that was just “another way” of talking of things, which tangled him endlessly in the same problems and the same discussions he so wanted to get past.


<1> Published as Hoffnung Statt Erkentniss: Eine Einführung in die Pragmatische Philosophie (Vienna: Passagen Verlag, 1994), the original text in English was later included in Philosophy and Social Hope (London: Penguin, 2000). Every quotation of Rorty is from these lectures.
<2> Peirce commented many times on aspects of Western philosophical tradition from the point of view of the categories, in ways that may seem to conflict sometimes with the views suggested here —as they seem to conflict sometimes among themselves. The general point that needs to be stressed is that the three categories only make sense when taken as a whole, and thus that any attempt at describing a philosophical position as recognizing only some of them has to lead to inconsistencies, and could be reformulated in other –also self-defeating– ways; the only sense of taking that perspective on tradition, of course, is to show it wrong, as I am trying to do here. See the third and fourth Harvard lectures (especially CP 5.77 ff.) for a much more nuanced account of the evolution of Western thought, lead by the attempt to show it as a systematic trial of the three categories on the maxim of parsimony. It could be noted, however, that Peirce focuses in those lectures on the stated intentions of each philosopher, not on how their theories effectively work, so the account is consistent with his insistence elsewhere that Western tradition is more generally under the spell of nominalism, or of a recognition only of secondness and firstness —a position that bears in those lectures only a small part of the story.
<3> In stricter terms not even that: things only exist in dyadic terms, and only in the sense that existence cannot be grasped at all, or is properly speaking “unreal.”
<4> The Essential Peirce, vol. II, p. 366.
<5> This experiment was proposed by Frank Jackson in “Epiphenomenal Qualia” (1982), Philosophical Quarterly, vol. 32, pp. 127-36.
<6> The Essential Peirce, vol. II, p. 367.
<7> On a New List of Categories (1867), CP 1.545 ff; all that Peirce had in print before was in formal logic. For some general indications on the evolution of Peirce's thought, see my paper Peircean Categories: an Old Name for a New Way of Thinking at the Digital Encyclopedia of C.S.P. (http://www.digitalpeirce.fee.unicamp.br/home.htm)
<8> See for example the discussion of Peirce's “remarkable theorem” at Christopher Hookway, Peirce (London: Routledge & Kegan Paul, 1985) p. 97 ff.


Do you think the author has it wrong? If so and you want to contribute a critical comment or commentary, brief or extended, concerning the above paper or its subject-matter, or concerning previous commentary on it, it will be incorporated into this webpage as perspicuously as possible and itself become subject thereby to further critical response, thus contributing to Arisbe as a matrix for dialogue. Your contribution could also be of the nature of a corroboration of the author, of course, or be related to it or to some other response to it in some other relevant way.

The URL of this page is:

This paper was last modified March 4, 2008

Queries, comments, and suggestions regarding the website to: