2. The Return of Nature

Since I have always censured and warned against the mistake of straying beyond the limits of the science at hand or mixing one science with another, this is the last fault I could be reproached with.

—Immanuel Kant, The Conflict of the Faculties

As soon as we allow two different callings to combine and run together, we can form no clear notion of the characteristic that distinguishes each by itself.

—Immanuel Kant, The Conflict of the Faculties

Let us suppose that Kant has succeeded in thinking through the crossing of the frontier between state of nature and state of law, in spite of all the aporias we have pointed out in passing. Let us even suppose that he has succeeded in thinking through everything to do with the internal organization of the State and that the problems that occupy the greater part of treatises on political philosophy have been resolved. Let us jump over everything that concerns sovereignty, legislation, the forms of government, suffrage, private property, everything that is the bread and butter of political philosophy. Once these problems have been solved (but we shall see that they can never be solved, and that is politics), the question arises of the relation between the State in question (which up to now will have been called simply the State, with a curious generic singular) and the other States we suddenly realize surround it. The Greek polis, which remains the fundamental model for the State in political philosophy, is essentially circular, the circle of citizens around the empty (“de-real”) space in the middle.1 It is as though the members (the men) of the city, finding themselves at the end of their immediate (i.e., internal) political preoccupations, turned away from the empty space that defines the circle in which they are standing and saw, to their great surprise, that the space around them is full of other groups, other circles. In fact, the existence of several States (and we shall see that there must be several) should prevent us from thinking—with the whole tradition—the analogical form of the State as being that of a circle: Circles can touch each other, but only tangentially, and cannot have frontiers in common. We shall need a more complex geometry to think politics.

The problem, simply put, is that the state of nature, left behind by the foundation (or at least the description) of the polis, keeps coming back as soon as that polis has to enter into contact with other poleis. One of the passages we quoted from the Metaphysics of Morals (§61), in fact, concerns not individuals in a state of nature but States in their relations among themselves. Here is the passage again, this time without omissions:

Since a state of nature among nations, like a state of nature among individual men, is a condition that one ought to leave in order to enter a lawful condition, before this happens any rights of nations, and anything external that is mine or yours that states can acquire or retain by war, are merely provisional. Only in a universal association of states (analogous to that by which a people becomes a state) can rights come to hold conclusively and a true condition of peace come about. (MM, §61, 119)

We thought we had crossed the (analogical) frontier separating state of nature from state of law, but this frontier returns as soon as we pose the question of the (real) frontier of the State. The frontier that surrounds the polis and marks it off from nature (even if that separation is inspired by nature, even if it is in a sense natural) also immediately marks the return of nature or the return to nature in the relations between States, as soon as we take its outer limit into account.

This situation brings formidable difficulties with it, and if I have chosen to follow them in Kant, this is because he seems to confront these difficulties more directly than others. Let me say immediately that it seems to me that this problem of the frontier is intractable and that it has been the object of a foreclosure in traditional political philosophy.2

If it is true that political philosophy constructs its polis out of nature, only to find that nature returns at its frontier, then it is reasonable to think that that nature, the one that returns, is in fact the primary problem. This quasi-Aristotelian reversal (for Aristotle too, what came last in the natural process was, in fact, first) implies, with Aristotle, that what is really first in nature, the natural origin, is not what appears first (individuals incapable of sustaining themselves in the state of nature without forming associations). But this reversal also implies, against or beyond Aristotle, that nor is it what appears at the end of the natural process that is, in fact, the origin (the polis and its as-if autarkeia as end and thus nature of the whole process). What seems to come first here is not the last stage, but the stage after the last, the postultimate stage, which, by an ultra-Aristotelian reversal, will show itself to be the preprimary stage, the origin of the supposedly natural origin. Aristotle says that the polis, which comes last in the story he tells, is, in fact, first: I am suggesting that what comes first can only be the state of nature that holds between States, after their supposed autarkeia has opened to the outside. Before the polis, that comes before the individual, there are relations (and therefore frontiers) between States (that do not yet exist). This possibility, allowed for in the “as if” or “all but” of the supposed autonomy of “the” State, is paradoxical enough to suspend terms such as “first” and “last” and brings with it a temporal complication that we are only beginning to glimpse.

Kant explains this very clearly in the “Universal History” text, still in terms of first and last. Recall the fifth proposition of that text, which claims:

the highest task which nature has set for mankind must therefore be that of establishing a society in which freedom under external laws would be combined to the greatest possible extent with irresistible force, in other words of establishing a perfectly just civil constitution. (KPW, 45–46)

And the sixth proposition opens on the claim that this highest problem is the last problem: “This problem is both the most difficult and the last to be solved by the human race.” Let’s read the rest of this famous proposition:

The difficulty (which the very idea of this problem clearly presents) is this: if he lives among others of his own species, man is an animal who needs a master. For he certainly abuses his freedom in relation to others of his own kind. And even although, as a rational creature, he desires a law to impose limits on the freedom of all, he is still misled by his self-seeking animal inclinations into exempting himself from the law where he can. He thus requires a master to break his self-will and force him to obey a universally valid will under which everyone can be free.3 But where is he to find such a master? Nowhere else but in the human species. But this master will also be an animal who needs a master. Thus while man may try as he will, it is hard to see how he can obtain for public justice a supreme authority which would itself be just, whether he seeks this authority in a single person or in a group of many persons selected for this purpose. For each one of them will always misuse his freedom if he does not have anyone above him to apply force to him as the laws should require it. Yet the highest authority has to be just in itself and yet also a man. This is therefore the most difficult of all tasks, and a perfect solution is impossible. Nothing straight can be constructed from such warped wood as that which man is made of. Nature only requires of us that we should approximate to this idea. A further reason why this task must be the last to be accomplished [am spätesten] is that man needs for it a correct conception of the nature of a possible constitution, great experience tested in many affairs of the world, and above all else a good will prepared to accept the findings of this experience. But three factors such as these will not easily be found in conjunction, and if they are, it will happen only at a late stage [nur sehr spät] and after many unsuccessful attempts. (KPW, 46–47)

Man’s “self-seeking animal inclinations [seine selbstsüchtige tierische Neigung]” bespeak an egoism that is more complex than it might appear (Luc Ferry’s French translation has, plausibly enough, “égoïstes” for selbstsüchtige here). In the Anthropology, Egoism appears right at the beginning (§2), under its learned German name Egoism, as a consequence of the very thing that supposedly distinguishes man from natural things (including animals, which means that a self-seeking inclination is already not quite animal)—that is, self-consciousness as an I. Given how paradoxical all these relations are in Kant, beyond the paradox they explicitly discuss, it is worth quoting at length:

While in philosophy we may not call up the judgments of others to confirm our own, as jurists do in calling up the judgments of those versed in the law, nevertheless each writer who finds no followers with his publicly avowed opinion on an important topic is suspected of being in error.

For this very reason it is a hazardous enterprise, even for intelligent people, to entertain an assertion that contradicts generally accepted opinion. This semblance of egoism is called paradox. It is not boldness to run the risk that what one says might be untrue, but rather than only a few people might accept it.—The predilection for paradox is in fact logical obstinacy, in which someone does not want to be an imitator of others, but to appear as a rare human being. Instead, a person like this often appears only strange. But because every person must have and assert his own thoughts (Si omnes patres sic, at ego non sic. Abelard), the reproach of paradox, when it is not based on vanity, or simply wanting to be different, carries no bad connotations.—The opposite of paradox is banality, which has common opinion on its side. But with this there is just as little guarantee, if not less, because it lulls one to sleep; whereas paradox arouses the mind to attention and investigation, which often leads to discoveries. (Anthropology, 17–18)

This is again the paradox of the legislator (we might say simply the paradox of paradox), but how could we avoid thinking of Kant’s own experience after the publication of the first Critique? We would need to read here the reflections (which are themselves highly paradoxical) in the Prolegomena to Any Future Metaphysics That Will Be Able to Come Forward as Science, and especially in the preface to that text, which presents the task of the Prolegomena, which could only come after the Critique, as being that of persuading the reader that critique “is a completely new science, of which no-one had previously formed merely the thought, of which even the bare idea was unknown, and for which nothing from all that has been provided before now could be used except the hint that Hume’s doubts had been able to give.”4 Kant continues imperturbably to formulate the principle of unintelligibility of his work:

To approach a new science—one that is entirely isolated and is the only one of its kind—with the prejudice that it can be judged by means of one’s putative cognitions already otherwise obtained, even though it is precisely the reality of those that must first be completely called into question, results only in believing that one sees everywhere something that was already otherwise known, because the expressions perhaps sound similar; except that everything must seem to be extremely deformed, contradictory, and nonsensical, because one does not thereby make the author’s thoughts fundamental, but always simply one’s own, made natural through long habit. (Prolegomena, 12)

We might try to formulate the paradox of the paradox of egoism as meaning that it is only in thus exposing oneself to noncomprehension and ridicule that anything nonegoistic (what has recently often been called community) becomes possible. And then generalize further and say that there is communication only in the paradox of this unintelligibility and claim that this is the principle of refutation of all philosophy of a Habermasian type, whether or not it invokes Kant.

