Meillassoux’s Reinterpretation of Kant’s Transcendental Dialectic

1 Introduction

Speculative realism shares with absolute idealism the claim that it is possible to maintain a speculative relation to the absolute. Contrary to the antirealist paradigm, dominant for so long in post-critical philosophy, speculative realism aims to turn “once more toward reality itself”.^[1] But, contrary to idealism, speculative realism also decidedly proclaims its realism as anti-idealist. In such programmatic declarations, we thus find expressed the intention to undo the steps taken by the German idealists, who set aside the Kantian reservations against unlimited speculation about the absolute, while at the same time not giving up the search for a proper speculative philosophy but rather looking for alternatives to the path taken by idealism. In this regard, it may prove fruitful to read such a program together with Kant’s Critique of Pure Reason – for conceiving in an anti-idealist way of a form of speculation about reality in itself can be understood as revisiting the origin of post-critical speculative thought, the Transcendental Dialectic.

Which aspects of this difficult and highly ambiguous text can be considered as being of particular interest from the perspective of contemporary philosophy? In the introduction and the first book, Kant sketches out a syllogistic “system of the transcendental ideas”^[2] by means of which it is supposed to become possible to think the systematic unity of reason. We find here that the three transcendental ideas – the thinking subject, the world, and God – are not only in themselves the absolute major premise of a logical syllogism of conditions but also that these three highest ideas can, in turn, again be arranged into a syllogism. As Kant points out, this syllogism of the absolute unity of reason can be read in two directions – either with God as the conclusion or as the major premise.^[3] Irrespective of the direction, the world (or freedom) always takes the place of the middle term. And it is the middle term, in particular, that poses the central difficulty for a unity of the system of ideas. This unity is particularly precarious insofar as it contains within itself both the problem of an “empirical use” of the ideas of reason, which is developed in the last section of the chapter on the antinomies, as well as the problem of a “regulative use” of these ideas, which is developed in the appendix. It is possible to think a unity of syllogistic system, empirical use, and regulative use of the ideas only if one is willing to accept that the Critique of Pure Reason is a treatise on logic (and metaphysics), and not a treatise on epistemology.^[4] But even with the proviso that one accepts that what is at stake here is some form of logical unity, one can hardly fail to register the internal unevenness of this unity. This unevenness is due to the fact that such a unity is logical not in the sense of referring to what Kant calls a “merely formal, i.e. logical use, where reason abstracts from all content of cognition” but rather in the sense of referring to a “real” use of reason as transcendental faculty.^[5] Because reason, however, does not simply derive its principles from the senses or from understanding, the unity of the real use of reason cannot have objective validity. Rather, Kant famously speaks of “objective but indeterminate validity”.^[6] These difficulties leave room for discussion on how Kant can be read here. Perhaps recent approaches in contemporary philosophy can help open up a new perspective.

The hypothesis underlying this article is that Quentin Meillassoux’s discussion of an absolute necessity of contingency, as developed in After Finitude: An Essay on the Necessity of Contingency, might serve as a proposal allowing for a re-reading of the Transcendental Dialectic. ^[7] I think that this is possible due to important structural affinities between the two systematic approaches. For one, in both texts, questions of how to refer to one another the categories of relation and of modality occupy a central position. For another, both texts touch on ontological questions about ethics, that is to say, they lay an ontological groundwork with regard to constructing a systematic unity between theory and practice. Based on this hypothesis, my aim in this article is to show how the theorem of a necessity of contingency can be used to think about Kant’s resolution of the dialectical illusion of the antinomy of reason and, specifically, to discuss whether the necessity of contingency can be considered to be included within a systematic unity of reason or whether a split/contradiction must remain even after the resolution of the antinomy. Methodologically, the question of whether the idea of a unity, or of a principle, of speculative experience can be grounded on such a critical comparison of Meillassoux’s theorem to the Kantian text will entail drawing a line across the Transcendental Dialectic in its entirety, from the introduction and the first book, to the second book, and the appendix.

To this end, I will proceed in the following steps. In the first step, I will retrace the basic points of Meillassoux’s critique of “correlationism”, by means of which he reconfigures the divisions between metaphysics, physics, and ontology in view of gaining rational access to an absolute in-itself (2). In the second step, against the background of the Kantian concept of hope, I will discuss Kant’s and Meillassoux’s respective operations in referring a logic of modalities, which results from the resolution of the antinomies, through the operator of the beautiful to an ethics (3). In the third step, I will then critically ask in how far absolute contingency can serve to think the concept of a speculative unity of experience, that is to say, if the resolution of the antinomies means that, logically, we must remain at the level of a noncontradictory distributive unity or whether it is also possible to envision a form that allows for tying such a distributive unity of experience to a universal subjective synthesis (4).

