An Assessment of Kant’s Transcendental Solution to the Third Antinomy

The preface of the second edition of Kant’s Critique of Pure Reason contains what I understand to be the most appropriate characterization of his greater philosophical project: he hopes to “carry out [a] plan of providing the metaphysics both of nature and of morals,” and ultimately to produce a “durable” theory of metaphysics (B xliii, B xliv). A durable metaphysical theory seems to be one that provides conclusions, or at least considerable insight, to the greatest number of metaphysical quandaries. Kant’s original starting point, or his own Copernican Revolution, is his suggestion that in order to “get farther with the problems of metaphysics,” we should “assume that objects must conform to our cognition,” a move that allows him to propose a priori structures that necessarily pertain to our experience of the world and that makes it possible to assert meaningful (non-tautological) propositions that can be necessarily true, or synthetic a priori truths (B xvi). Kant also attempts to maintain a certain epistemic humility through his assertion of transcendental idealism, which proposes that the objects of human experience are simply appearances which do not share the same predicates as the objects of fundamental reality, or things-in-themselves.” It is this sense of idealism which Kant relies upon to make sense of certain “antinomies,” or seemingly unresolvable problems “into which reason falls of itself and even unavoidably” (A407/B434). Kant’s Third Antinomy concerns a philosophical problem as old as Aristotle, the “unmoved mover” or “initial cause.”

Each of Kant’s antimonies concern a set of two contradictory propositions that arise from the metaphysical endeavor, namely a thesis and an antithesis, which together form a “dialetical theorem of pure reason” (A421/B449). Such a dialectic, according to Kant, “is one that every human reason must necessarily come up against in the course of its progress” (A422/B449). In the case of the Third Antinomy, this species of dialectic arises from the empirical understanding that all events are conditioned by a previous event, or that every event has a cause. This understanding is the basis of Kant’s concept of “causality … according to nature” (A532/B560). The opposite of this understanding, or Kant’s concept of “causality … from freedom,” allows that some events are unconditioned (A532/B560). In this paper I intend to explain Kant’s argument for both the causality of nature and the causality of freedom, elaborate his transcendental solution to this antinomy, especially as it relates to the problem of free will, and discuss what I understand to be unsatisfactory or problematic elements of both this application of transcendental idealism and of such an idealism in general.

I find it appropriate to begin by elaborating the antithesis of Kant’s Third Antinomy, which suggests “everything in the world happens solely in accordance with laws of nature” (A445/B573). While this concept is familiar to anyone who has considered causal determinism or even basic physics, it will help to frame it within Kant’s specific understanding of the a priori laws which govern human experience. In the Transcendental Aesthetic, Kant suggests that every object of human experience, or every “appearance,” is structured by a spatiotemporal form. In this sense, every experienced object or event exists within a temporal sequence, such that empirical events can and do precede (or succeed) others. This temporal nature makes it possible for a traditional understanding of causality, one in which all events are preceded by causes which necessitate them according to a rule, to apply to empirical experience. Kant suggests this phenomenon is not only possible but necessary — through “[natural causality] alone a thoroughly connected experience is possible” (A448/B476). To elaborate this conception of natural causality, it will be helpful to consider Kant’s suggestion that “the whole elapsed past time is thought of as given necessarily as the condition for the given moment” (A412/B439). In this sense, every event is understood as conditioned by the entire chain of causal events that preceded it.

While this picture of an entirely conditioned chain of events is certainly aligned with both a simple understanding of natural laws and with Kant’s own formally structured empirical reality, it seems to rely on an infinite regress that begs to be explained. The proof for the thesis of the Third Antinomy concerns this problem:

If … everything happens according to mere laws of nature, then at every time there is only a subordinate but never a first beginning, and thus no completeness of the series on the side of the causes descending one from another. But now the law of nature consists just in this, that nothing happens without a cause sufficiently determined a priori (A446/B474).

The thesis of the third antinomy ultimately suggests that natural causality, when understood as the only possible causality, is self-contradictory, such that another kind of causality is necessary in order for natural causality to have the nature that science (and Kant) have ascribed to it. Kant proposes this causality is “an absolute causal spontaneity beginning from itself,” or the causality of freedom (A446/B447). In addition, as part of this same thesis, Kant is also careful to assert that “here we are talking of an absolute beginning not as far as time is concerned but as far as causality is concerned,” a clarification which reiterates a certain separation between a free cause and the necessary spatiotemporal structures of the empirical world (A450/B487).

It is now appropriate to discuss how Kant’s transcendental idealism can be understood to offer a resolution to the paradox inherent in this contradictory set of propositions. As mentioned earlier, transcendental idealism concerns a distinction between spatiotemporal appearances, which constitute the empirical experience that is understood to obey natural causation in accordance with certain laws, and things-in-themselves, which are understood as specifically non-spatiotemporal. The most basic requirement of the thesis of the Third Antinomy is that the chain of natural causes can be attributed to at least one unconditioned cause, and since transcendental idealism allows for the existence of an non-spatiotemporal entity not bound by the causality of nature, it seems this requirement is satisfied. This understanding is in many ways parallel to a conception of an unmoved mover, found in both Classical philosophy from Aristotle and Medieval philosophy from Aquinas. Kant’s precise solution, assuming transcendental idealism, asserts an intelligible, non-intuitive, and therefore non-empirical, but also free (spontaneous) cause, which exists outside the series of natural causation but can nevertheless have effects within the law-governed series, of the empirical, experienced world.