Back in the sixth proposition, “This problem is both the most difficult and the last to be solved by the human race.” The last problem, the most difficult, and thus, in Aristotelian vein, the first problem, the primary problem, the problem of problems, the supreme problem, the highest, das höchstes. If this problem were not there to finalize them, the other problems would not even be real problems. So we might be surprised by the beginning of the seventh proposition, which subordinates this supreme problem to another problem, which (at least in the order of the text) follows the last problem, like a postultimate problem that, however, has apparently to be solved before the last problem that we have just quoted:

The problem of establishing a perfect civil constitution is subordinate to [ist . . . abhängig] the problem of a law-governed external relationship with other states, and cannot be solved unless the latter is also solved. (KPW, 47)

Obviously, in spite of the order of exposition, this problem must precede the last problem, that of internal political organization. But we can see that this order of priority cannot be simple and that in order to preserve the last problem as last problem, Kant opens the perspective of an unthinkable priority of the relations between States over the States themselves. To prepare for a reading of this unthinkable, let’s try to explain as simply as possible this problem of the relations between States, whether it be the first or the last, the penultimate or the postultimate, always secondary or preoriginary.

Nature returns, then, at the frontiers of States and, still wanting discord, as Kant has said, spreads antagonism and warfare in a state of anarchy that seems formally identical to the one suffered by individuals in the state of nature. To the extent that this situation ought to push the individual toward civil organization, it would seem that the return of Nature between States ought to push them toward a supranational or international organization that would be the counterpart, on the planetary scale, of the constitution of the individual State, and a super-State of this sort (which would then realize the fantasy not only of every head of state5 but of political philosophy itself, which talks of The State in the singular, which would like there only to be one) would seem to resolve the problem of frontiers (there would simply be none) and merge again with a certain naturality of the planet, which, it will be remembered, does not present any natural frontiers that cannot be crossed, if only in a ship or on a camel.

The end of nature, according to Kant, is that mankind achieve through discord the full development of all natural faculties. War, which is the form par excellence of this discord, has as its result the dispersion of mankind across the whole surface of the earth,6 so that they might inhabit even those zones where only natural providence (in the form of moss, reindeer, driftwood, etc.) provides them with the means to live. This movement of violent dispersion is also provoked by the invocation of the prelegal right we discussed, according to which my neighbor must or ought to leave my vicinity if he refuses to enter into a legal association with me: To leave my vicinity, he must be able to go somewhere. One might wonder how this tendency to dispersion is going to lead to the end of nature (i.e., perpetual peace). Why would war not simply continue forever? Or else, why would mankind not continue to disperse indefinitely to avoid war (all the while remaining in the state of nature, for even if they found a provisional peace by fleeing war, war would always threaten so long as that peace was not perpetual, according to the asymmetry of the war-peace couple we pointed out earlier)?

The somewhat surprising reply is to be found in the globe itself, the globe whose form has a quasi-transcendental status in this argument: Dispersion reaches its limit when those fleeing war westward, for example, come back round from the east. As the surface of the globe is finite and continuous, it appears that dispersion has, after all, its limits: “Nature has enclosed them all together within determinate limits (by the spherical shape of the place they live in, a globus terraqueus)” (MM, §62, 121).

This form of the globe gives Kant a great deal of food for thought throughout his work, and always at moments that touch (if only tangentially) on the problems that interest us here. Already in the first Critique, especially in the chapter on “The Final Aim of the Natural Dialectic of Human Reason,” which closes the appendix to the “Transcendental Dialectic,” and which vaunts the advantages for theoretical research of admitting the existence of a supreme wisdom, we find the following:

The advantage created by the earth’s spherical shape is well known; but few know that its flattening as a spheroid is the only thing preventing the elevations on the dry land, or even smaller mountains perhaps thrown up by earthquakes, from continuously displacing the earth’s axis and perhaps appreciably so in not too long a time; this might happen if the swelling out of the earth at the equator were not such a mighty mountain that the centrifugal force of every other mountain can never noticeably bring it out of place in regard to its axis. And yet without scruples we explain this wise arrangement from the equilibrium of the formerly fluid mass of the earth. (A687/B715n)

Or else, in more complex fashion, but with obvious analogical importance for everything that concerns us here (including the possibility of analogy itself), the following, from the “Discipline of Pure Reason,” not by chance under the title “On the Impossibility of a Skeptical Satisfaction [Befriedigung, translated by Tremesaygues and Pacaut7 as “trouver la paix,” finding peace] of Pure Reason That Is Divided against Itself” (we shall come later to the question of perpetual peace in philosophy):

If I represent the surface of the earth (in accordance with sensible appearance) as a plate, I cannot know how far it extends. But experience teaches me this: that wherever I go, I see a space around me in which I could proceed farther; thus I cognize the limits [Schranken] of my actual knowledge of the earth [Erdkunde, geography] at any time, but not the boundaries [Grenzen] of all possible description of the earth [Erdbeschreibung]. But if I have gotten as far as knowing that the earth is a sphere and its surface the surface of a sphere, then from a small part of the latter, e.g., from the magnitude of one degree, I can cognize its diameter and, by means of this, the complete boundary [Begrenzung], i.e., surface of the earth, determinately and in accordance with a priori principles; and although I am ignorant in regard to the objects that this surface might contain, I am not ignorant in regard to the magnitude and limits [Schranken] of the domain that contains them.

The sum total of all possible objects for our cognition seems to us to be a flat surface, which has its apparent horizon, comprehends its entire domain and which is called us the concept of unconditioned totality. It is impossible to attain this empirically, and all attempts to determine it a priori in accordance with a certain principle have been in vain. Yet all questions of our pure reason pertain to that which might lie outside this horizon or in any case at least on its borderline.

[. . .]

Our reason is not like an indeterminably extended plane, the limits of which one can cognize only in general, but must rather be compared with a sphere, the radius of which can be found out from the curvature of an arc on its surface (from the nature of synthetic a priori propositions), from which its content and its boundary [Begrenzung] can also be ascertained with certainty. (A760–62/B788–90)8

Now we might wonder whether transcendental-type arguments can find a solid footing on this globe.9 What if nature, still in dispersive mode, had as its end to send mankind to inhabit other planets? What transcendental difference could there be between crossing the desert on a camel, crossing the ocean in a boat, and crossing interplanetary and even intergalactic space in a spaceship (one imagines Kant: “the camel of space”)? More than once, Kant envisages the possibility that there be other inhabited planets in the universe (it was a fashionable subject in the eighteenth century, following Fontenelle’s earlier book on the plurality of worlds10): “That there could be inhabitants of the moon, even though no human being has ever perceived them, must of course be admitted.” Or again:

Since, however, even though we might not be able to undertake anything in relation to an object, and taking something to be true is therefore merely theoretical, in many cases we can still conceive and imagine an undertaking for which we would suppose ourselves to have sufficient grounds if there were a means for arriving at certainty about the matter; thus there is in merely theoretical judgments an analogue of practical judgments, where taking them to be true is aptly described by the word faith, and which we can call doctrinal faith. If it were possible to settle by any sort of experience whether there are inhabitants of at least some of the planets that we see, I might well bet all I have on it. Hence I say that it is not merely an opinion but a faith (on the correctness of which I would wager many advantages in life) that there are also inhabitants of other worlds. (A825/B853; tr. mod.)

But if he does indeed often speak of other planets and the possibility that they be inhabited,11 it seems that Kant never envisages the possibility that mankind might leave the spherical surface of Earth to flee the state of nature (while remaining indefinitely in it) toward space, “the final frontier.” For example, still in the sixth proposition of the “Universal History” text, Kant writes in a note:

Man’s role is thus a highly artificial one. We do not know how it is with the inhabitants of other planets and with their nature, but if we ourselves execute this commission of nature well, we may surely flatter ourselves that we occupy no mean status among our neighbors in the cosmos. Perhaps their position is such that each individual can fulfill his destiny completely within his own lifetime. With us it is otherwise; only the species as a whole can hope for this. (KPW, 47n)

Let us, then, suppose that we try to broaden the limits of the space available to mankind to take in more than the spherical surface of the earth. A transcendental argument ought to be able to accept such a broadening. If this line of argument is to succeed, however, one would have to suppose that, like the earth, the universe as a whole offers a finite space (however vast) to mankind. There is no inevitability in humans’ having to form civil associations under the rule of law if they can continue to flee forever without ever returning to their point of departure. Now, such a supposition cannot fail to remind us of the Antinomies of Pure Reason, and more especially the first antinomy that precisely concerns this question of the finitude or infinitude of the universe.

