Kant and the Problem of Knowledge

Rethinking the Contemporary World

Luigi Caranti, Alessandro Pinzani, Luigi Caranti, Alessandro Pinzani

• Publisher: Routledge India
• Year: 2022
• ISBN:
  9781000606881
• Edition: 1
• Topic: Filosofía
• Subtopic: Historia y teoría filosóficas

This book examines Kant's contributions to the theory of knowledge and studies how his writings can be applied to address contemporary epistemological issues. The volume delves into the Kantian ideas of transcendental idealism, space, naturalism, epistemic normativity, communication, and systematic unity. The essays in the volume study Kant's theories from a fresh perspective and offer new arguments for assenting that knowledge cannot account for itself without acknowledging the fundamental role of the cognitive subject. In doing so, they suggest that we reconsider Kant's views as a powerful alternative to naturalism.

Featuring readings by well-known Kant specialists and emerging scholars with unorthodox approaches to Kant's philosophy, the volume fills a significant gap in the existing scholarship on the philosopher and his works. It will be of great interest to scholars and researchers of knowledge, philosophy, and epistemology.

1 THE TRANSCENDENTAL DEDUCTION AND TRANSCENDENTAL IDEALISM A retrospective

Henry E. Allison¹

DOI: 10.4324/9781003050742-2

The publication of P. F. Strawson’s The Bounds of Sense in 1966 was a major event in Anglophone Kant interpretation, because it appeared to make it possible for philosophers trained in the analytic tradition to have their Kant without the disreputable idealism that was generally thought to be inseparable from it. Strawson accomplished this by distinguishing sharply between what he regarded as the “analytic achievement” (Strawson 1966: 271) of the Critique and the transcendental idealism with which it is intertwined. Strawson’s positive focus was on the Transcendental Analytic, and he considered Kant’s achievement insofar as it is contained in this portion of the Critique “analytic”, since, in his reading, it attempted to demonstrate that certain general features are necessary ingredients in any coherent conception of experience that one can ascribe to oneself. Paramount among these is that the experienced world must contain at least relatively permanent re-identifiable objects and an overall regularity in their interactions and changes (Strawson 1966: 24, 28). Strawson characterized Kant’s idealism in numerous ways, but the most striking is as the doctrine that “reality is supersensible and that we can have no knowledge of it” (1966: 38) According to Strawson, the reason Kant appeals to transcendental idealism in support of his claims about the structure of experience is to account for their presumed necessity. In this widely shared view, the fundamental idea underlying Kant’s so-called “Copernican Revolution” is that necessity can be established only on the basis of the metaphysically idealistic premise that the order of nature is imposed by the human mind, which Strawson describes as the theory of the “mind making Nature” (1966: 22). And closely related to this is his dismissal of Kant’s account of the mental processing through which this “making” is supposedly achieved, which he refers to as “the imaginary subject of transcendental psychology” (Strawson 1966: 32).

Strawson’s work has spurred ongoing debates regarding the nature of Kant’s idealism and its relation to the Transcendental Deduction, as well as the broader issue of the force of transcendental arguments in which, to paraphrase Robert Stern, it is claimed that the possibility of some supposedly indisputable fact about us or our mental life, for example, that we have experiences, presupposes some state of affairs, for example, the existence of enduring material objects, that the skeptic has called into question (Stern 2000: 6). One of the familiar objections to such arguments is that they rest on some undefined idealistic commitment, which is frequently considered sufficient to undermine their force (Stern 2000: 49–58). My concern here, however, is with the bearing of this objection on Kant’s Transcendental Deduction, and my intent is to defend Kant against it by showing that this idealism, properly understood, is neither disreputable nor separable from the argument of the Transcendental Deduction. The discussion is divided into three parts: (1) an analysis of Strawson-inspired criticisms of the Deduction, which turn on the propositions that Kant’s mistake was to attempt to demonstrate the unconditional necessity that appearances conform to the categories, when all that is either required or possible is a merely hypothetical or conditional necessity; (2) an explanation of why the stronger claim is required; (3) a consideration of the connection between the argument of Transcendental Deduction and transcendental idealism from the perspective of Lucy Allais’s discussion of the issue in her recent book (2015), in which she both defends this idealism and argues that certain aspects of the Deduction are separable from it.