2 “Correlationism” vs the necessity of contingency

Meillassoux’s philosophical operation can be generally understood as the attempt to reject and overcome the commonplace of post-critical philosophy that all possible knowledge of objects is inexorably caught within the boundaries of a “correlation” between subject and object. Any philosophy subscribing to the idea that the “correlation” is the necessary presupposition of possible knowledge is defined by Meillassoux as “correlationist”.^[8] For correlationist philosophers, thus, any pretension “to consider the realms of subjectivity and objectivity independently of one another”^[9] must always be false. Or, put differently again, correlationism’s fundamental tenet is that any in-itself can only ever be understood as-such-for-us. The majority of post-Kantian philosophies can be understood as adhering to some form of correlationism, regardless of possible differences in terminology or specific elaboration. In this regard, what Meillassoux calls correlationism functions as an overarching modus operandi – a transcendental level, if one likes, within which conditions of knowability can be further specified. For example, these may be put in terms of a Kantian distinction between things in themselves and appearances for us, in terms of the distinction in analytic philosophy between language and referent, or in terms of a phenomenological distinction between man and being.

By contrast, in Meillassoux’s own model, the subjective unity underlying objectivity as the necessary condition for the latter’s cognition is vastly surpassed by the reality of the world. In what is referred to as “speculative materialism”,^[10] it nonetheless becomes thinkable to venture beyond finite subjectivity and relative thought, to claim that it is possible to rationally demonstrate the possibility of knowledge of an absolute truth about the world.^[11] It is noteworthy that in the attempt to disprove the correlationist paradigm, correlationism is not simply rejected but turned against itself.^[12] In other words, the project of speculative materialism can be understood as a kind of inverted transcendentalism, or as a question about the “retro-transcendental”^[13] conditions of thought. In a demonstration of the necessary presuppositions for the argument that subject and object can only be known in their mutual co-implication, Meillassoux shows that the very possibility of this argument against the knowability of an absolute independent of its givenness must itself presuppose an absolute, namely, the mere facticity of the correlation.^[14] This then allows for showing that the logic of the correlationist paradigm itself is contingently given, that is to say, it can be proved that the only existing necessity is the necessity of an absolute contingency of everything.^[15]

In the next step, this finding serves as the hypothesis for answering a question that remains open in Kant’s transcendental philosophy: Meillassoux claims that Kant in the Critique of Pure Reason presupposes but does not prove that the thing in itself exists and that it is noncontradictory.^[16] However, based on the premise of absolute contingency, it becomes possible to enforce the postulate of the existence and noncontradictory being of the in-itself through a strict logical deduction of these two qualities.^[17] The absolutely contingent sphere of the in-itself can no longer be primarily described by logical laws, but rather can only be grasped by its mathematization. On the basis of the mathematical-real of the transfinite proven in the set-theoretical investigations of Cantor and Zermelo-Fraenkel, it is possible to go beyond the realm of potentialities indexed by totality and to think the real possibility of a sudden changeability of everything ex nihilo.^[18] This existent and noncontradictory, contingent absolute is associated with originary time, which is characterized as “virtual”^[19] and which emerges beyond any division of physics and metaphysics.

Although the question of division between metaphysics and physics has never been the primary topic of contemporary philosophy, it has nonetheless been prominently thematized, for instance, in Derrida’s deconstruction of the temporal synthesis of metaphysics. In “Ousia and Gramme”,^[20] Derrida sets out from Heidegger’s critique of the “vulgar”, i.e., “metaphysical” concept of time and then undertakes a reading of the Book IV of the Physics of Aristotle in order to develop an aporetic concept of time as the fundamental condition of (im-)possibility of metaphysical sense. Here, one should bring to mind that, when considered from a systematical perspective, the Physics provides the temporal synthesis on the basis of which it then becomes possible to ground the metaphysical order that Aristotle develops in the Metaphysics, the fundamental groundwork that remains structuring for Western metaphysics – as Derrida claims, even through Heidegger’s attempt of a “destruction” of it. Derrida’s equation of the aporetic impossibility of temporal synthesis with the impossibility of metaphysics indicates an inherent impossibility to refer the logical and the empirical to one another in a consistent manner. But in the terms of Kantian transcendental philosophy, empirical time cannot be assumed as ‘exterior’ to subjective temporal synthesis. ‘Outside’ of metaphysical time, it is here only possible to assume an ontological time, not an empirical time.^[21] In this regard, the hypothesis of an absolutely contingent reality is in fact closer to transcendental philosophy than to Derrida’s model: the absolute in-itself can be understood as a kind of extended temporality outside of possible synthesis, enabling the latter.^[22] Absolutely contingent being thus designates a realm prior to any aporetic relation of physics and metaphysics, from which an aporetic temporality first emerges. And Meillassoux believes that it is possible to relate to this absolute realm in a fully consistent manner, through logics and mathematics.