Let me recall as simply as possible the place and principle of the antinomy because, as we shall see, everything here depends on an “analogical” language of nature, frontiers, and violence. Kant takes his time leaving behind the terrain of the understanding in the first Critique, for reasons that inspire in him an unusual burst of poetic writing that is worth quoting at length in the context of natural frontiers we are slowly setting up. I will simply emphasize certain expressions:

We have now not only traveled through the land of pure understanding, and carefully inspected each part of it, but we have also surveyed it, and determined the place for each thing in it. This land, however, is an island, and enclosed in unalterable boundaries by nature itself. [Dieses Land aber ist eine Insel, und durch die Natur selbst in unverändliche Grenzen eingeschlossen.]12 It is the land of truth (a charming [reizender, attractive, seductive] name), surrounded by a broad and stormy ocean [rather than the eminently traversable and driftwood-carrying sea of the political writings], the true seat of illusion, where many a fog bank and rapidly melting iceberg pretend to be new lands and, ceaselessly deceiving with empty hopes the voyager looking around for new discoveries [den auf Entdeckungen herumschwärmenden Seefahrer], entwine him in adventures from which he can never escape and yet also never bring to an end. But before we venture out on this sea, to search through all its breadth and become certain of whether there is anything to hope for in it, it will be useful first to cast yet another glance at the map of the land that we would now leave, and to ask, first, whether we could not be satisfied with what it contains, or even must be satisfied with it out of necessity, if there is no other ground on which we could build [we do not yet know]; and, second, by what title we occupy even this land, and can hold it securely against all hostile claims [so it appears we have enemies]. (A235–36/B294–95)13

So we are dealing with an island the frontier of which extends indefinitely. The sea around the understanding, beyond the Grenze in principle determined from the inside (to talk like the Wittgenstein of the Tractatus [4.114; 5.6; 5.61–5.62]), the place of illusion, will be a thick frontier that is never crossed. What will occupy the Transcendental Dialectic (remember that for Kant dialectic is always a “logic of illusion” [CPR, A293/B349]) will be this frontier as such, insofar as one cannot avoid trying to understand it as a frontier giving access to something beyond it. It is not by chance that the language of frontiers and territories returns in force here, in the discussion of the transcendental illusion:

We have to do only with transcendental illusion, which influences principles whose use is not ever meant for experience, since in that case we would at least have a touchstone for their correctness, but which instead, contrary to all the warnings of criticism, carries us away beyond the empirical use of categories, and holds out to us the semblance of extending the pure understanding. We will call the principles whose application stays wholly and completely within the limits [Schranken] of possible experience immanent, those that would fly beyond these boundaries [Grenzen] transcendent principles.14 But by the latter I do not understand the transcendental use or misuse of categories, which is a mere mistake of the faculty of judgment when it is not properly checked by criticism, and thus does not attend enough to the boundaries of the territory in which alone the pure understanding is allowed its play [die Grenze des Bodens, worauf allein dem reinen Verstande sein Spiel erlaubt ist]; rather, I mean principles that actually incite us to tear down all those boundary posts [Grenzpfähle] and to lay claim to a wholly new territory that recognizes no demarcations anywhere. (CPR, A295–96/B352)

So, no more boundaries, either internal or external: We are on the high seas. No more nature here. The reassuring coastline that surrounds the island of the understanding (seen from the inside, from the understanding, it is an island) and separates it from a different element becomes here, seen from the vantage of Reason, already part of the island, a terrain surrounded by fences (and therefore by an eminently artificial, threatened, deconstructible frontier), with what lies beyond it forbidden but, in fact, relatively accessible, and the difference of which from the terrain situated within the fences is no longer naturally clear.15 What is more, there will now be not only a desire or temptation to go and look at what is on the other side (like the Seefahrer who could not resist the call of the high seas16) but, in fact, an incitation to escape (or invasion), an incitation that can even be of the order of order itself:

Hence transcendental and transcendent are not the same. The principles of pure understanding we presented above should be only of empirical and not of transcendental use, i.e., of a use that reaches out beyond the boundaries of experience [Erfahrungsgrenze]. But a principle that takes away these limits [Schranken], which indeed bids us [gebietet] to overstep them, is called transcendent. If our critique can succeed in discovering the illusion in these supposed principles, then those principles that are of merely empirical use can be called, in opposition to them, immanent principles of pure understanding. (CPR, A296/B352–53)

So the principles in question are transcendental, but their use must be no more than immanent. The transcendent use of the transcendental is to be avoided. The principles of pure understanding as such are not to be found in experience, but they must be applied only to experience, from which they departed only the better to return. The ineluctable character of the transcendental illusion comes from the fact that the transcendental cannot fail to tend to become transcendent, that reason cannot prevent itself not only from crossing the frontier but from enjoining us to cross it. We have only just established the limit of the use of the concepts of the understanding (against a certain barbarity),17 and now here we are wanting to pull down the fence we just put up and charge off into the beyond.

The Antinomy of Pure Reason has a privileged place in the exploration of this frontier, at least for us, because it concerns “the realm of appearances,” where there is a world (and a cosmology) and where planets and stars call to us. The Antinomy thus occupies a troubling place with respect to the island (or the planet) of the understanding (of the truth as we know it), because it would seem that only it could claim to establish the frontier of that island or planet, if there is a frontier. And indeed we shall see that the Antinomy is all about the frontier and that all the themes we have picked out of the political writings, where frontier seems to have a literal meaning, return, in a way that seems to be analogical, to haunt Kant’s text here.18

The apparently central position of the Antinomy in the Transcendental Dialectic (between the paralogisms and the ideal, or between the subject and God) is itself a frontier position. The place (and we cannot use this term with impunity, nor indeed terms such as sphere, terrain, realm, territory, or even zone, because all these terms—and the term term itself—are all more or less mortgaged to the problems we are trying to elucidate) of the Antinomy is that of the world (places, spaces, space, what is spaced in general) as realm of phenomena (Erscheinungen, appearances [but not yet illusions]) that must present themselves as being conditioned and that therefore lead one to think that there must be a totality of conditions for a given phenomenon, a totality that, because it is total, must itself be unconditioned. Reason cannot prevent itself from wanting to grasp this totality (which obviously is never given in experience, because everything that is given in experience is necessarily conditioned and therefore limited) and, in so doing, falls into the Antinomy. According to Kant, there are four inevitable antinomies of reason: of the finitude or infinitude of the world, of the simplicity or nonsimplicity of substance, of causality, and of the existence of a necessary being. We are going to look at the first of these, which touches very directly on the problem posed by the status of Kant’s argument in the political writings about the sphericity (or at least the finitude) of the terrestrial globe.

The first antinomy, then, concerns the finitude or infinitude, in space and time, of the world (the idea of the totality of all appearances). The thesis of the first antinomy: The world has a beginning in time, and it is finite in space; for if the world did not have a beginning, that would imply that an infinity of time had elapsed before any given moment—but as an infinite series (here that of the successive states of things) is infinite only on condition of never being completed, it follows that this is impossible and that the world must have had a beginning in time. And the world must be finite in space, because if it were not, given that it would not be given to intuition within limits, its size would have to be thought by successive enumeration of its parts, which would require that an infinite time must have elapsed for the world to be thought of as infinite, which is impossible. The antithesis: If the world had a beginning in time, there would have had to be an empty time before this beginning; but as nothing can come from an empty time, the world cannot have a beginning in time. And as for space: If the world were finite, it would have to exist in a limitless empty space; this space could not be part of the world (totality of objects), which would then be in relation with a nonobject, which is impossible.

Even if the antinomies in general do not exactly lead to war, they are certainly not at peace either. Before going into the detail, Kant presents them as jousts, which are not very dangerous but during which one realizes sooner or later that victory always goes to the one who goes last.19 He then insists, after repeating them, on the fact that they are not for all that a game (Spielgefechte) and also that reason cannot simply request or demand peace, because it is implicated in the conflict itself. And yet, we can hope that the conflict rests on a mere misunderstanding, ein blosser Missverstand, which would, once dissipated, allow a durable peace to be instituted, a durably calm regime (ein dauerhaft ruhiges Regiment), in which reason would dominate the understanding and the senses (CPR, A464–65/B492–93).

We were seeking the consequences that would result from extending to the cosmos in general the space into which, fleeing the intolerable state of nature, mankind can disperse to avoid violent confrontation. Just as, on the global scale, Kant needs the finitude of the earth’s surface to support his argument that perpetual peace is inevitable, the result of violence itself, here we find that the interest of reason (which would guide me if I absolutely had to take sides in this conflict) is better served, for practical purposes, by the thesis (according to which the world as totality of appearances is finite). It is also on the side of the thesis that we find in the third antinomy the assertion that man is free. It is, however, insufficient to invoke this reassuring convergence, first because what Kant says about the interest of reason is, to say the least, ambiguous, as we shall see, and then because the interest of reason is not in itself sufficient to solve the problem posed by the antinomy.

Let’s leave for now (as does Kant himself)20 the “resolution” of the antinomy and its “political” consequences. Like mankind in the political writings, let’s return, after a long detour around the globe, to Kant’s arguments in the “Universal History” text. We are getting to the point where we will be able to talk about cosmopolitanism (remember that the title of the text we are reading is “Idea for a Universal History with a Cosmopolitan Purpose”). We should not believe that the problems addressed by the first Critique—problems we have addressed primarily via the apparently analogical language of territories and frontiers—are to be held at a distance from these political questions. On the contrary, Kant himself, in the “Canon of Pure Reason,” still talking about the “interests of reason,” invokes—the passage is a famous one—the three questions that contain the whole interest of my reason: “What can I know?,” “What ought I to do?,” and “What may I hope?” Commenting on this passage in Kant and the Problem of Metaphysics, Heidegger affirms that these three questions (and therefore the interests of reason) define man not as a natural being but as a cosmopolitan, a citizen of the world, ein Weltbürger, and, to that extent, that they “constitute the domain of authentic philosophy.”21 If ever, then, the cosmo-political argument of the “Universal History” text were to prove insufficient, there would be nonnegligible consequences for true philosophy thus defined.