1. Kant after Strawson

Until recently, much of the post-Strawson Anglophone Kant literature has been devoted to the project of separating as much of Kant’s theoretical philosophy as possible from the supposedly excess and disreputable baggage of transcendental idealism. I shall here consider the views of two philosophers who have played a significant role in the development of this line of thought but, as far as I know, do not consider themselves Kantians: Ross Harrison and Quassim Cassam. ²

(A) Harrison: Harrison’s concern is to defend transcendental arguments, understood in roughly the manner described by Stern, namely as anti-skeptical arguments in which the negation of a skeptical conclusion is shown to be a necessary condition of a premise that the skeptic denies, against the objection raised by Bernard Williams and others that they presuppose some kind of idealism (Harrison 1982: 211–24). Assuming that a dependence on idealism in any form would undermine such an argument, because it allows the skeptic to reply that all it could show is how things must seem to us, his self-appointed task is to demonstrate that these arguments can succeed without any dubious idealistic premises But though his primary concern is with contemporary forms of transcendental argumentation, he does focus a significant amount of attention on Kant, whom he considers the philosopher who both invented this form of argumentation and the exemplar of the folly of attempting to provide it with an idealistic grounding.

Harrison’s major move, which was anticipated by Strawson and subsequently endorsed by many others, is to isolate and call into question the underlying assumption on the basis of which idealism is supposedly introduced into Kant’s core transcendental argument that any world experiencable by beings like ourselves must be subject to the categories. Characterizing Kant’s “transcendental” claim in general terms as the proposition that the world is necessarily judgable, Harrison accuses Kant of a modal fallacy by conflating the merely hypothetical necessity that the world must be itself judgable if we are able to judge or experience it, with the quite different proposition that the world is necessarily judgable tout court (Harrison 1982: 215–16). And, in explanation of this fallacy, he appeals to Kant’s concern to establish certain synthetic a priori propositions about possible experience, which he can only do by assuming that the truth conditions of these propositions are imposed by the mind. As Harrison puts it,

Because appearances [for Kant] depend on us there is not a crucial distinction between hypothetical necessities in which something is only true if it is judged and categorical necessities which are true in themselves. For appearances are not things in themselves; their existence depends on judgment.

(Harrison 1982: 216)

Harrison’s proposal for saving transcendental arguments from the fatal infection of idealism consists of lowering the bar by insisting that a successful argument of this type need not establish such necessities. According to him, “a transcendental argument works without establishing the necessity of anything apart from hypothetical statements. It therefore does not need the idealist premise that what exists is necessarily judged” (Harrison 1982: 216). Harrison attempts to illustrate this by considering an argument of the Kantian type, which shifts attention from the objects of our knowledge to the “medium by which they are apprehended by us” (1982: 220). On this basis, he suggests that it is possible to establish indirectly propositions about the nature of whatever is cognized by beings like us through this medium, but he insists that the procedure goes awry when, instead of drawing from this conclusions about how things will necessarily appear to us in virtue of this medium, one draws conclusions about how they are necessarily in themselves, for then the objects do not have any existence apart from this medium, which is full-fledged idealism.

The most interesting feature of Harrison’s account is his response to a worry that he thinks it might generate, namely the merely contingent nature of the agreement between the cognitive requirements of the mind and the nature of the world. He suggests that without idealism, there appear to be two alternative explanations for this contingency: either it “was pure luck or that it was the product of the conscious design of some external cause” (Harrison 1982: 222). But rejecting an appeal to luck as a non-explanation and citing Kant’s own skepticism regarding the appeal to pre-established harmonies to dismiss the latter, Harrison suggests a third explanation, namely natural selection. In this view, the agreement between our forms of judgment and the world is the result of “normal causal processes”, such that “any individual who thinks (or acts) as if there is no causation in the world, or that the future does not resemble the past, is an unsuccessful mutation, ill-adapted to survive or breed” (Harrison 1982: 223). In a concluding paragraph, Harrison acknowledges that this explanation holds only up to the present time and that if causal relations cease to hold in the world we will no longer be able to judge it, but he insists that this is beside the point, since “[the] present harmony is all that it is required to explain and for which idealism was thought to be the only possible explanation” (1982: 223).

(B) Cassam: Although Cassam expresses skepticism regarding the viability of transcendental arguments under the best of conditions, the bulk of his account consists of a defense of a version of Strawson’s “objectivity argument” purified of any idealistic component (Cassam 1987: 355–78). As with Harrison, Kant is used as the model to illustrate the futility of appealing to idealism in order to answer a skeptic by means of such an argument. But inasmuch as Cassam’s account pays far greater attention to Kant’s text, particularly to the Transcendental Deduction, which he considers “as profound as it is opaque” (1987: 356), it calls for a separate treatment.

