Immanuel Kant’s Critique of Pure Reason and the critical philosophy it created generated a storm of responses, ranging from the near-deifying to the dismissively critical. In 1792, Gottlob Schulze anonymously published Aenesidimus in response to both Kant’s original text and an interpretation of it by Karl Reinhold. Aenesidimus was an attack from the side of the Humean sceptics, claiming that Kant and Reinhold’s attempt to move past Hume’s scepticism begs the question. Although based on a misunderstanding of several of critical philosophy’s key points[1], it was highly influential at the time in turning the public tide against critical philosophy (Di Giovanni 24-25). Kant’s main thesis is as to the existence and nature of a priori concepts, or (as Schulze often calls them) necessary synthetic concepts. These have their ground in the mind, are applicable only to experiential cognitions, and provide the form for all our experience (Schulze 136). Unfortunately, Schulze says that these judgements, which attempt to resolve Hume’s sceptical doubt, claim to do so by simply assuming exactly what he was doubting to begin with and what Kant himself claims is impossible(Schulze 132-133): that we can have knowledge of things outside our experience, such as things-in-themselves.
In particular, Schulze claims that there are two related assumptions made by the critical philosophers that were taken as objects of scepticism by Hume. The first is the assumption that the principle of sufficient reason extends to things-in-themselves as opposed to simply our representations; the second is the assumption that we can infer anything about the nature and composition of objects-in-themselves from our representations of them (Schulze 101, 132-133). According to Schulze, both of these questions are begged multiple times by Kant in the exposition of several of Kant’s points regarding necessary synthetic judgments. Furthermore, Schulze finds Kant contradicting his own philosophical principles at times, for Kant similarly claims that we cannot have any knowledge of the thing-in-itself and that the categories of cause and effect can only be applied to objects obtained through the senses—yet he proceeds in depending on knowledge of the thing-in-itself and applying the laws of causality to things-in-themselves (Schulze 160).
In order to understand Schulze’s argument, it is first necessary to explain the principles of Hume’s scepticism and how Kant intends to refute them. Di Giovannni sums up Hume’s basic principle as follows: “the only distinction in intuition is between the subject and its representation” (24). All consciousness is assumed to be present only within ourselves, and there is currently no scientific or logical reasoning that allows us to make any definitive claims about anything outside of experience, such as a thing-in-itself (Di Giovanni 20-21)[2]. Related are his arguments on cause and effect. If the principles of cause and effect hold absolutely and come from our human reason, Hume argued, we should be able to understand how this necessity arrives a priori. This does not seem to be the case, however. As Schulze puts it, “it is quite impossible to see how, just because something is, something else must be also necessarily” (135). Therefore our experience of the concept of cause and effect is based merely on our experiencing these things together and associated on a consistent basis and assuming from that that they must be necessarily and objectively linked (Schulze 135). Our sense that synthetic judgments have necessity, according to Hume, is due merely to a process of induction and not due to their a priority. Kant, on the other hand, claimed to have shown that a priori synthetic concepts—of which cause and effect is but one example—exist and provide this necessity due to having their ground and cause in the mind (Schulze 137). Schulze claims, however, that these arguments cannot refute Hume, as they do not actually argue against his points—they merely assume that his doubts as to the nature of cause and effect and our knowledge of things-in-themselves have already been satisfied and continue from there. Furthermore, Kant seems to ignore his own assumptions in the construction of many of his arguments.