Let’s now suppose that natural teleology has exploited discord in order to disperse mankind across the whole surface of the globe, that there are people living in the most inhospitable places, that nature is exercising its providence by providing moss, sending driftwood, and so on. The man who fled west has come back from the east. The quasi-mechanical result of this process of dispersion in a finite space is that frontiers are produced, the state of nature returns at them, and that arrangements must be found, to the extent that flight is no longer possible. Here is Kant in the seventh proposition of his text, the one that poses the preprimary or postultimate problem:

What is the use of working for a law-governed civil constitution among individual men, i.e. of planning a commonwealth? The same unsociability which forced men to do so gives rise in turn to a situation whereby each commonwealth, in its external relations (i.e. as a state in relation to other states), is in a position of unrestricted freedom. Each must accordingly expect from any other precisely the same evils which formerly oppressed individual men and forced them into a law-governed civil state. Nature has thus again employed the unsociableness of men, and even of the large societies and states which human beings construct, as a means of arriving at a condition of calm and security through their inevitable antagonism. Wars, tense and unremitting military preparations, and the resultant distress [Not] which every state must eventually feel within itself, even in the midst of peace—these are the means by which nature drives nations to make initially imperfect attempts, but finally, after many devastations, upheavals and even complete inner exhaustion of their powers, to take the step which reason could have suggested to them even without so many sad experiences—that of abandoning a lawless state of savagery and entering a federation of peoples [Völkerbund] in which every state, even the smallest, could expect to derive its security and rights not from its own power or its own legal judgment, but solely from this great federation (Foedus Amphictyonum), from a united power and the law-governed decisions of a united will. (KPW, 47)

Kant, as we just saw, insists on the analogy between the previous stage of the process (individuals in the state of nature are obliged or constrained to leave it) and this one (States in the state of nature are in an identical situation). But this analogy is not without its problems. We immediately notice an obvious asymmetry: Individuals in the state of nature were faced with a literally intolerable situation—it was quite simply impossible for the individual to subsist without entering into a lawful civil state with other individuals. But one imagines that a State is not exactly in the same situation, for it could in principle survive alone (if we accept the Aristotelian definition whereby the State is defined by its autarkeia, or all but), without neighbors. What is more, the State in Kant is defined precisely by the fact that man has left the state of nature, and if he now falls back into it, because of the sphericity of the globe and the inevitability of the formation of frontiers, he cannot be back in it in the same way. Nothing allows us to think that the individual left the state of nature in order to find himself back in it in spite of himself, whereas this is exactly the situation of the State, which finds in spite of itself that the state of nature returns at the frontier.

Let’s allow a suspicion to form that this analogy, upon which Kant is so insistent, and which sustains the whole argument in favor of the inevitability of perpetual peace, hides an asymmetry that is troubling for all Kant’s thought—or at least his political thought. What is at stake here is at any rate clearly indicated by Kant from the very opening of the “Universal History” text. Given that human actions are neither immediately in conformity with reason (as they would be if humans were true rational cosmopolitans) nor the result of mere instinct (like the actions of animals), they always might present a spectacle of contingency and chaos that would be upsetting for the philosopher (who is here simply a man endowed with reason). To avoid this “dismay” (Unwillen, discontent, displeasure, depression), the philosopher must, without strictly theoretical justification but in the interest of reason, suppose that there exists a guiding thread (the natural providence we have been trying to understand from the start). This choice between the rational optimism of the guiding thread and the pessimism of chaos returns in a still more urgent tone in the middle of this seventh proposition, at the moment when the analogy between individuals and States begins to be a problem. But where the preamble suggests a two-branched dilemma (either chaos forever, which is depressing, or else the guiding thread of providence that will lead us inevitably to perpetual peace), here a third possibility seems to be on offer: namely, that perpetual peace arise from chaos by chance, by the pure chaotic action of chaos itself. Before, it seemed undeniable that in chaos there was only chaos, depressing for the philosopher (and therefore for humanity itself considered as rational), and order, if order there were, was hidden; now we have, through philosophical argument and the operator of the guiding thread, found at least some sense of order. The new possibility now invoked by Kant is that this order might in fact be born of the fortuitous action of chaos itself, without our even needing to postulate the existence of the rational guiding thread. The new possibility, troubling for Kant’s thinking, is that chaos might order itself in the absence of all teleology. Which is perhaps why Kant dismisses this possibility as soon as he mentions it, so as to return to the two possibilities mentioned in the preamble to the text:

Whether we should firstly expect that the states, by an Epicurean concourse of efficient causes, should enter by random collisions (like those of small material particles) into all kinds of formations which are again destroyed by new collisions, until they arrive by chance at a formation which can survive in its existing form (a lucky accident which is hardly likely ever to occur); or whether we should assume as a second possibility that nature in this case follows a regular course in leading our species gradually upwards from the lower level of animality to the highest level of humanity through forcing man to employ an art which is nonetheless his own, and hence that nature develops man’s original capacities by a perfectly regular process within this apparently disorderly arrangement; or whether we should rather accept the third possibility that nothing at all, or at least nothing rational, will anywhere emerge from all these actions and counter-actions among men as a whole, that things will remain as they have always been, and that it would thus be impossible to predict whether the discord which is so natural to our species is not preparing the way for a hell of evils to overtake us, however civilised our condition, in that nature, by barbaric devastation, might perhaps again destroy this civilised state and all the cultural progress hitherto achieved (a fate against which it would be impossible to guard under a rule of blind chance, with which the state of lawless freedom is in fact identical, unless we assume that the latter is secretly guided by the wisdom of nature)—these three possibilities boil down to the question of whether it is rational to assume that the order of nature is purposive in its parts but purposeless as a whole. (KPW, 48)

The Epicurean hypothesis, then, is dismissed as soon as it is invoked and already foreclosed by the end of this passage (too improbable a chance, on the one hand; always likely to reproduce barbarity, on the other) in which Kant ends up reformulating his question in the same terms as he did in the preamble.

Now the invocation of Epicurus and chance here does not happen at all by chance. Here we are deciding the problem of universal history in the name of the interest of reason, and almost tautologically: It is reasonable to prefer reason, is what Kant is basically saying, reason should choose itself rather than its other. And yet, the interest of reason is harder to discern when it is invoked on its own account in the first Critique, around the antinomy we were just reading, and this is because reason itself is harder to discern. The antinomy, as we recalled, comes about because of the action of reason itself, acting according to its nature, once it crosses the frontier of the understanding to find itself again in its apparently own proper domain or field, that of reason itself, where, however, it turns out to be a bit less than reasonable, at least to the extent that it gives rise to propositions that are perfectly contradictory but apparently irrefutable: about, for example, the question of the finitude of the world. If reason were really rational (if it were reduced to the understanding, the true domain or field of reason that has returned to itself),22 it would not need to consult its interest to know how to conduct itself. Reason has an interest, then, only to the extent that it is not (yet) fully itself, has not come back to itself, driven as it is to leave on a quest for its systematic circle.23 The interest of reason in this conflict is therefore not obvious, the more so in that reason in fact has several interests to consider and that these interests are most often themselves in contradiction or at least in competition with each other. Kant, wanting rapidly to characterize the two sides of the antinomy—the thesis, which is dogmatic (the world is finite; there are simples; there is a causality of freedom; there is a necessary being), and the antithesis, which is empiricist (the world is not finite; nothing is simple; the only causality is natural causality; there is no necessary being)—appeals for the antithesis to the very same Epicurus we have just seen ironically dismissed in the “Universal History” text.

In this part of the first Critique, at any rate, there are in fact three interests of reason to be considered: the speculative interest, the practical interest, and the popular interest. The thesis has going for it the practical interest, in that it sustains the “cornerstones of morality and religion,” whereas the antithesis robs us of all these supports, “or at least seems to rob us of them” (CPR, A466/B494). The thesis can also lay claim to a certain speculative interest, in that the thesis seems to allow for the complete grasp of the chain of conditions for a given conditioned, whereas the antithesis can only perpetually relaunch the search for the unconditioned that it never finds. Finally, it has in its favor the popular interest, which is easily satisfied with “a firm point to which it may attach the reins guiding its steps” (A467/B495). Everything leads one to think that this must be an invincible advantage and that Kant could hardly fail to support the thesis (at least insofar as he is considering the interests of reason), which also seems to be fully in accordance with what we have been reading in the political writings.

The thesis also has a proper name: Plato. The antithesis, which seems defeated in advance, is called Epicurus. Kant repeats that is has no practical interest, because it seems to undermine the bases of morality and religion, but he immediately recognizes that it has a considerable advantage on the side of the speculative interest of reason, if only because it firmly places knowledge in its proper domain, that of experience, where it allows an indefinite extension of conceptual knowledge without ever jumping or crashing through the fence and getting into the field or ocean of the Ideas of Reason. If empiricism limited itself to modest acceptance of its limits, while allowing practical reason to appeal to intellectual presuppositions and to faith (without claiming to have theoretical knowledge of them), all would be well, but it cannot help itself from crossing those same limits to denounce everything beyond them as illusion and fiction.24

Why does Kant choose Epicurus to illustrate the empiricism of the antithesis? In fact, the name Epicurus had already figured (again under the sign of exclusion) in this part of the first Critique, in the remark on the third antinomy, the one that bears on causality. He is mentioned in discussion of the thesis:

The confirmation of the need of reason to appeal to a first beginning from freedom in the series of natural causes is clearly and visibly evident from the fact that (with the exception of the Epicurean school) all the philosophers of antiquity saw themselves as obliged to assume a first mover for the explanation of motions in the world, i.e., a freely acting cause, which began this series of states first and from itself. For they did not venture to make a first beginning comprehensible on the basis of mere nature. (A450/B478)

Epicurus, then, and the school that bears his name, would be the most empiricist of empiricists, the only true empiricist in fact, or at least the only consistent one,25 the only one to push empiricism to the extreme consequences enumerated in the antithetic itself. Only Epicurus had the courage to be truly empiricist, to give up the origin. And this is how Epicurus appears regularly throughout Kant’s philosophy, often invoked, almost always denounced, a kind of ancient version of Hume, and yet more worrisome than Hume, more present.

