Notes on <Kant's Transcendental Idealism> - The Thing in Itself and the Problem of Affection
First part deals with the general problem of finding a justification for referring to things as they are in themselves. The second traces the relationship between the concept of the thing in itself and noumenon and transcendental object. Third attempts to provide a solution to the problem of affection, by suggesting a sense in which Kant can consistently maintain that things as they are in themselves affect us.
The Problem of the Thing in Itself
The problem of the thing in itself can be described as the problem of having to provide a legitimate, nonpolemical use for the concept. The polemical use is justified by the critique of transcendental realism; it enables Kant to explain the errors of his predecessors and to show how the objects of human cognition are not to be considered in a philosophical account. However, this does not justify any positive use for the concept within the “critical philosophy” itself. There are numerous places in which Kant does speak affirmatively about things in themselves, for example:
The word appearance must be recognized as already indicating a relation to something. …must be something in itself, an object independent of sensibility [A251-52]
These passages speak of the “thing in itself” in two different ways. For the most part it is the object that appears. But there are places where they refer to entities ontologically distinct from the sensible objects of human cognition.
It should not be inferred that Kant is confused about the transcendental distinction. The appearance - thing in itself distinctions indicate a contrast between two ways in which the objects of human experience can be considered in transcendental reflection. The conception of a noumenon as an ontologically distinct entity is required only in order to allow for the possibility of conceiving of God (and perhaps rational souls). This conception is important for Kant’s metaphysics, but it does not enter directly into a transcendental account of the conditions of the possibility of human knowledge. The task of a transcendental justification of the concept of the thing in itself is to explain the possibility and significance of considering “as they are in themselves” the same objects which we can know only as they appear, it is not to license the appeal to a set of unknown entities distinct from appearances.
Unfortunately, the search for a justification in the Kantian texts does not appear to meet with much initial success. There are two distinct lines of argument in “defense” of the thing in itself. One account is the need to acknowledge a “cause” or “ground” of appearances. This can be termed the “causal interpretation”. It construes the relationship between an appearance and a thing in itself to be effect and its cause. An obvious problem with this is that it requires that we take the appearance and the corresponding thing in itself as two distinct entities and the notion of “noumenal causality”. This cannot provide the needed justification, since it presupposed that we can refer to things in themselves when we take them as causes of appearance.
Other passages suggest that Kant’s claim is semantic. On this reading, Kant is affirming a relation of logical implication between the concept of an appearance and the concept of a thing in itself. The basic thought here is that the expression “appearance” is parasitic upon the expression “thing in itself”, to use the former expression is already to presuppose the legitimacy of the latter.
It is also clear that the straightforward semantic approach is no more successful than the usual argument. First it assumes the expressions “appearance” and “thing in itself” refer to two distinct entities. The claim is that reference to entities of the former sort presupposes the possibility of reference to those of the latter sort. Therefore it is not applicable to the transcendental distinction between two ways of considering one and the same thing. Second, the attempt to modify the argument so as to make it relevant to the transcendental distinction seems to lead to incoherence. The problem here is that the notion of considering a thing as it is in itself is presented by Kant in essentially negative terms: to consider a thing in this way is just to consider as it is apart from the condition under which it appears to us and thus as not being an appearance. Consequently if we apply the semantic argument to this distinction, we arrive at the claim that the designation of a thing as an appearance requires that we also designate the very same thing as nonappearance, which means Kant is requiring us to contradict ourselves.
The distinction is not between a thing considered as an appearance and the same thing considered as a thing itself; it is rather between a consideration of a thing as it appearance and a consideration of the same thing as it is in itself. The relevant term functions adverbially to characterize how we consider things i n transcendental reflection rather than what it is that’s being considered. To consider something as it appears, we must also consider it as it is in itself.
We can assert the nonspatiality and nontemporality of things considered in themsevles without violating the principle of “critical” agnosticism. We can do so because such claims do not involve any synthetic a priori judgments about how things really are in contrast to how they merely seem to us. On the contrary, they involve merely analytic judgments, or methodoloigcal directives, which specify how we must conceive things when we consider them in abstraction from their relation to human sensibility and its a priori forms.
The Noumenon and the Transcendental Object
We are now in a position to treat the concept of the noumenon and the transcendental object, which are both intimately connected with the concept of a thing considered as it is in itself. Noumenon is the epistemological concept par excellence, characterizing an object of whatever ontological status, considered in a non-sensible manner of cognition. To know an object in this manner is to know it as independent of its relation to the structure of cognition. This is equivalent to know it as it is in itself.s
The “understanding” must think noumena is the critical understanding, or the human understanding qua engaged in transcendental reflection. The critical understanding must think noumena because this concept is a correlate of the transcendental concept of appearance and is thus intimately connected with the doctrine of sensibility. In fact, it is just this connection with sensibility that enables it to function as a limiting concept. Its specific task is to “curb the pretensions of sensibility” by referring to a different manner of knowing (intellectual intuition) with respect to which the objects that appear to us as subject to sensible condition would be known as they are in themselves independently of these conditions. An object known in this way is, by definition, a noumenon.
The “unknown something” into which the rich concept of the noumenon at work in the Inaugural Dissertation is transmuted is now called the “transcendental object”. Kant uses the term in at least two distinct ways. For the most part the transcendental object seems to be equated with the thing in itself. There are some places it is not the case. These two ways corresponds to two ways one can speak transcendentally about an object distinct from our representations.