Schulze begins his argument by claiming that the entire concept of necessary synthetic judgments assumes the nature of causality that Hume doubted. Kant looks at judgements that we have and says that they must be the “effect of something” (Schulze 137). Kant then assumes that given that they are an effect, they must have a cause. This cause is the mind (making them a priori), leading to all the other properties they have, such as being the form of sensibility and only being able to be applied to empirical intuition (Schulze 137). This, however, assumes that Hume’s doubts are no longer valid, since we cannot know if judgments even have a cause or ground and if, more generally, the laws of causality apply to actual things. Schulze speaks for Hume:
Rightly he could say: “As long as…the concepts and principles of causality…are still uncertain and disputed…it is pointless to want to enquire into the sources of the various parts of human knowledge, or to establish anything about them. For before we have the right to ask ‘What are the sources and causes of our knowledge’, we must already have established that for every actual thing there exists a ground and its cause, and, specifically with respect to our knowledge, that all its determinations are the effect of particular causes.” (Schulze 139)
Even if we were to acknowledge that assume that the claims as to the causes of our knowledge are given, Kant will beg the other sceptical question (of what we can know about objects-in-themselves from our representations of them). Kant, Schulze claims, begins with the claim that necessary synthetic judgments cannot be thought to have necessity if they do not have a ground in the mind (140). From this, Kant makes a leap to claiming that they therefore are grounded in the mind (Schulze 141). This is assuming that we can make claims about things outside of our representation (in this case, that necessary synthetic judgments are based in the mind) based on our representations of them (in this case, that necessary synthetic judgments cannot be represented as possible unless they are based in the mind). This is a clear violation of Hume’s sceptical principle, and seems especially problematic because it also does not mesh with Kant’s own insistence that we cannot know anything about the realm of things-in-themselves (141-142).
Kant runs into similar problems when he tries to speak of the nature of a priori judgments as only applicable to empirical intuitions. Schulze frames Kant’s argument as similar to his argument about the mind being the source of a priori cognitions: we can only think of one way in which it is possible that we can have concepts that precede yet refer to a specific representation: to have those concepts exist as the form of our cognition of those objects (Schulze 150). Therefore, this is how these a priori concepts actually are. This, like the first case, assumes that the nature of things-in-themselves depends on our representations of them, “that something can be only so constituted—objectively and actually—as we are capable of representing it to ourselves” (Schulze 151, emphasis in original), which completely ignores Hume’s doubt that we could ever know such things and Kant’s own claim that we cannot know the thing-in-itself.
Schulze also finds problematic Kant’s description of the what exactly the mind is that supposedly underlies these judgments. Kant does not make it clear whether his system assumes the mind as a thing-in-itself, a noumenon, or a transcendental idea (Schulze 166). The first options, that the mind is a thing-in-itself or noumenon, follow a common line of thought, according to Schulze: that something “real” (the representations) also need something “real” as a ground (154). The problem here, of course, is that attempting to say that the source of our necessary synthetic judgments derives from either of these goes against both Kant and Hume’s philosophy. The first and familiar objections are that a) if the mind or subject is a thing-in-itself or noumenon, we cannot know anything about it, as we cannot know the nature of things outside of experience and b) we cannot assume that it can even be a cause, given that the laws of causality may not apply to things-in-themselves (Schulze 155). It also, yet again, goes against Kant’s own philosophy, as it would make no sense from a critical philosophical point of view to depend on a supposedly unknowable thing-in-itself or noumenon as the ground of a priori synthetic judgments.
The third option is that the mind exists not as a thing-in-itself or noumenon but as a “transcendental idea.” This is a Kantian concept of an a priori concept that brings “unbounded unity and completeness to our experiential cognitions” (Schulze 164). One of Kant’s proposed ideas is that of an “absolute complete subject which is not, in turn, the predicate of another thing” (Schulze 163). This, however, also runs into an issue. The purpose Kant puts these transcendental ideas towards, according to Schulze, is to move experiential knowledge towards completion or perfection—they are not applicable to non-experiential knowledge. The use of these concepts to ground the necessary parts of our knowledge—the a priori cognitions—is therefore seriously flawed (Schulze 171). Yet again, Kant seems to assume that we can have knowledge of things outside of experience even though his own philosophy denies this very claim.
[1] I will not go into the faults with Schulze’s understanding of the critical position in this paper, but as Di Giovanni has pointed out, it fundamentally misunderstands the nature of the thing-in-itself for Kant and Reinhold (24-25). Their position is that the thing-in-itself exists merely as an unknowable “something” which must be the cause of our representations, for if we cannot ascribe the representations to an object outside of us, we cannot be sure the representations themselves are separate from us and we thus doom ourselves to solipsism.
[2] Hume does not claim that the thing-in-itself is completely unknowable in principle, simply that right now we cannot necessarily know anything about it through scientific methods or say that the laws of cause and effect apply to it (Di Giovanni 20). This is a major difference between Kant (at least as viewed by Schulze) and Hume.