The other side, the thesis, is, as we said, associated with Plato. He immediately has an advantage over Epicurus in Kant’s presentation, to the extent that it is via a reference to Plato that Kant had presented the very concept (or Idea) of a concept (or Idea) of reason as the very object of the whole Transcendental Dialectic. Epicurus may well have had the merit of recalling the understanding to its proper domain in experience, but he runs the risk of seeming out of place in the dialectic, where Plato is truly at home, to the point that it is from him that Kant borrows the very term “Idea,” which is the keystone of the whole discussion of reason here.

Now it is exactly here that Kant makes his remark about legislation in the domain of language and on dead and learned languages, which we quoted earlier. This remark, which will continue to give us food for thought, is followed immediately by another one, no less precious and enigmatic, on the relations between philosophy and philosophical reading (and, indeed, reading in general):

Plato made use of the expression idea in such a way that we can readily see that he understood by it something that not only could never be borrowed from the senses, but that even goes far beyond the concepts of the understanding (with which Aristotle occupied himself), since nothing encountered in experience could ever be congruent to it. Ideas for him are archetypes of things themselves, and not, like the categories, merely the key to possible experiences. [. . .] I do not wish to go into any literary investigation here, in order to make out the sense which the sublime philosopher combined with his word. I note only that when we compare the thoughts that an author expresses about a subject, in ordinary speech as well as in writings, it is not at all unusual to find that we understand him even better than he understood himself, since he may not have determined his concept sufficiently and hence sometimes spoke, or even thought, contrary to his own intention. (CPR, A313–4/B370)

So Plato, without exactly understanding what he meant, posited the importance of Ideas. It is perhaps no accident that Kant will defend him not only on the level of practical reason but also, and above all, on that of politics and, in particular, with regard to the Republic (the only Platonic text that Kant cites here). For—and this is in line with what we saw in the section on the interest of reason—these Platonic Ideas flourish in the domain of reason itself, where it is most itself (i.e., where it is not entirely reasonable), in its proper domain or field or sphere, beyond the understanding. Not that Plato entirely knew or understood this, which is why we needed an active and well-intentioned reader (i.e., Kant) really to understand what he meant without exactly saying it. Do not, then, make fun of Plato’s Republic on the grounds that it is impracticable, as does Brucker, but, just where Plato abandons us to such a miserable and shameful reading (CPR, A316/B373), take up the baton and think better than Plato what Plato gave us to read and think. The passage that follows (excerpted in Kant’s Political Writings and described by its editor as “the kernel of his political philosophy” [KPW, 15]) indeed contains the essential elements of the fifth proposition of the “Universal History” text, of “Theory and Practice,” and the appendix on morality and politics in Perpetual Peace (which we shall be reading in detail later):

The Platonic republic has become proverbial as a supposedly striking example of a dream of perfection that can have its place only in the idle thinker’s brain; and Brucker finds it ridiculous for the philosopher to assert that a prince will never govern well unless he participates in the ideas. But we would do better to pursue this thought further, and (at those points where the excellent man leaves us without help), to shed light on it through new endeavors, rather than setting it aside as useless under the very wretched and harmful pretext of its impracticability. A constitution providing for the greatest human freedom according to laws that permit the freedom of each to exist together with that of others (not one providing for the greatest happiness, since that would follow of itself) is at least a necessary idea, which one must make the ground not merely of the primary plan of a state’s constitution but of all the laws too; and in it we must initially abstract from the present obstacles, which may perhaps arise not so much from what is unavoidable in human nature as rather from neglect of the true ideas in the giving of laws. For nothing is more harmful or less worthy of a philosopher than the vulgar appeal to allegedly contrary experience, which would not have existed at all if institutions had been established at the right time according to the ideas, instead of frustrating all good intentions by using crude concepts in place of ideas, just because these concepts were drawn from experience. The more legislation and government agree with this idea, the less frequent punishment will become, and hence it is quite rational to assert (as Plato does) that in perfect institutional arrangements nothing of the sort would be necessary at all. Even though this may never come to pass, the idea of this minimum is nevertheless wholly correct when it is set forth as an archetype, in order to bring the legislative constitution of human beings ever nearer to a possible greatest perfection. For whatever might be the highest degree of perfection at which humanity must stop, and however great a gulf [Kluft] must remain between the idea and its execution, no one can or should try to determine this, just because it is freedom that can go beyond every proposed boundary. (CPR, A316–17/B372–74)

From the point of view of the Ideas, then, of reason itself, it would seem obvious that the interest of reason (at the very least its political interest) would be attached to Plato rather than Epicurus, to the thesis as against the antithesis, just as we saw the “Universal History” text rapidly dismissing Epicurus when it comes to the possibility of peace.

And yet, things are not so simple. Just as Plato can suffer at the hands of too rapid a reading, so Epicurus has been badly served by his readers (and translators). First of all, as one would expect, Epicurus, on the side of the antithesis, has the advantage when it comes to the speculative interest of reason, when reason folds itself back onto the terrain of the understanding, the domain par excellence of reasonable reason, and leaves that terrain only to prevent the understanding from trying to escape its bounds. So once we have recognized the uselessness of empiricism when it comes to the practical interest:

On the contrary, however, empiricism offers advantages to the speculative interests of reason, which are very attractive and far surpass any that the dogmatic teacher of the ideas of reason might promise. For with empiricism the understanding is at every time on its own proper ground, namely the field solely of possible experiences, whose laws it traces, and by means of which it can endlessly extend its secure and comprehensible cognition. Here it can and should exhibit its object, in itself as well as in its relations, to intuition, or at least in concepts an image for which can be clearly and distinctly laid before it in similar given intuitions. Not only is it unnecessary for the understanding to abandon this chain of natural order so as to hang onto ideas with whose objects it has no acquaintance because, as thought-entities, they can never be given; but it is not even permitted to abandon its business and, under the pretext that this has been brought to an end, to pass over into the territory of idealizing reason and transcendent concepts, where there is no further need to make observations and to inquire according to the laws of nature, but rather only to think and invent, certain that it can never be refuted by facts of nature because it is not bound by their testimony but may go right past them, or even subordinate them to a higher viewpoint, namely that of pure reason. (CPR, A468/B496)

So everything seems to be clear, divided up, clearly delimited on both sides of the antithetical frontier: on the one side, Plato and morality; on the other, Epicurus and science. Or, at least, Platonism on the one side, Epicureanism on the other. If we need to extend Plato a little by an effort of reading, one should perhaps do the same for Epicurus:

Each of the two says more than it knows, but in such a way that the first encourages and furthers knowledge, though to the disadvantage of the practical, the second provides principles which are indeed excellent for the practical, but in so doing allows reason, in regard to that which only a speculative knowledge is granted us, to indulge in ideal explanations of natural appearances, and to neglect the physical exploration of them. (CPR, A472/B500)

We know how Plato says more than he realizes, and what that more is, when given a little help from the reader. What about Epicurus? How does he go beyond the obvious or first reading one might be tempted to go in for? Well, by detaching himself, perhaps, from Epicureanism itself. Where Plato goes beyond himself by going beyond experience when it comes to practical reason, and thus confirms and comforts the position of the (“dogmatic”) thesis in the antithetic, Epicurus is detached from himself in a way that destabilizes the symmetry of the two sides, which will mean that the antithetical character of the antithetic will be a little disturbed. This happens not so much in the main text but in a footnote called for by the short dramatic paragraph that interrupts the analysis of empiricism (the antithesis, then) after the first two interests of reason (the practical and the speculative), unable to wait for the analysis of the third interest, the popular (whereas nothing of the sort happens on the side of the thesis: no little paragraph, no footnote had interrupted the presentation of the thesis as to all three interests of reason). This little paragraph, a single sentence that sits on the page like a sentence from Bouvard and Pécuchet, reads simply: “Dies ist der Gegensatz des Epikureisms* gegen den Platonisms” (CPR, A471/B499: “This is the opposition of Epicureanism* and Platonism”).

The élan of this sentence, a summary or dramatic tagline for everything that is being developed in these pages of the first Critique, is cruelly interrupted by the footnote asterisk, which refers the reader, even before the end of this clear and decisive sentence, to a long note immediately concerned to separate Epicurus from what is here being asserted. Through an effort of reading perhaps analogous to what was demanded in the case of Plato, we can remove Epicurus “himself” from the antithesis by exempting him from the thesis of the antithesis (because an anti-thesis is still a thesis, it tends to become dogmatic, maintains its clearly defined place on its side of the antithetical fence only by entering into a more or less secret agreement with the thesis as to the proper—thetic—form of its statements). Epicurus himself does not, perhaps, belong on the side of the antithesis, and so perhaps does not become thetic and dogmatic, because he would perhaps not have maintained the thesis of the antithesis as a thesis but only as a hypothesis or working maxim for the conduct of the understanding. And, by saying “perhaps” in this way, by saying it is still a question (whereas in recommending that we read Plato beyond Plato or better than Plato on the side of the thesis, no such hesitation appears, and the remarks on reading have a slightly dogmatic tone, precisely, a thetic tone that brings out the thesis more clearly), Kant, discussing the hypothetical character of Epicurus’s thought, himself remains, as is only right, in the hypothetical, without really advancing a thesis about the hypothesis of Epicurus’s hypothesis—a hypothesis that, if it were true, would make of Epicurus a true philosopher.26

Here, then, is the note from the dramatic little sentence about Epicureanism and Platonism:

There is still a question, however, whether Epicurus ever presented these principles as objective assertions. If they were perhaps nothing more than maxims of the speculative employment of reason, then in them he would have shown as genuine a philosophical spirit as any of the sages of antiquity.27 That in the explanation of appearances one must go to work as though [als ob] the field of investigation were not cut off [abgeschnitten] by any boundary or beginning of the world [we are in the first antinomy]; that one must assume the material of the world as it has to be if we are to be taught about it by experience; that no other way of generating occurrences than their determination through unalterable natural laws, and finally that no cause distinct from the world are to be employed: even now these are principles, very correct but little observed, for extending speculative philosophy while finding out the principles of morality independently of alien sources; if only those who demand that we ignore those dogmatic propositions, as long as we are concerned with mere speculation, might not also be accused of trying to deny them. (CPR, A471n/B499n)

So Epicurus seems to withdraw a little from the antithetic by resisting the becoming-dogmatic or the becoming-thesis of the antithesis.28 We advance the hypothesis of the antithesis for the good of theoretical knowledge, according to an as if that cannot fail to recall the formulations of the categorical imperative and that troubles the very distinction between speculative and practical interest to the extent that the very concept of an interest of reason already pushes it toward the practical domain, as Kant indeed recognizes. Once they have been taken as hypo-theses, Epicurus’s principles leave the practical domain to practical reason (whereas if they are taken as theses, they cannot fail to invade the practical domain and declare it a domain of pure fiction), while allowing the development of the speculative according to an interest that will itself end up being practical too.29

Is it by chance that Epicurus, the philosopher of chance, should also appear in the antinomy of practical reason? Let’s recall that this antinomy, which is as natural and naturally beneficial as that of speculative reason, is born from the Idea of the Supreme Good, the unconditioned in the practical domain, which cashes out into contradictory relations between happiness and virtue. In this antinomy, Epicurus, or the Epicurean, shows up again (as the one who believes that virtue consists in the pursuit of happiness) but is this time opposed not to Plato but to the Stoics, who believe the opposite. And even though the Epicurean position will be dismissed as “altogether false” (CPrR, 105), an effort of reading like the one we have noted in the first Critique will again be brought to bear in order to separate Epicurus himself a little from his own doctrine or canonic, which also softens the rigor of the antinomy itself:

For, Epicurus as well as the Stoics extolled above all the happiness that arises from consciousness of living virtuously; and the former was not so base in his practical precepts as one might infer from the principles of his theory, which he used for explanation and not for action, or as they were interpreted by many who were misled by his use of the expression “pleasure” for “contentment”; on the contrary, he reckoned the most disinterested practice of the good among the ways of enjoying the most intimate delight and included in his scheme of pleasure (by which he meant a constantly cheerful heart) such moderation and control of the inclinations as the strictest moral philosopher might require. (CPrR, 96–97)30

My point is not at all to save Epicurus (who will, in any case, be back to cause more trouble in our reading of the Antinomy of Judgment in Chapter 5), or even to vaunt his thinking, but to take this proper name as the index of trouble in Kant, trouble that is perhaps the whole of Kant’s thought (once we have decided to make an effort of reading and to continue with this effort even when he deserts us, etc.). This trouble—which here affects the frontier between thesis and antithesis in the antinomy, and thereby the strictly antinomic character of the antinomy, the dialectical character of the transcendental dialectic, and, step by step, the distinction between pure speculative and pure practical reason, etc.—can also be seen under the sign of nature (she wanted the discord), which returns as the (at least potential) violence that lurks along the frontiers of States formed when mankind, now dispersed across the terrestrial globe, is obliged to re-encounter and negotiate this “chaotic” (KPW, 49) situation.

Is there in all this a natural process, like the one that is supposed to push human reason beyond all reasonable limit? We can judge this analogically, following natural (here cosmic) teleology:

The real test is whether experience can discover anything to indicate a purposeful natural process of this kind. In my opinion, it can discover a little; for this cycle of events seems to take so long a time to complete, that the small part of it traversed by mankind up till now31 does not allow us to determine with certainty the shape of the whole cycle, and the relation of its parts to the whole. It is no easier than it is to determine, from all hitherto available astronomical observations, the path which our sun with its whole swarm of satellites is following within the vast system of the fixed stars; although from the general premise that the universe is constituted as a system and from the little which has been learnt by observation, we can conclude with sufficient certainty that a movement of this kind does exist in reality. (KPW, 50)

So we judge the natural purposiveness of human affairs according to the analogy of purposiveness in the natural world, that purposiveness itself being freely postulated from the point of view of the human in order to cheer up reason, which does not like chance and finds it dismaying. And yet, the same reason that loves nature and its immutable laws when it comes to phenomena (recall that this is the forte of Epicurean empiricism) runs the risk of allowing natural analogism to settle in a domain that ought in principle to come under the heading of the noumenal rather than the phenomenal—namely, the political history of humanity. We are slowly approaching the thought that “politics” might be the name for the zone in which that analogism cannot be decided according to the major Kantian distinctions between phenomenon and noumenon, between speculative and practical reason, and “law” might be the name for the undecidable utterance that is spoken in that place. Thus, for example, just before the passage from the “Universal History” text that seemed to dismiss Epicurus, the explanation in terms of natural purposiveness is shadowed by a troubling mechanistic explanation. Kant has just said that nature pushes the States toward the formation of a Society of Nations:

However wild and fanciful this idea may appear—and it has been ridiculed as such when put forward by the Abbe St Pierre and Rousseau (perhaps because they thought that its realisation was so imminent)—it is nonetheless the inevitable outcome of the distress in which men involve one another. For this distress must force the states to make exactly the same decision (however difficult it may be for them) as that which man was forced to make, equally unwillingly, in his savage state—the decision to renounce his brutish freedom and seek calm and security within a law-governed constitution. All wars are accordingly so many attempts (not indeed by the intention of men, but by the intention of nature) to bring about new relations between states, and, by the destruction or at least the dismemberment of old entities, to create new ones. But these new bodies, either in themselves or alongside one another, will in turn be unable to survive, and will thus necessarily undergo further revolutions of a similar sort, till finally, partly by an optimal internal arrangement of the civil constitution, and partly by common external agreement and legislation a state of affairs is created which, like a civil commonwealth, can maintain itself automatically [so wie ein Automat sich selbst erhalten kann]. (KPW, 47–48)

This Idea of Mechanical Reason seems to be in every sense the end of nature and the end of politics. There is no nature left in this Idea, insofar as nature puts an end to itself in it or, by the same token, becomes a perfected nature that embraces everything in its mechanics, politics included, which also puts an end to itself here. Nature ends in politics and vice versa. The end of nature is the end of nature; the end of politics is the end of politics. (No) more nature: (no) more politics. The point is not to criticize these propositions of Kant’s, because we want to suggest on the contrary that this is the truth of a whole tradition of political thought, which finds itself obliged to think of politics in view of an end that would put an end to it. Political thought is drawn toward the vision of a future in which politics would have disappeared in its self-realization, which would also be the end of nature and the end of history. And by a fatality inscribed in the very form of this thinking, what supervenes when there is no more politics is nature, which, as we have sufficiently seen, is violence itself. So nature, coming to its end in mechanical politics, would have to reappear as absolute violence.32

It is the law twice over: We must leave the state of nature, and we ought to leave it. The States must find their external (and thus internal) equilibrium, and they ought to find it. Perpetual peace must come about, and it ought to come about. It is the law. And the coincidence of the two senses of the word law is called nature. Opposed to physis, nomos will always have been called and absorbed by it.

This mechanical language will dominate the end of the “Universal History” text. After the passage in which Epicurus and chance are dismissed, Kant continues (my emphasis except on the last two lines):

While the purposeless state of savagery did hold up the development of all the natural capacities of human beings, it nonetheless finally [endlich]33 forced them [sie nötigte], through the evils in which it involved them, to leave this state and enter into a civil constitution in which all their dormant capacities could be developed. The same applies to the barbarous freedom of established states. For while the full development of natural capacities is here likewise held up by the expenditure of each commonwealth’s whole resources on armaments against the others, and by the depredations caused by war (but most of all by the necessity of constantly remaining in readiness for war), the resultant evils still have a beneficial effect. For they compel [nötigen] our species to discover a law of equilibrium to regulate the essentially healthy hostility which prevails among the states and is produced by their freedom. Men are compelled to reinforce this law by introducing a system of united power [eine vereinigte Gewalt] hence a cosmopolitan system of general political security. This state of affairs is not completely free from danger, lest human energies should lapse into inactivity, but it is also not without a principle of equality governing the actions and counter-actions of these energies, lest they should destroy one another. (KPW, 49)

This language of force, or rather of forces in the plural that resist each other, of a dynamic equilibrium to be found, gives a certain primacy to the relations between States, even if, as Kant says, this is still happening in view of the internal organization of the State, itself done with a view to the culture and, ultimately, the morality of the citizens. For, as we learn from Religion within the Boundaries of Mere Reason, we are still in the ethical state of nature even if we have left behind the juridical state of nature.34 But all of that depends on this mechanical action of the States confined within the finite space of the globe. In the Perpetual Peace text, this priority of external relations (whereby the postultimate problem becomes the preprimary problem) is made still clearer and allows us to see what I would like to call the “external abyss” of this argument:

Even if people were not compelled by internal dissent to submit to the coercion of public laws, war would produce the same effect from outside. For in accordance with the natural arrangement described above, each people would find itself confronted by another neighbouring people pressing in upon it, thus forcing it to form itself internally into a state in order to encounter the other as an armed power. (KPW, 112)

From the political point of view, one could always start with this mechanical logic, which works in terms of an external force, even though that would mean forgetting the story Kant tells us elsewhere, beginning with the individual in the state of nature. This argument must be valid in general, but if every State were to be formed this way, we would have to refine its logic. Kant is thinking of one virtual State surrounded by already-constituted States, which would make it into a State by pure mechanical pressure. But if we try to think every State was formed in this way, we have to suppose a general “not yet,” in which all neighboring States would be formed by the pressure of an absolute exterior that would precede the formation of any State at all. In which case we would no doubt have to look toward catastrophic morphogenesis (or perhaps Epicurean physics) to find a mechanics capable of describing such a situation.