First the object in the “weighty” sense, or the object “corresponding to, and also distinct from our representations”. Since we can’t stand outside of our representations, such object can be thought only in general = x. IN the context, the concept functions as a kind of transcendental pointer.
Second the object considered as it is in itself. Here the object is not simply considered as distinct from our representations, but also as distinct from the sensible conditions under which an object can alone be intuited by the human mind, from our capacity to represent objects. Here the reference to the transcendental object serves to underscore the point that the consideration of an object as it is in itself does not yiled the determiante concept of a knowable object.
Kant distinguishes between transcendental object and noumenon saying that “the transcendental object, that is, the completely indeterminate thought of something in general. This cannot be entitled the noumenon, for I now nothing of what it is in itself, and have no concept of it save as merely the object of a sensible intuition in general, and so as being one and the sam for all appearances” (A253). It’s strange to see Kant denying that the transcendental object is the noumenon on the grounds that he knows nothing of what it is in itself, as if he could know what the noumenon is in itself. His point, however, is simply that the object to which I refer my representations must be described merely as a transcendental object, not as a noumenon, because I am lacking a faculty of nonsensible intuition. The underlying assumption is that if I had such a faculty, the object would be a genuine noumenon, and I would know it as it is in itself.
In Second Edition Kant introduces the distinction between a positive and a negative sense of the noumenon. The former is understood as “an object of a nonsensible intuition”, and the latter “ a thing so far as it is not an object of our sensible intuition” (B307).
Affection
Despite the unknowability of things in themselves, there are passages in which Kant characterizes the thing as it is in itself as the nonsensible “cause” or “ground” to appearance or our sensible representations.
F.H. Jacobi gives a classical formulation for the problem of affection. Starting with the uncontroversial premise that sensibility requires that the human mind be affected by objects if it is to have any material for thought, he points out that there are only two possible candidates for the affecting object: appearance and transcendental object (which he equates with the thing in itself). His strategy is to show that neither can do the job. The former can’t do it because it is defined by Kant as a mere representation in us; the latter can’t do it because of its unknowability which precludes the application to it of any of the categories, including causality.
The issue is whether the affecting object is an appearance, a thing in itself, or perhaps both. This is based on the assumption that the distinction of appearance and thing in itself is a distinction between two entities.
Before addressing the problem, it is necessary to dismiss the objection that the very notion of an empirical affection is incompatible with the Kantian philosophy because empirical objects are appearances and appearances are “mere representations in us”. The problem is that whether there is any warrant for assuming that a statement about an object affecting the mind involves a reference to the object considered as it is in itself.
Such a warrant is provided by the fact that affection is taken as a necessary condition of the possibility of experience and in this sense as part of a “transcendental story”. The temptation is to argue against Kant that even if one grants that the assertion “something must affect the mind” has a transcendental status in that it expresses a necessary condition of the possibility of experience, this does not require us to construe the expression “something” or its equivalents as doing anything other than referring indifferently to one or more of the members of the class of empirical objects. All that’s been established by this is the entirely general claim that some object or other must affect the mind, if the mind is to have any experience. If this is the Kantian warrant for the injection into the transcendental story of a reference to the transcendental object, then the whole account rests on a failure to distinguish between “something”, construed as having an indefinite reference, with “a something in general”, construed as a name or referring expression.
The problem with this criticism is it ignores one half, the most important half, of the transcendental story. Since Kantian theory of sensibility not only requires that something be “given to” the mind, it also maintains that this something becomes part of the content of human knowledge only as the result of being subjected to the a priori forms of human sensibility. But this something that affects the mind cannot be taken under this empirical description. The thought of such an object is the thought of something non-sensible, non-intuitable and hence “merely intelligible”.
If it is a necessary condition of human experience that something affect the mind, it is a necessary condition of a transcendental account of such experience that this something be viewed as a “something in general = x”, that is, as transcendental object. But this does not commit Kant to the illegitimate postulation of any super-empirical, unknowable entities. The point is that insofar as such entities are to function in a transcendental context as material conditions of human cognition, they cannot be taken under their empirical description.
[A494-95/B522-23] falls into two parts, each asigns a different role to transcendental object. First part notes immediately the reference to the “non-sensible” and thus “unknowable” cause of our representations, which is equated with the “purely intelligible cause of appearance in general.” “It provides something corresponding to sensibility viewed as receptivity.”
In the second part, Kant seems to expand the role of the transcendental object. “the whole extent and connection of our possible perceptions” and “the real things of past time are given in the transcendental object of experience”. Transcendental object functions as “conceptual repository” for our way of referring to the remote past or distinct regions of space. The concept of the transcendental object is simply a higher-order empirical concept, one which refers to experience as a whole.
To say this is not to commit ourselves to view that Kant is offering some kind of metaphysical explanation in the manner of Berkeley’s divine mind, which allows us to “save” the reality of unperceived objects and events, instead it repudiates any such metaphysical explanation, thereby enabling us to se that we must define the reality of past events in terms of their connection with the present “in accordance with the laws of the unity of experience”.
Far from providing a metaphysical story about how the mind or noumeal self is somehow mysteriously affected by the transcendental object, they merely stipulate how the affecting object must be conceived in the transcendental, or non-empirical account of the affection required for the explication of the Kantian theory of sensibility. These claims involves the use of the categories, especially causality. However it dos not justify the frequently voiced criticism that Kant is guilty of an illicit application of the categories to things as they are in themselves. The point here is simply that the function of the categories in these transcendental contexts is purely logical and does not carry with it any assumptions about their objective reality with respect to some empirically inaccessible realm of being.