The reflections on the conditions of the possibility of statements about an “ancestral” reality can be seen as an example of such an operation. “Ancestrality” is defined as any reality that precedes the emergence of human consciousness, or indeed any form of life on earth at all.^[23] As discussed above, the aim here consists in proving the possibility of a positively determinable speculative knowledge about an absolute reality, independent of its givenness for a consciousness. The existence of ancestral material ‘objects’, which Meillassoux calls “arche-fossils”, functions as a “material support”, which seems to force us to adopt this perspective.^[24] The age of these entities, several billion years old, can be dated by the modern empirical sciences – for example, certain fossils whose age can be determined by the decay rate of their radioactive isotopes, or the light from distant stars that can be dated by the laws of thermoluminescence. But what is remarkable from a philosophical point of view about such datings is that they refer to objects which are so old that they already existed at a time when it was utterly impossible to conceive of them as an ‘object’ for a consciousness. The characteristic of these objects thus lies in their irreducibly material quality, and they are originally defined by the fact that they cannot be ‘objects for us’ but rather exist ‘by themselves’. Meillassoux asks how it is possible to ascribe truth to statements about such objects.^[25] His answer is that the arche-fossil points to the reality that every sort of relation to the world – irrespective of whether of thought, life, etc. – can be understood “as a fact inscribed in a temporality within which this relation is just one event among others, inscribed in an order of succession in which it is merely a stage, rather than an origin”.^[26] This means considering an absolute, cosmological time that precedes any transcendental form, a time that describes a realm independent of whether or not we, as thinking beings, exist or not, and that provides the conditions of possibility for a “lacunary givenness” to first emerge from a “lacuna of all givenness.”^[27] In this sense, ancestrality must be conceived not just strictly in terms of chronological antecedence but also in terms of logical precedence. Here, one can therefore note how Meillassoux’s critique of correlationism mirrors Kant’s ‘constructive critique’ of metaphysics in the Transcendental Dialectic.^[28] In his critique, Meillassoux relies on an extended temporality, that is, a cosmological, albeit absolutely contingent temporal synthesis, while at the same time undermining the very notion of the possibility of such a synthesis.

Finally, in order to be able to assert an absolute necessity of contingency, by which even a change of the laws of nature becomes possible, Meillassoux points out that we must also ask ourselves why these laws have not changed so far, or why they do not change all the time but rather exist as relatively stable constants, so that they are even generally declared to be necessary. Following the answer provided by the author, we should make sense of this experience by means of understanding that in the absolute realm of detotalized mathematics, other laws apply than those of probability. Hence, one cannot conclude from the mere fact that, as far as we know, the laws of nature have not changed so far, that they might not suddenly change after all.^[29] Absolute contingency thus denotes the possibility of a passing away of necessity itself. This possibility guaranteed by absolute contingency, however, “designates a pure possibility; one which may never be realized”.^[30] This amounts to saying the following: If contingency is the only absolute necessity, I must assume the possibility that everything could radically change at any moment. But this possibility is itself in no way necessary and rather remains contingent.

3 Hope and the beautiful: Ethics and the Transcendental Dialectic

Kant’s aim in the appendix of the Transcendental Dialectic is to develop how the thought of the highest formal unity of reason as a “purposive unity”^[31] can be brought about. What this means is that in order to expand our cognition of nature, we must presuppose the latter’s “systematic unity”^[32] as a guiding principle. The form of unity of reason which Kant had presented as the syllogistic system of the transcendental ideas in the first book of the Transcendental Dialectic is now put into practice, so to speak, by means of the “regulative use” of these ideas. This regulative use consists in applying the idea of God as a hypothetical major premise to the middle term of the world, i.e., to “every ordinance in the world as if it had sprouted from the intention of a highest reason”,^[33] thereby directing our processes of cognition through assuming a purposiveness of nature. It has been noted in the majority of the literature that the deduction of this form of systematic unity can be seen as problematic. In the appendix, Kant seems to directly contradict the claim he first made in the introduction^[34] and repeats again in the appendix,^[35] namely, that a deduction of the ideas of reason is impossible. The earlier claims notwithstanding, Kant now lays out a possible deduction of the ideas by means of distinguishing between constitutive and regulative principles and by showing that the transcendental ideas can be used as regulative principles of a systematic unity of empirical knowledge.^[36]

The problem critically noted in the literature is due to the fact that Kant, after the unfolding of the main part of the Transcendental Dialectic in the second book, in the appendix is faced with the task of having to justify a kind of turn from a critique of dialectical illusion to showing the possibility of a systematic unity of reason. If now, in the appendix, a deduction is actually to be possible, the question therefore arises how the elimination of the problem of transcendental illusion^[37] can be precisely understood, so that it then becomes possible for the ideas to function as regulative principles of a systematic unity.