When suggested in regards to an object which is understood as necessary only to save natural causality from infinite regress, and is therefore only a condition of possible experience instead of something that also corresponds (in some manner) to specific appearances in time, Kant’s appeal to a thing-in-itself (or to transcendental idealism) seems like a necessary step in the resolution of this paradox. However, Kant’s resolution to the Third Antinomy contains an additional claim, one that concerns a certain species of human freedom, which suggests that many objects, namely humans, have “a faculty of determining oneself from oneself, independently of necessitation by sensible impulses” (A534/B562). With this, his argument expands in a direction obviously oriented towards an account of free will.

During his discussion of a transcendental resolution to the Third Antinomy, Kant elaborates further on the relationship between appearances and things-in-themselves:

… since these appearances, because they are not things in themselves, must be grounded in a transcendental object determining them as mere representations, nothing hinders us from ascribing to this transcendental object, apart from the property through which it appears, also another causality that is not appearance, even though its effect is encountered in appearance (A538-539/B566-577).

Here we encounter a more determinate understanding of the transcendental object, or more specifically, a distinction between this object as an appearance and as a thing-in-itself, where the former contributes an “empirical character,” which performs in the empirical world in “accordance with natural laws” in the “single series of the natural order,” while the latter contributes an “intelligible character,” which is essentially free and spontaneous, outside the temporal, law-determined series (A539/B567). The empirical character of an event, as part of the unified experienced given to our minds, is always entirely predictable given enough knowledge of preceding events and natural law — to our intuition, every event is sufficiently determined. Nevertheless, Kant asserts an intelligible character, understood as rational, which is both a “faculty through which the sensible condition of an empirical series of effects first begins,” and “the persisting condition of all voluntary actions under which the human beings appears” (A552-553/B580-581). It is essential to a recreation of this argument, however, to acknowledge the intelligible character of the transcendental object, or Kant’s thing-it-itself, can “never be known immediately” — because it is necessarily non-empirical, it is shrouded in mystery (A540/B568).

Kant’s illustration of free will suggests that while a human action, as something empirical, is always attributable to natural causality, the same action can also be considered free — it can also be attributed to a transcendental object acting outside the series of temporal events. This understanding raises a concern for transcendental idealism surrounding the nature of the relationship between an appearance and the transcendental object which Kant suggests serves as its ground, but also, in cases of causality from freedom, as its cause. This concern is understood in the literature as the problem of affection. In a chapter on this topic, albeit one primarily oriented towards an understanding of the transcendental objects as a ground of appearances and not a cause of them, Henry Allison ultimately suggests that the difficulty in understanding this relationship arises from a tendency to want to understand Kant’s transcendental idealism in transcendentally real terms, or from the fallacious assumption that “Kant owes us, yet cannot provide, some ultimate metaphysical story about affection: a God’s-eye account of what it is that really supplies the matter of cognition” (73). According to Allison, we are instead supposed to understand transcendental idealism as “as idealism of epistemic conditions” (71). On Allison’s behalf, in the context of his explanation of human freedom, Kant is careful to clarify that he is not attempting to establish the “reality of freedom, as a faculty that contains the causes of appearance in our world of sense,” but rather that he only wants establish freedom as a “transcendental idea” (A558/B586).

While Allison’s interpretation of the problem may be entirely appropriate, I would argue that in light of Kant’s own characterization of his project — one which will provide the metaphysics of both nature and morals — it seems a reorientation towards epistemic concerns leaves us unsatisfied. If Kant is to provide a durable metaphysics, one that includes a metaphysical solution to the problem of free will, and not just an epistemic one, it seems he indeed owes us an explanation that accounts for the nature of a causal relationship between an ideal entity and a real one. Descartes, who might be understood as the father of the traditional idealism or mind-body dualism which Kant’s argument is certainly aware of, was posed a similar question concerning mind-body interaction in a letter from Princess Elisabeth. However, Kant might be understood to escape this problem because his transcendental idealism is drawn along different lines than Descartes’ mind-body substance dualism. Perhaps Kant is not responsible for providing a clear response to this problem specifically because he maintains a strict lack of insight into the nature of transcendental objects, or things-in-themselves. Nevertheless, if we expect to look to Kant for a durable metaphysical theory, it seems appropriate that a causality to which we can attribute empirical events, even if they can also be attributed to natural causes, should be better explained, but it seems no explanation is offered.