Kant, of course, does not go in this direction. The guiding thread of all this thinking, that of teleological nature, pushes him, rather (in the eighth proposition) to assert that all human history (in this optic chosen by the philosopher to avoid the dismay caused by the spectacle of the empirical chaos of history) can be considered to be the realization of nature’s secret design. This is where Kant launches himself into the cosmic analogy we saw, which allows him to see already, in Europe—where commercial relations are tending to replace warfare as the ordinary means of communication between States—the sketch of “a great political body of the future, without precedent in the past,” which encourages the thought that one day will see the arrival of “the highest purpose of nature, a universal cosmopolitan existence . . . the matrix within which all the original capacities of the human race may develop” (KPW, 51).

We shall see in a moment what to think of the “cosmopolitan situation” announced here. Before we get there (it is true that it is starting to keep us waiting), what of this mechanistic teleology of Kant’s? We asked whether the state of nature was to be abandoned through obligation or through necessity. It is henceforth obvious that it is both at once, or else (but that is nature) that obligation cashes out as natural necessity. Practical reason, qua rational, prescribes cosmopolitan existence as alone worthy of human reason, and natural necessity will make it inevitable. Even if mankind does not care to leave nature for the cosmopolitan automaton out of duty, nature will see to it anyway. What ought to happen will infallibly happen, if at least we follow the guiding thread (which does not give rise to theoretical proof)—but we ought to follow it at any rate, and in any case we must follow it, because following the guiding thread according to which nature will lead mankind to its end is part of the natural mechanism that leads mankind to its end. We keep our freedom of choice (we can even prefer the Epicurean hypothesis), but the interest of practical reason prescribes that we follow the idea that renders that same prescription pointless, because it will be followed in any case.

We can read this situation in at least two ways. The first reading will find that all is well that ends well, that morality is shored up by the thought that even if humans do not act according to reason, then natural mechanism will, a little slowly it is true, end up doing it for them. But the second reading might not be so sanguine, for a moral end that comes about by necessity is no longer moral at all; the cosmopolitan machine that functions like an automat would leave no place for morality at all. Before the arrival of the cosmopolitan state, the States do not behave morally, for they are constantly waging war on each other. But when war has finally brought about peace in cosmopolitan form, there will be no morality anymore, because that state will maintain itself automatically, in perpetuity. The end of the end is the end of the end.

So it is not for nothing that the text that deals explicitly with the perpetual peace of this cosmopolitan situation should be placed literally under the sign of death:

the perpetual peace

A Dutch innkeeper once put this satirical inscription on his signboard, along with the picture of a graveyard. We shall not trouble to ask whether it applies to men in general, or particularly to heads of state (who can never have enough of war), or only to the philosophers who blissfully dream of perpetual peace. (KPW, 93)

This perspective of death will turn out to be powerful enough to cast doubt on the whole argumentation we have been following. If the “Universal History” text, which has guided us up until now, closes on hope, and the projection—as freely inevitable, so to speak—of the cosmopolitan situation of perpetual peace, the text bearing that name, the worrying opening of which we have just read, will show us not only that such a peace really is death, and that—contrary to all classical readings of this text—we must do everything to avoid it, but that all Kant’s frontiers and distinctions are threatened in the very tracing of their line, and that the definition of critique itself will not survive unscathed.


1. See Jean-François Lyotard, Des dispositifs pulsionnels (Paris: Union générale d’édition, 1973), 268. In fact, this empty space at the heart of the City can only be thought as resulting from an invagination of the outside—indeed, a radical exteriority—that we are, precisely, calling “the frontier.”

2. I use the word “foreclosure” here to name a problem, because it seems necessary to have understood the frontier to understand foreclosure (in the psychoanalytical sense). See Lyotard, Le différend, §218. What Lyotard calls “differend” is, moreover, not so different from what Kant calls “state of nature,” as suggested by Kant’s further analysis in the “Doctrine of Right” (§42): “It is true that the state of nature need not, just because it is natural, be a state of injustice (iniustus), of dealing with one another only in terms of the degree of force each has. But it would still be a state devoid of justice (status iustitia vacuus), in which, when rights are in dispute (ius controversum), there would be no judge competent to render a verdict having rightful force. Hence each may impel the other by force to leave this state and enter into a rightful condition.” Kant’s refusal of the differend motivates, among other things, his denial of any right to revolt, because in such a case there would be no judge competent to decide between the right of the sovereign and the right of the people. See KPW, 81, a passage to which we return in Chapter 4. In Lyotard, the Kantianism of the definition of the differend is tempered by a Hegelianism whereby it wants to be expressed, if only as a litigation (see Le Différend, §§22–23).

3. In the “Doctrine of Right,” Kant shows that the truly political (civil) association cannot be essentially communitarian: “The civil union (unio civilis) cannot itself be called a society, for between the commander (imperans) and the subject (subditus) there is no partnership [Mitgenossenschaft]” (MM, §41, 85). We will later draw the consequences of this at the level of international politics.

4. Immanuel Kant, Prolegomena to Any Future Metaphysics That Will Be Able to Come Forward as Science, tr. Gary Hatfield, rev. ed. (Cambridge: Cambridge University Press, 2004), 11.

5. “It is nonetheless the desire of every state (or its ruler) to achieve lasting peace by thus [i.e., by creating a universal monarchy] dominating the whole world, if at all possible” (KPW, 113).

6. Recall also the link of violence and dispersion in Aristotle: The separated individual must be a lover of polemos. But dispersion, scattering, also happens in order to avoid warfare. There is scatter whether or not there is violence in the narrower sense: archi-violence. This is where we would need to reread in Rousseau the supposedly peaceful scatter of the state of nature.

7. Immanuel Kant, Critique de la raison pure, tr. A Tremesaygues and B. Pacaud (Paris: Vrin, 1944).

8. See also the use of the horizon to explain the (only) apparent moral decline of mankind in “Theory and Practice”: “It can be shown that the outcry about man’s continually increasing decadence arises for the very reason that we can see further ahead, because we have reached a higher level of morality. We thus pass more severe judgments on what we are, comparing it with what we ought to be, so that our self-reproach increases in proportion to the number of stages or morality we have advanced through during the whole of known history” (KPW, 89). See the Appendix for a reading of this extended image from the first Critique specifically as it resorts to the (unstable) distinction between the concepts of Schranke and Grenze.

9. We might, for example, invoke the fictitious philosopher de Selby in Flann O’Brien’s novel The Third Policeman, who shows without difficulty that the world is, in fact, sausage-shaped. See my essay “Introduction to Economics I: Because the World Is Round,” in Bataille: Writing the Sacred, ed. Carolyn Bailey Gill, 46–57 (London: Routledge, 1995).

10. Bernard de Fontenelle, Entretiens sur la pluralité des mondes (Paris: Blargeart, 1686).

11. Especially in the precritical text Universal Natural History and Theory of the Heavens or Essay on the Constitution and the Mechanical Origin of the Whole Universe According to Newtonian Principles (1755) (in Immanuel Kant: Natural Science, ed. Eric Watkins, 182–308 [Cambridge: Cambridge University Press, 2012]), the third part of which develops conjectures as to the inhabitants of the different planets, because, even though Kant is prepared to think that some planets may be uninhabited, or not yet inhabited, he maintains that “most of the planets are certainly inhabited and those that are not will be at some stage” (297). See also §91 of the third Critique (to which we return in Chapter 5): “To assume rational inhabitants of other planets is a matter of opinion; for if we could approach more closely to other planets, which is intrinsically possible, we could determine by means of experience whether they exist or not; but we never will come close enough to other planets, so this remains a matter of opinion.”

12. Note the naturalism here: We have said that there are no natural frontiers, but an island can give a strong impression that there are anyway (we are going to see how things stand in this regard with a whole planet) and that they are unchangeable and necessary. In spite of himself, Lyotard inherits something of this naturalism when he takes up Kant’s description and talks of an archipelago of discursive genres. See Le différend, 189ff.

13. See also the echo of this passage later, in the “Paralogisms of Pure Reason” (but only in the first edition version): “Nothing but the sobriety of a strict but just criticism can liberate us from these dogmatic semblances, which through imagined happiness hold so many subject to theories and systems, and limit all our speculative claims merely to the field of possible experience, not by stale mockery at attempts that have so often failed, or by pious sighing over the limits of our reason, but by means of a complete determination of reason’s boundaries according to secure principles, which with the greatest reliability fastens its nihil ulterius on those Pillars of Hercules that nature has erected, so that the voyage of our reason may proceed only as far as the continuous coastline of experience reaches, a coastline that we cannot leave without venturing out into a shoreless ocean, which, among always deceptive prospects, forces us in the end to abandon as hopeless all our troublesome and tedious efforts” (A395–96).

14. I look more directly at the complex (and inconsistent) relation between the terms Schranke and Grenze in Kant’s understanding of frontiers in the Appendix.

15. For the language of enclosure, compare this somewhat incoherent analogy from the “Universal History” essay: “But once enclosed within a precinct [Gehege, an enclosure, a paddock] like that of civil union, the same inclinations [those toward “wild freedom”] have the most beneficial effect. In the same way, trees in a forest [in einem Walde], by seeking to deprive each other of air and sunlight, compel each other to find these by upward growth, so that they grow beautiful and straight—whereas those that put out branches at will, in freedom and in isolation from others, grow stunted, bent and twisted” (KPW, 46).