The question of how to think the regulative use of the ideas must therefore be preceded on a preliminary stage by the resolution of the problem of the illusion that arises from the false empirical use of the ideas. In this sense, an understanding of the regulative use of the ideas presupposes a consideration of their empirical use, discussed in the main section of the Transcendental Dialectic.^[38] With respect to this empirical use of the ideas,^[39] however, Kant does not yet speak of deduction for the purpose of a systematic unity, but of the resolution of an antinomy for the purpose of a “distributive unity of the use of the understanding in experience”.^[40]

In this regard, the central merit of the questions raised by Meillassoux’s project of speculative materialism in general, as well as, in particular, of the proof he presents of the noncontradictory character of the thing in itself, lies in a call to re-examine the relationship between the resolution of the antinomy and the systematic unity of reason. On the one hand, as we will see, one can detect between Kant and Meillassoux a certain structural similarity as pertains to grounding ethical principles on a logical derivation that entails the resolution of a logical antinomy. But, on the other hand, in the resolution of the antinomy, Kant, strictly speaking, is no longer discussing the conditions of possibility of experience, but rather possibility in experience in the sense of an empirical use of the ideas. In this regard, there is a strong connection to the third Critique, which discusses the principle and conditions a priori of the faculty of the power of judgment, that is, the conditions of experience. However, before further discussing this point, in the following, I will first shortly look into how these questions are systematically discussed within the first Critique in the step from the Transcendental Dialectic to the Doctrine of Method.

Kant notes that the resolution of the antinomy with regard to the idea of the world as the objective synthesis of the series of appearances poses a particularly difficult problem. Unlike the only one-sided illusion that results from reason trying to think the absolute unity of the thinking subject’s synthesis, here the “natural and unavoidable illusion” of reason shows itself in the form of a “wholly natural antithetic”.^[41] This antithetic is due to the fact that, with regard to the world, the reason is not only confronted with its own inner conditions but also with the conditions that make it possible to think an external reality. Instead of relying on theological or cosmological models, Kant tackles this difficulty through a strictly logical operation, which in the resolution of the third antinomy amounts to a discussion of the antithetic between the concepts of nature and freedom. These concepts, that are of central importance for practical philosophy, are not derived conceptually but rather bound to the logical construction of the resolution of the antinomy, operating only with the categories of relation and modality, which Kant had introduced in the Analytic. Precisely because of this purely logical, not conceptual, determination, the relation of freedom and nature, which emerges in the resolution of the antinomy, is not determined but remains radically indeterminate. Only in the next step, i.e., Kant’s Doctrine of Method, are these determinations of logical relations made with respect to the distributive unity of experience taken up and, in the Canon, further processed in a metaphysical system by means of a logic of concepts.

The ethical intention of the appendix of the Transcendental Dialectic can be understood in relation to the second section of the Canon, titled “On the ideal of the highest good, as a determining ground of the ultimate end of pure reason” in Kant’s Doctrine of Method, which follows the Transcendental Dialectic. One finds here the famous summary, according to which the interest of reason in both theoretical and practical regard is united in the three questions of what I can know, what I should do, and what I may hope.^[42] These three questions, which clearly echo the syllogism of the three ideas of reason, can be seen as presenting the preparatory steps for what Kant in the Critique of Pure Reason considered possible with regards to developing his own system of metaphysics, post critique, so to speak. However, it has to be remarked that there is no simple progress from knowing to doing, and from doing to legitimate hope. Instead, Kant begins here from freedom as the middle term of the syllogism, claiming that the practical necessity of moral principles corresponds to a theoretical necessity “that everyone has cause to hope for happiness in the same measure as he has made himself worthy of it in his conduct”.^[43]

What I may hope is thus premised on applying reason in practice to the world as “the absolute unity of the series of conditions of appearance”,^[44] i.e., on the “real” use of reason in the field of hypothetical certainty. Hope, Kant points out, “is simultaneously practical and theoretical”.^[45] The practical aspect of this perspective ultimately leads “to the inference that something is (which determines the ultimate final end) because something ought to happen”.^[46] The theoretical aspect of this perspective concerns the issue “that something is (which acts as the supreme cause) because something does happen”.^[47] Kant defines this difference between the “ought-to” and the “is” as the difference between the worthiness to be happy, which can be grounded on the moral law and can thus be known a priori, and happiness, which can only be grounded on empirical principles. In their practical, that is, their moral use, the principles of pure reason have “objective reality”.^[48] They can function as principles of possible experience, understood in a straightforward common sense manner. For these principles of pure reason, condition “actions in conformity with moral precepts which could be encountered in the history of humankind”.^[49] One should note the hypothetical mode of possibility stressed here, once the principles of pure reason are not only restricted to the subjective necessity of the moral law but also extended into the objective, contingent space time of the “history of humankind.” While the emphasis in the typeset may at first perhaps seem insignificant, this focus placed on the modality of possibility points to a dense problem regarding a relation between logic, ethics, and aesthetics, which I only want to briefly outline.