In order to be charitable to Kant, it is important to acknowledge his exploration of what he refers to as the “practical” consequences of transcendental idealism in the same second edition preface quoted earlier, consequences which he understands as metaphysically significant. Kant suggests that by explicitly narrowing the extent of sensibility to appearances, what initially appears to lead only a negative understanding of things in themselves as non-spatiotemporal in fact also yields a positive understanding of the “practical use of reason” (Bxxv). By reason’s practical use, he means its role in morality. Here it is fitting to consider what is really at stake in any discussion of free will — moral responsibility — and why I consider accounts like Kant’s to be problematic.

It is appropriate that a concern with moral responsibility is usually discussed in metaphysical terms, because it seems the human preoccupation with moral responsibility is, in many contexts, traditionally linked to religious understandings of divine judgement, in which an individual’s metaphysical, or, in this case, transcendental, ability to direct their own actions is essential to the divine appraisal of character necessary for a dualistic conception of the afterlife. Outside of this context, however, an understanding of moral responsibility becomes more nuanced, because it ultimately determines the way we think about and act towards instances of moral transgression. Kant’s conception of free will seems to suggest that while its exact nature, means of affection, and even possibility are unknowable, its status as a transcendental idea is enough to secure a belief in moral responsibility. While this seems like a desirable end, I’d like to suggest that our seemingly dogmatic obsession with salvaging moral responsibility within an entirely conditioned empirical world is actually impractical, in the commonplace understanding of the term. If a practical concern, within the same commonplace understanding, can be understood as one that is concerned with achieving a desired end given the parameters of a situation (it is practical to carry an umbrella when it’s about to rain, so as to avoid getting wet), and morality can be understood as a means to an ideal end in which “each can carry on his own affairs in peace and safety,” it seems apparent that practical concerns should be especially responsive to the real empirical conditions of the world (Bxxv). When Kant suggests a certain transcendental freedom can ultimately influence a world which he believes is, in an empirical sense, necessarily and entirely determined, he ignores the extent to which an emphasis on certain transcendent ideas in relation to morality might actually deemphasize the realization of a practical moral understanding. By a practical moral understanding, I mean one that is interesting in achieving moral ends. When we acknowledge that an individual’s “malicious lie” is empirically the result of that individuals “bad upbringing” or “bad company,” but choose to understand this action as “entirely unconditioned in regard to the previous state,” it seems that our choice should be understood as impractical, or irrational if it is rational to advance own our moral ends. My suggestion is that the deemphasis of empirical causes within our moral understanding might be considered counterproductive, because it prevents us from focusing on the effective causes of bad or malicious behavior. While I acknowledge that an understanding of freedom is still important to a realization of moral ends, I do find it problematic that compatibilist theories of freedom, a grouping in which I believe Kant’s theory belongs, are irresponsive to this dilemma.

I also find that a similar critique is applicable to Kant’s transcendental idealism in general. While a separation between appearances and things-in-themselves seems like a move towards an epistemic humility, it has certain significant consequences for our worldview. Kant’s suggestion that certain rational, a priori structures given by the human mind are the source of a spatiotemporal reality seems to imply the consequence that spatiotemporal reality is contingent on our existence. While I certainly cannot claim a thorough understanding of the nuanced literature surrounding Kant’s position on the reality of spatiotemporal appearances outside of human experience or existence, it seems appropriate to suggest that a reorientation of our metaphysical worldview around our own minds can be considered explicit not humble, but rather exceedingly assuming and anthropocentric. The same anthropocentrism is furthered by Kant’s suggestion, made in his discussion of the Third Antinomy, that humans are transcendentally unique in the sense that they can assume “independence of the power of choice from necessitation by impulses of sensibility,” an independence which is explicitly not granted to animals or other forms of life, and one that secures a certain one-to-one relationship with a thing-in-itself, which Kant’s greater position seems to suggest is somehow metaphysically superior. It will be helpful to explain why I find this understanding problematic. By granting metaphysical significance to a boundary drawn along the lines of reason, we become comfortable in limiting our moral concerns along the same lines, such that individuals who can be portrayed as less than rational, such as mentally-handicapped humans, are apt to fall short of meriting the moral rights we extend to rational beings. In historical examples of slavery and genocide, the same less-than-rational characterizations are made of our victims. My suggestion is that a genuine concern with morality in the empirical world is potentially incompatible with the acceptance, especially on a metaphysical level, of discrete, hierarchical boundaries drawn between types of living things, as this presents a slippery slope away from our moral ends.

I acknowledge that these arguments do not enter into a dialogue with Kant’s remarkably rigorous framework in a manner that is remotely comparable to his own rigor. With this acknowledgement, I find it important that philosophy at least occasionally concern itself with the extra-disciplinary implications of its most influential works, and it is my hope that more rigorous attention is paid to these matters in philosophy’s future. Philosophy’s project, as I understand it, must always involve the maintenance of humility before a world that is unimaginably greater than the individual human subject who attempts to understand it.

Works Cited

Allison, Henry. Kant’s Transcendental Idealism. Yale University Press, 2004. Web.

Kant, Immanuel. Critique of Pure Reason. Edited by Paul Guyer and Allen W Wood, Cambridge Univ. Press, 2009. Print.