16. But who always might run aground, deliberately or not, like Hume: “Nothing from all that has been provided before now could be used except the hint that Hume’s doubts had been able to give; Hume also foresaw nothing of any such possible formal science, but deposited his ship on the beach (of skepticism) for safekeeping, where it could then lie and rot, whereas it is important to me to give it a pilot, who, provided with complete sea-charts and a compass, might safely navigate the ship wherever seems good to him, following sound principles of the helmsman’s art drawn from a knowledge of the globe” (Prolegomena, 11–12).

17. This is the famous and eminently political story told in the preface to the first edition of the first Critique, which give metaphysics the status of a “battlefield,” the history of which is hard to organize according to a linear development or teleological fulfillment.

18. It goes without saying that the first Critique never stops talking about limits, boundaries, and frontiers, from beginning to end, to the point where it would be pointless to give references; it is precisely this ubiquity of the frontier that we are trying to understand. Let’s say for now that Kant is always trying to (de-)limit the limit or draw a frontier on or in the frontier. But if there is one place where the frontier is concentrated, that place is the Antinomy.

19. “These sophistical assertions thus open up a dialectical battlefield, where each party will keep the upper hand as long as it is allowed to attack, and will certainly defeat that which is compelled to conduct itself merely defensively. Hence hardy knights, whether they support the good or the bad cause, are certain of carrying away the laurels of victory if only they take care to have the prerogative of making the last attack, and are not bound to resist a new assault from the opponent. One can easily imagine that from time immemorial this arena has often been entered, both sides gaining many victories, but that each time the final victory was decisive merely because care was taken that the champion of the good cause held the field alone, his opponent having been forbidden to take up his weapons again. As impartial referees we have to leave entirely aside whether it is a good or a bad cause for which the combatants are fighting, and just let them settle the matter themselves. Perhaps after they have exhausted rather than injured each other, they will see on their own that their dispute is nugatory, and part as good friends” (CPR, A422–23/B450–51).

20. “For now we will postpone this fundamental inquiry a little longer” (CPR, A465/B493).

21. Martin Heidegger, Kant and the Problem of Metaphysics, 5th enlarged ed., tr. Richard Taft (Bloomington: Indiana University Press, 1997), 145. Heidegger also takes up a fourth question (“What is the Human Being?”) from the much less reliable text of the Logic, which will allow him to embark on what it is tempting to call the anthropologizing drift of the fourth part of the Kantbuch.

22. “In its speculative use reason led us through the field of experiences, and, since it could never find complete satisfaction for itself there, it led us on from there to speculative ideas, which in the end, however, led us back again to experience, and thus fulfilled its aim” (CPR, A804/B832).

23. It is indeed Kant and not Hegel who is speaking: “Reason is driven by a propensity of its nature to go beyond its use in experience, to venture to the outermost bounds of all cognition by means of mere ideas in a pure use, and to find peace only in the completion of its circle in a self-subsisting systematic whole” (CPR, A797/B825).

24. See the Prolegomena, 7, and the Appendix in this text.

25. “Epicurus on his part at least proceeded more consistently in accord with his sensual system (for in his inferences he never exceeded the bounds of experience) than Aristotle and Locke” (CPR, A854/B882).

26. Having denounced hypotheses and all that even looks like them in the preface to the first edition of the Critique as “forbidden commodities” (Axv), in the “Discipline of Pure Reason,” Kant asserts that hypotheses have a role in the exercise of pure reason only as polemical instruments, to be used only to show that the other has as little reason to maintain that his claim is of the order of knowledge as we would to claim it of the opposite. This is precisely what happens in the Antithetic, and in both cases there is a language of endless and unresolved battles and of weapons we should not be afraid of using, precisely in the interests of perpetual peace: “Hypotheses are therefore allowed in the field of pure reason only as weapons of war, not as grounding a right, but only for defending it. However, we must always seek the enemy here in ourselves. For speculative reason in its transcendental use is dialectical in itself. The objections that are to be feared lie in ourselves. We must search them out like old but unexpired claims, in order to ground perpetual peace on their annihilation. External quiet is only illusory” (CPR, A777/B805). The only true peace of reason is perpetual peace, and we have to pass through the violence of hypotheses in order to achieve it. See too the famous comparison with Copernicus in the preface to the second edition of the first Critique: “In the same way, the central laws of the motion of the heavenly bodies established with certainty what Copernicus assumed at the beginning only as a hypothesis, and at the same time they proved the invisible force (of Newtonian attraction) that binds the universe, which would have remained forever undiscovered if Copernicus had not ventured, in a manner contradictory to the senses yet true, to seek for the observed movements not in the objects of the heavens but in their observer. In this Preface I propose the transformation in our way of thinking presented in criticism merely as a hypothesis, analogous to that other hypothesis, only in order to draw our notice to the first attempts at such a transformation, which are always hypothetical, even though in the treatise itself it will be proved not hypothetically but rather apodictically from the constitution of our representations of space and time and from the elementary concepts of the understanding” (Bxxii). Kant does not use the word revolution about Copernicus here, but as we shall see later, the radical hypothesis does indeed have a revolutionary character that will trouble Kant’s analogy between his critical thought and his political thought.

27. “[Die] Weltweisen des Altertums”: Norman Kemp Smith’s classic 1929 translation (London: Macmillan, 1929) has “philosophers of antiquity” (as indeed does Werner Pluhar’s more recent translation (Indianapolis: Hackett Publishing Company, 1996). This would give Epicurus’s radical empiricism greater philosophical standing than Kant is strictly allowing—as a specialist of Hume, Kemp Smith might have enjoyed the more eminent status his translation gives to Epicurus, but Kant’s term is less committal, making Epicurus a philosopher among “sages” (who are not, perhaps, in general philosophers) rather than among philosophers, and thus maintains the atmosphere of the “perhaps” we are pursuing here.

28. This privileged position of Epicurus can already be seen in the precritical Theory of the Heavens. In his preface, Kant, more Seefahrer or astronaut, and less prudent that in the first Critique (“I have dared to undertake a dangerous journey on the basis of a slight supposition and already see the foothills of new lands. Those who have the courage to pursue the exploration, will step onto those lands and have the pleasure of bestowing their own name upon them” [194]), envisages, even if only to reject it, the hypothesis that “if the blind mechanism of the powers of nature knows how to develop so magnificently and to such perfection all of its own accord” (ibid.) and says that if this were the case, then, “Epicure [sic] lives again in the middle of Christendom” (ibid.). A little later, he accepts some degree of convergence between his own system and Epicurus’s, taking his distance on a single point—namely, that Epicurus “was even so impudent that he insisted that the atoms deviated from their straight motion without any reason in order to be able to encounter one another” (198), whereas this very absence of apparent cause, given the order and regularity of the laws of nature, would be the sign of “an all-sufficient highest mind in which the natures of things were designed in accordance with unified purposes” (199). There is a definite proximity here between truth and error that reappears in the Antithetic we are reading.

29. As Kant puts it, “All interest is ultimately practical” (CPrR, 102). See also CPR, A797/B825 (“If there is to be any legitimate use of pure reason at all [. . .] this will concern not the speculative but rather the practical use of reason,”) and CPrR, 100–2, and more especially the following, which prepares the ground for our discussion of teleological judgment in Chapter 5: “The ultimate aim of nature which provides for us wisely in the disposition of reason is properly directed only to what is moral” (CPR, A801/B829). This practical end of reason in general is “foreign to transcendental philosophy,” which can only tend to pull reason back toward the understanding (ibid.). Kant appends a note to the word foreign (fremd) to explain that because all practical concepts in the end concern feelings, and therefore pleasure and pain, they do not belong to the domain of transcendental philosophy, which deals with knowledge and not feelings. This motif of the foreign also plays a role in the discussion of this issue in the second Critique in a subsection, the title of which proclaims the primacy of the practical over the speculative: “Whether speculative reason, which knows nothing about all that which practical reasons offers for its acceptance, must accept these propositions and, although they are transcendent for it, try to unity them, as a foreign possession [wie einen fremden Besitz] handed over to it, with its own concepts, or whether it is justified in obstinately following its own separate interest and, in accordance with the canon of Epicurus, rejecting as empty subtle reasoning everything that cannot accredit its objective reality by manifest examples to be shown in experience, however much it might be interwoven with the interest of the practical (pure) use of reason and in itself not contradict the theoretical, merely because it actually infringes upon the interest of speculative reason to the extent that it removes the bounds which the latter has set itself and hands it over to every nonsense or delusion of imagination?” (CPrR, 101). The foreigner in the house of reason is clearly enough the moral law, which is entirely within the paradoxical logic of the legislator. There is legislation only of the foreign.

30. See also the end of the Doctrine of Virtue in the Metaphysics of Morals, where the “ever-cheerful heart,” recommended by “the virtuous Epicurus,” is part of the practice of virtue (§53).

31. A page earlier Kant states with some confidence that we are “a little beyond the half-way mark” (KPW, 49).

32. This is again the principle of science fiction, to which Kant opens the (milky) way.

33. One senses the awkwardness of Kant’s position: The state of nature is a state without finality (and thus in a sense not even natural), but finally this finality has to happen to it, though this must above all not happen by chance.

34. Immanuel Kant, Religion within the Boundaries of Mere Reason and Other Writings, tr. Allen Wood and George di Giovanni (Cambridge: Cambridge University Press, 1998), 106–7.