When first introducing the transcendental ideas as totality of conditions in the first book of the Transcendental Dialectic, Kant states that it is not impossible for two modes of the modality of the possible to coincide: “what is possible in itself (internally), which is in fact the least one can say of an object” and “what is possible in all respects in every relation, which is again the most that I can say about the possibility of a thing”. However, for the vast majority of cases, these two modes of possibility are “infinitely far apart”.^[50] This passage can be read together with a passage in § 59 of the third Critique, where Kant discusses the relation of the power of judgment to the intelligible in the experience of the beautiful, that is, how the power of judgment gives the law to a “heteronomy of the laws of experience.”^[51] In this giving of the law to heteronomy, the power of judgment “sees itself, both on account of this inner possibility in the subject as well as on account of the outer possibility of a nature that corresponds to it, as related to something in the subject itself and outside of it, which is neither nature nor freedom, but which is connected with the ground of the latter, namely the supersensible, in which the theoretical faculty is combined with the practical, in a mutual and unknown way, to form a unity”.^[52]

The beautiful functions here as the mediation between inner possibility and outer possibility, between the theoretical and the practical, by forming “in a mutual and unknown way” a unity of the supersensible which, in turn, serves as the ground of practical freedom – i.e., the beautiful acts as a form of transcendental synthesis. In this function, the beautiful figures as “symbol of morality”.^[53] As such a symbol the beautiful is said to be able to mediate between what for the most part must be assumed as lying infinitely far apart, between inner and outer possibility.

Against this background, we can now turn to Meillassoux’s attempt to ground an “immanent ethics” on the ontology of absolute contingency, i.e., to articulate an immanent inscription of the ethical values of courage, goodness, or justice into the ontological order of being.^[54] Meillassoux has developed this idea of an immanent ethics in the published excerpts of his dissertation, titled L’inexistence divine.^[55] Just as in After Finitude, certain structural similarities to the Kantian system are striking here as well. Specifically, one can find the same two relevant systematic operators at work as in Kant: hope and the beautiful.^[56] The beautiful is conceived of as the “phenomenal emergence without reason of an accord between reason and the real”, that is, as a new, contingent, “factial” symbol.^[57] Hope, in turn, circumscribes the desire of every human being as a rational being and is defined as “desire crossed by thought”.^[58] However, here we do not find the beautiful as that which connects inner and outer possibility, but rather possibility as that which connects the beautiful and hope. Meillassoux’s concept of the beautiful comprises the contingent emergence of a “non-necessary encounter of phenomenal mechanisms and our rational ends”,^[59] a coincidence of reason and reality.

Ultimately, though, beauty can only occur through “the conjunction between being and act”.^[60] But this is not the same as saying that it could thereby be brought about, or caused. Not unlike the Kantian distinction between the worthiness to be happy and happiness, Meillassoux thus introduces a difference between condition and cause.^[61] Thought’s knowledge of contingency is crossed with desire, belonging to practical reason, in order to think of the possibility of a completely contingent emergence of the correspondence between reality and reason as the emergence of the beautiful, understood as a just world. Although this attempt to ground an immanent ethics in many respects recalls the transcendental principle of the third Critique, purposiveness as the principle of the beautiful, Meillassoux’s version differs from Kant in the respect that the beauty of the (just, beautiful) world is not retraced as if the intelligence of a divine were its cause. Instead, we are told that a base for the beautiful world of justice must rather be sought in the hope of just people in the past, who through their hope enabled the possibility of the contingent appearance of this world as if hope were its source.^[62] This marks a difference to Kant’s system: “None the less, the beauty of the fourth World [i.e., the just world] would not be strictly identifiable with the transcendental beautiful”.^[63] This is noteworthy insofar as tying hope – i.e., desire crossed by thought – to the concept of the beautiful – i.e., the contingent emergence of the ultimate order of justice – through the principle of purposiveness structurally amounts to conceiving of such a principle not as a transcendental principle but rather in the sense of a regulative principle. As a regulative principle, it is related to practical reason, which here circumscribes the realm of what we may hope for, in order to think the possible congruence between reason and reality.

The ramifications of this can be understood if we recall what was discussed above with regard to the steps inherent in the Kantian systematic of the first Critique, where the regulative principle only comes into play after the transcendental principle, i.e., the regulative use only after the empirical use of the transcendental ideas. Questions regarding happiness follow questions regarding the worthiness to be happy. Meillassoux, as we have seen, adopts the difference between condition and cause. But he then claims that “the ethical mistake par excellence consists in founding the end in the origin”.^[64] The problem he sees in such an order is that it thereby becomes impossible to think “that the contingent thought of the eternal surpasses the eternal, that the human being surpasses the being of beings”.^[65] For this reason, Meillassoux introduces the thought of a God who only enters at the very end.^[66] This God is understood as both an incarnation of the universal and a “human mediator” who after having resurrected the dead as the advent of ultimate justice abolishes His own absolute power in order to become a human child, and thereby equal to everyone else.^[67] What this amounts to is directly identifying the beautiful with the regulative idea. Through the collapsing of the transcendental into the regulative, and from there into the systematic, i.e., through the shifting from an “as if a divine will were the cause” to an “as if the hope of just people were the cause”, the discourse of the antinomy between freedom and nature disappears. As a consequence of this omission, it is hardly possible to avoid implicitly assuming an identity between nature and freedom – an identity that Kant does not presuppose in the Transcendental Dialectic despite the resolution of the antinomy. I will further discuss this point in the next section. Here, it is sufficient to note that presupposing such an implicit identity between nature and freedom equals effecting a shift of proportion in how to think an indeterminate relation between desire and rationality.