According to Cassam, “Transcendental arguments are concerned with the specification of conceptually necessary conditions of the possibility of experience” (1987: 356). After dismissing as a side issue Stroud’s objection that such arguments unavoidably reduce to some form of verificationism, he maintains that any argument of this type must contain two components, which he names a “Conceptual Component” and a “Satisfaction Component” (Cassam 1987: 357). The former is a claim about what counts as experience, while the latter refers to a state of the world, which must hold if experience, so understood, is to be possible. But recognizing that, so described, this argument-schema is too weak, since it leaves it open to the skeptic to challenge the assumed conception of experience, Cassam rede-scribes the Conceptual Component as the necessity of the possibility of distinguishing within experience “a component of recognition or judgment, which is not simply identical with, or wholly absorbed by, the particular item which is recognized, which forms the topic of the judgment” (1987: 359). ³ Correlatively, the Satisfaction Component is a degree of regularity in experience that is sufficient to make this possible. And it is assumed that the skeptic must accept the former, because it is either analytic or some other form of conceptual truth (Cassam 1987: 358–9).

Cassam tests his proposal by applying it to the A-version of the Transcendental Deduction. He admits that his reading is “somewhat artificial”, since it deals with only one aspect of the Deduction (its “objective side”) while setting aside for later consideration another aspect (its “subjective side”), and his intent is to show how closely his schema corresponds to this aspect of the Deduction. Eschewing details, the “Conceptual Component” is equated with the necessary unity of consciousness, and it is claimed that,

The Satisfaction Component then makes a claim about how the world must be if the unity of consciousness is to be possible, the upshot being that the truth of something which is explicitly claimed to be doubtful is necessary if the unity of consciousness requirement which with luck, the sceptic will already have been persuaded to accept, is to be met.

(Cassam 1987: 358–9)

Correlatively, the Satisfaction Component is that “appearances must display such unity and interconnectedness as is possible only if they are appearances of objects” (Cassam 1987: 361). Accordingly, there is a close fit between Cassam’s proposal and the objective side of the A-Deduction. But the most interesting part of Cassam’s account is the portion that he had initially neglected, which corresponds to the “subjective side” of the Deduction, where idealism makes its appearance. At this point, Cassam’s account echoes Harrison’s and Strawson’s and consists of two parts. First, suppressing for dialectical purposes a worry about the viability of transcendental arguments as such, he claims that,

It is quite sufficient to say that what a transcendental argument shows is that the world must be a certain way for experience to be possible; it is a further unnecessary step to insist that the world referred to is merely the world of appearances, and that it is in a way constructed by the understanding.

(Cassam 1987: 364)

Having dismissed any need for transcendental idealism, Cassam attempts to explain why Kant could have failed to realize this obvious point. His, answer, like Strawson’s and Harrison’s, is that it resulted from a conflation of two senses of necessity, namely between a merely conditional and an unconditional necessity (Cassam 1987: 366). The former, which is assumed to be warranted by Kant’s objectivity argument, is that appearances must conform to the conditions of the unity of consciousness, that is, contain sufficient order and coherence, if they are to constitute experience. The latter is the unconditional necessity that appearances conform to this unity tout court. And this conflation is deemed responsible for the introduction of idealism, because it is only on the assumption that the requisite orderliness of appearances is imposed by the mind that their conformity to the conditions of the unity of consciousness can be guaranteed.

The final step is to account for Kant’s erroneous assumption that such a guarantee is needed. Cassam sees this as grounded in what he, like Harrison, regards as Kant’s misguided worry about contingency, and he attributes to this worry Kant’s insistence on the necessity for a transcendental synthesis of the imagination as the objective ground of the affinity or associability of appearances (Cassam 1987: 366–7). Moreover, he dismisses this worry by distinguishing between two questions: (1) “What must appearances be like if they are to provide a basis for the unity of consciousness?” and (2) “[W] hy are appearances such as to provide for the unity of consciousness?” (Cassam 1987: 370). Since it is precisely the kind of question that a transcendental argument on his view is designed to answer, it is not surprising that Cassam finds the first a “genuine philosophical question” (1987: 366–7). The second question, however, is not treated so kindly, for he questions its legitimacy, despite the fact that Harrison appears to have provided a naturalistic answer to it. Instead, in dismissing the second question, Cassam appeals to a Witt-gensteinian quietism, claiming that,

It is enough… that appearances do conform to the conditions of the unity of apperception; they might not have done and they might not continue to do so, and whilst this might give rise to a certain sense of insecurity, this degree of insecurity simply has to be tolerated.

(1987: 370)

2. Response to Cassam

Although I accept Cassam’s d...