The basic logic of the beautiful, both in Kant and Meillassoux, consists in a congruence of freedom and nature, or of ethical harmony between my freedom and the freedom of others. In this sense, the beautiful fulfills an important ethical dimension in both transcendental philosophy and the system of the necessity of contingency. But the immanent ethics of a necessity of contingency envisions a form of transfer between hope and justice, even if hope is seen as just the condition, not as the contingent, absolute cause for justice. What is missing here entirely is the necessary determination of the relation between nature and freedom. When compared to Kant, this difference is all the more relevant because the outer logic of the systematic framework seems to function in a quite similar way to the steps taken in the first Critique. However, I would argue that it is problematical to assume a function of hope as the carrier that allows for passing from inner possibility to inner necessity and then to outer possibility.

4 Contingency, necessity, and the unity of experience

As I have indicated above, with the intent of a critique of the model of the necessity of contingency, I will now discuss in which way the problem of absolute contingency arises in the antinomy chapter, before the concept of a regulative use of the ideas is introduced. The issue of the absolute necessity of contingency is of major interest for the resolution of the antinomies in the final section of the antinomy chapter, titled “On the empirical use of the regulative principle of reason, in regard to all cosmological ideas”.^[68] This section ends with Kant stating that the “unconditionally necessary” and “unbounded contingency” are in no way opposed to each other.^[69] Kant devotes a separate analysis to absolute necessity in the second Critique, where he examines the subject in relation to the moral law. But in the first Critique, absolute necessity is considered instead within the framework laid out in the first book of the Transcendental Dialectic,^[70] i.e., within the parameters of a strictly logical constitution of absolute conditions, that are here referred to human action. In section nine, Kant ties the category of possibility to the power of choice (Willkür), which he characterizes as sensible or “pathologically affected”, but not “pathologically necessitated”. That is to say, although human beings cannot escape sensible impulses, it is nonetheless possible that they determine their actions independently of these impulses.^[71]

Kant states that it is only through the transcendental idea of freedom that it becomes possible to ground the concept of practical freedom.^[72] And although it is perhaps not easy to understand how to keep the empirical power of choice apart from the subject that is determined by its inner necessary conditions, one should take care to differentiate here between three levels: the transcendental idea of freedom, the power of choice, and the will. The latter two pertain to practical freedom and should be distinguished from each other. Depending on how the power of choice determines its own maxim, it can either act according to the free will or not.^[73] In the resolution of the third antinomy, on the other hand, Kant is concerned with a more fundamental principle that presents the (transcendental) conditions of practical freedom. While the pure moral law of which Kant speaks in the second Critique is predicated on its universal form alone and devoid of any determinate content, one can say that the entire content of the form of the freedom of pure practical reason is developed in this concluding section of the antinomy chapter. The form of reason’s practical freedom has to be thought in a twofold way: on the one hand, as the faculty of positing a nondetermined beginning, and, on the other hand, as the faculty of acting in the phenomenal world as a consistent constituent within a chain of causally determined links. Hence, in order for an empirical subject to instantiate this form, it must be accorded a twofold character. One part of this consists in the subject’s “intelligible character”,^[74] which only functions as the cause of the subject’s actions as appearances but to which the conditions of sensibility do not apply. This noumenal component of the subject has often been directly tied to practical reason since that is clearly toward where it is ultimately oriented.^[75] The other part of the twofold subjectivity consists in the subject’s “empirical character”^[76] through which the subject’s actions as well as the (empirical) subject itself can be derived as appearances from a relation to other appearances following causality as the law of nature.

The idea of freedom can therefore not be fully understood if we restrict its meaning to actions guided by practical reason. The empirical use of the regulative ideas is first of all directed at the determination of the links between appearances. By thinking and speaking these links and by thinking and speaking about these links, we are able to think the idea of freedom – which is why Kant says that reason is determining, but, in regard to how it links appearances, not determinable.^[77] This also means that reason cannot be directly stated as a cause, but structurally remains located outside the chain of appearances. The question of why reason has not determined itself otherwise is not appropriate when we try to think freedom. According to Kant, the correct question is only, “Why has it not determined appearances otherwise through its causality?”^[78] This second form of the question aims at the fact that the regulative idea of freedom can be applied here only in the relatively limited function of a critique of natural necessity, in the sense of a negation that judges the past as a condition of possible freedom. There is no direct practical purpose involved (what should I do?), but rather a critical question or attitude directed toward the causality of the link between appearances. Kant’s argument for why it is not possible to answer conclusively this critical question about the possibility of a different determination of these appearances by the causality of reason in the past is compelling: the question is pointless because it overlooks the tautology that “another intelligible character would have given another empirical one”.^[79] Although this is not yet the domain of pure practical reason, the critical question regarding freedom nonetheless serves a practical purpose. The critical cut into the formed link of appearances allows for calling into question a certain state of affairs that cannot legitimately lay claim to being normatively justified by the law. Thus, the relation between freedom and nature as laid out by Kant in the resolution of the antinomy is critical but not yet in itself practical. Through the uneven relation of appearances to nature and freedom, it rather becomes possible to understand the meaning of “objective but indeterminate validity”.

One can see why the Transcendental Dialectic remains such a contested passage when recalling Hegel’s critique of it. For example, in the introduction to the Science of Logic, Hegel praises the general idea of the Transcendental Dialectic, Kant’s emphasis on “the objectivity of reflective shine and the necessity of the contradiction which belongs to the nature of thought determinations”.^[80] However, Kant’s resolution of the antinomy as the way to relate appearances and things in themselves to each other after their strict separation is criticized by Hegel. In the logical unity of the philosophy of absolute spirit, things in themselves and appearances are immanently related to each other. Contradiction occurs when reason is related to things in themselves. But insofar as logic can recognize the contradiction as something other lying within itself, reason is able to give a unity that encompasses and includes the contradiction. “This result, grasped in its positive aspect, is nothing else but the inner negativity of the determinations”.^[81] For Hegel, the dialectic, consistently thought through to its end, is self-determination of reason – something Kant, as we have seen, explicitly denies. From the point of view of the Hegelian logic, the Kantian resolution of the antinomy of nature and freedom appears inconsistent or even inconclusive.^[82]

In the context of this broader philosophical lineage, Meillassoux’s account of the absolute necessity of contingency can be regarded as a decidedly anti-Hegelian attempt at resolving the antinomies of pure reason. Absolutely contingent becoming, i.e. the real possibility of change of everything, even of the laws of nature, is opposed by Meillassoux to a form of thought that conceives of becoming in terms of real contradiction.^[83] But despite the fact that the model of the necessity of contingency sidesteps the internal negativity of the Hegelian dialectic, it does not suffice to grasp the entire structure of the Kantian resolution of the antinomies. The brief outline sketched above may serve to provide the context within which one can understand where the Kantian resolution of antinomy differs from the Hegelian resolution of dialectical contradictions. In the question of the empirical use of the regulative ideas, Kant speaks of a unity only insofar as it concerns principles of a real experience. Such a unity is not “merely logical”.^[84] Rather, in the “real use” of reason, the unity of reason is constituted through its very use.^[85] Kant grounds this unity on a “dynamical” synthesis of appearances, which allows for thinking the synthesis of grounds of determinations as “things not homogeneous” [Synthesis des Ungleichartigen].^[86] As such, the dynamical synthesis of experience can be superimposed on the “mathematical” synthesis, which synthesizes “homogenous things”,^[87] without the former type of synthesis excluding the latter. In this regard, the question whether a contradiction-free resolution of the antinomy therefore also entails that it is not necessary to think any contradiction cannot be answered generally. The resolution of the antinomy means constituting a synthesis of different determinants, and this will sometimes lead to a contradiction while at other times it will not. We can only say so much that the resolution of the antinomy of freedom and nature contains a constant tension and even collision between the dynamical and mathematical forms of synthesis as principles of our explanation of the relation of appearances.

So, this collision of principles then allows for considering the connection between the first Critique and the third Critique not just with regard to the ties between hope and the beautiful but also with regard to the ties between freedom and the sublime. Seeing that Kant in the Transcendental Dialectic not only speaks of the empirical use of the regulative principle but that he also deduces freedom as the principle through which it only first becomes possible to ground the difference between the dynamical and mathematical use of the syntheses of the concepts of understanding “under the condition of empirical thinking in an experience”,^[88] mentioned already in the Analytic, we should not hesitate to ask how the resolution of the antinomies of the Transcendental Dialectic has to be tied back to the faculty of understanding. Kant claims that the use of our faculty of understanding as determined by the ideas of reason means that “it will be guided better and further” in its cognition.^[89] In this regard, it may then at first glance seem surprising that in the third Critique, Kant discusses the mathematical sublime^[90] and the dynamical sublime^[91] as two modes of failure of the synthesis of the imagination. In contrast to the perfect auto-affection of a mind capable of intellectual intuition (intellectus archetypus) that would allow for a completely successful synthesis, in finite beings, synthesis must necessarily fail, thereby giving rise to the experience of the two types of the sublime.^[92] While this might seem to be at odds with “being guided better and further” in our cognition, after what has been discussed above regarding the meaning of objective but indeterminate validity, it is no longer inconceivable how a failure of complete synthesis and an increase in knowledge can go hand in hand. For what follows from this incompleteness of our synthesis of understanding is a contingency of necessity itself. By no means should this be understood as indicating that gaps in our knowledge result from our inability to reach some sort of deeper, substantial reality of an essentially unknowable thing, quite the contrary. Precisely through signaling missing knowledge, these gaps are productive. In this sense, the gaps in our knowledge do not point to unknowable things, but instead to things we do not (yet) know.^[93]

There is thus a certain incompleteness to the model of the necessity of contingency. This consists in the fact that although real contradiction is removed in order to think absolute becoming, what is not sufficiently taken into account is that we can never think the realization of change in the real without an unevenness, or even collision, in the dynamic synthesis of experience. In this sense, the important question of the necessity of contingency, taken up by Meillassoux, should perhaps be expanded to ask if Kantian “unbounded contingency”, which results after the resolution of the antinomy, might not have to be considered as being split internally.

Accordingly, it would then be possible to distinguish between two forms of contingency: The first form of contingency, the necessity of contingency, functions adhering to the logic of universality and constitutive exception – everything is contingent, except for contingency itself, which is necessary.^[94] In Kantian terms, we are here speaking of God, who follows from a mathematically unconditioned unity of the world, that is, God figuring as inexistent conclusion of the syllogism of the three transcendental ideas, presented in its inductive form. By contrast, the second type of contingency is the contingency of necessity, contingency turned back onto itself. This form, in turn, corresponds to God as the non-existent major premise of the syllogism in its deductive form – that is, God, not as inexistent necessity outside of (or at the end of) an immanent world, but rather as existing logical modality that is impossible to prove in nature.

5 Conclusion

In this article, I have attempted to read Meillassoux’s speculative materialism as an interpretation of Kant’s Transcendental Dialectic in order to discuss a possible route one might take to reinterpret the meaning of speculative thought. I have tried to show that the theorem of a necessity of contingency contains the crux of the relation between the empirical use and the regulative use of reason.

To briefly summarize, first, it was discussed whether an opening of the boundaries of metaphysics in the necessity of contingency should be understood in the sense of a move toward a physics or toward an ontology of absolute being. Insofar as Kant himself understands his transcendental philosophy as ontology, the fact that Meillassoux opts for the latter places him firmly on Kantian ground. The second step consisted in an analysis of how the link between the empirical use of reason and the regulative use of reason must be understood through the operator of the beautiful. Pointing out the difference between the principle of purposiveness as bound to the regulative ideas and as transcendental principle, respectively, it was shown that although a conception of the beautiful order of justice derived from the principle of purposiveness can be made to conform with the absolute contingency of being, such a model does not yet suffice to think how the content of rational desire might be brought forth. Against this background, the third step therefore consisted in turning back once more from the regulative use to the empirical use of reason so as to analyze in more detail Kant’s resolution of the antinomy between freedom and nature. It was shown that the transcendental idea of freedom allows for resolving the antinomy without thereby relinquishing the central importance of contradiction itself. Consequently, it was possible to point out a split in Kantian “unbounded contingency” into the necessity of contingency and the contingency of necessity.

As a result, we can note that Meillassoux’s theorem of absolute contingency can serve as a productive perspective through which it becomes conceivable how one might interpret the Transcendental Dialectic in terms of a speculative unity of experience. But at the same time, it also became apparent that speculative materialism may have to be supplemented by a question regarding an additional synthesis. The necessity of contingency allows for opening up this question, but it does not allow for conclusively answering it. We can also note that the fact that Kant designed the Transcendental Dialectic as an indispensable means of critical philosophy to resolve the transcendental illusion does not mean that this can only be understood in the limited frame of a critique of dogmatic metaphysics, as has often been the case in the reception of Kant’s works. Rather, I want to claim that the Transcendental Dialectic can be read as containing in itself a logic of an unfolding of a synthesis that has to be presupposed in experience. What remains to be further discussed might be in which sense one can actually refer to this dynamical synthesis as a unity, strictly speaking. In any case, what I have here provisionally referred to as speculative (and not dogmatically restricted “empirical”) unity of experience should make one thing clear: If a reading of the Transcendental Dialectic like the one proposed could be sustained, this would then maybe even allow for beginning to lay out a counter-objection to the philosophically most profound objection to Kant’s system, as put forward in Hegel’s critique of Kant. Instead of referring speculative experience to the absolute idea as internally divided unity of opposed determinations of thought, the speculative unity of experience could be understood as a subjective synthesis of objective indeterminacy.