The Thing-in-itself: A Problem Child
Never have I had a text unravel so quickly as when I asked the Critique of Pure Reason what role the thing-in-itself plays within it. I’ve become firm in my belief that Kant himself didn’t know. This view is financed by the fact that in the First Edition of the Critique, Kant goes for it, and reaches a point where he’s almost completely dispelled the thing-in-itself (an undertaking we will look at in this article). But, then, it is precisely these passages that seem to be the target of the Second Edition. No deletion is more substantial between the editions than the purge of this metaphysical bravado. Furthermore, in the introduction to the Second Edition, he refers to the newly written sections that are intended to stand in for what has been deleted, and provides a quick rewrite (“it should say…”) of one of the sentences (the text itself obviously already on the wagon to the printers), and then, in a footnote directly following this, he counters a number of objections that could be raised with his new, more conservative characterization. One has the sense that the courier who was sent to collect the text of the introduction from Kant was stopped again on the garden path on his way out so Immanuel could add just one more paragraph. Kant never worked it out. It truly is his “Problem Child”.
In this article I want to return to those First Edition passages, where Kant was bold and headstrong, and for the sake of completeness in my research into this question, sketch out the characterization of the thing in itself as it appears there. I think you’ll be surprised by just how Hegelian the First Edition Kant is.
· · ·
So what’s the problem? To quickly outline it, Kant is generally characterized as advocating for the following positions:
- Appearance needs to be the appearance of something itself not appearance, the thing-in-itself.
- We cannot know anything about things-in-themselves.
In fact, these two theses are so widely attributed to Kant, we can (and will) call them the TLDR version of the entire Critique of Pure Reason. However, as should be obvious in the way that I have worded them above, they are in direct contradiction with each other! This contradiction only exists in the Second Edition. No matter how many footnotes on introductions, post-it notes appended and last minute revisions, Kant couldn’t resolve it. There is a metaphysical issue of how something beyond space and time can be the cause of appearances, and there is an epistemological issue of how we can know that something is unknowable. The Second Edition does a good job of dealing with the second of these issues, but the first remains.
However, this difficulty doesn’t exist in the first edition, where Kant actually criticizes this (the TLDR version) view of things. Let’s check it out.
The First Edition Argument, Part 1
In the “Paralogisms of Pure Reason” (in the first edition of the Critique of Pure Reason) Kant actually attributes the “TLDR reading of Kant” to a position he dubs “Empirical Idealism” and instantly contrasts his own position of “Transcendental Idealism” in opposition to this (Kant, 2007, A369).
The issue of the role of the thing-in-itself, Kant argues, only becomes an issue for the view that the objects “out there” are only accessible as representations, not as things-in-themselves. Thus, the objects “in here” (which include the representations of objects “out there”, as well as thoughts, etc) are given immediately, but the reality of external objects is always only ever an inference, inferred from our representations as being their cause. But as an inference from an effect can admit of multiple causes, we are left with the possibility of skepticism towards external objects, things-in-themselves as they are in themselves, with no way of allaying this skepticism, each of us in Descartes’ melodramatic meditations. However, Transcendental Idealism avoids this uncertainty by making of our representations the objects themselves as such.
The argument works like this:
All external objects are given to us in space via sensation. Sensation occurs within the representation of space, and is then, within imagination, constituted as this or that object occupying this space. As space is ideal (as per the arguments in the Transcendental Aesthetic) and formal, it is not real as such but merely represents the possibility of co-existence. The sensations, though, are real because the givenness of sensation is the definition of “real” within the ideal frame of space, differentiating “outer” objects from purely “inner” objects (imagined ones for example). The self-presence of my representations, taken as “inner”, is what provides the surety of the cogito, ergo sum, assuring its reality, and the immediate givenness of sensation provides the reality of “outer” objects as it is the basis upon which these objects are “outer”. Sensation, then, is thus being taken as immediate and simple.
As Kant puts it “space itself is nothing but mere representation, so that nothing in it can be taken as real, except what is represented in it” (2004, A375). That is to say, if we inquire into the reality of an object, that is, a thing in space, then there is no further question of its reality than its actual representation in space. Or as Collins (1999) puts it “the only thing about which we might even raise the question “Is it real?” is the thing that exists in space, the thing that sensation marks.” (p.79)
Underpinning Cartesian skepticism, and the mystery of metaphysical interactionism (non spatial-temporal objects causing appearance), is the desire to peer behind our representations and confirm the nature of the things-in-themselves. Kant’s reply amounts to saying: ‘if you want to know what it really looks like, then look at it, it’s right there. If you want to know what it really feels like, then touch it. So on and so forth. There is nothing more that your curiosity can sensibly lay claim to.’ This is why Kant in this section claims that a Transcendental Idealist is also an Empirical Realist: if space is represented in us, all that is represented in it is also in us, that is, appearance. Sensation (also “within us”) marks the distinction between the inner and outer sense, and provides the only reality that can be inquired after. That is to say, the objects of empirical intuition are perfectly and wholly real, without the need for an inferred object beyond them being real in order for them to be real. To be “real” means to actually (as determined by sensation) be in space, our appearances are determined by sensation to be in space (which the ideality of space allows us to be certain of), and thus possess all of the reality needed to be called “fully real”.
The First Edition Argument, Part 2
However, Kant is careful to distinguish the two senses of “outer” (Kant, 2004, A373). His argument applies to “outer” in these sense of “objects out there in space”, that is “outer” in all and any spatial terms. The other sense still remains, and that is “outer” in the sense of independent of us. This he refers to as the “transcendental object” (yet another term easily conflated with but distinct from thing-in-itself, and noumenon). It still hovers.
“The transcendental object, which forms the foundation of outer appearances, and the other transcendental object (the cogito), which forms the foundation of our inner intuition, is in itself neither matter (sensation), nor a thinking being, but is simply a ground (to us unknown) of appearances which supply to us the empirical concept of both” (Kant, 2004, A379–380, my elucidations in italic)
The Transcendental Object
The transcendental object is introduced as the “concept of an object in general” in A109. Its theoretical function is allowing the categories to interact with the manifold of sensible intuition. For example, I take the lamp in front of me as a particular object, and take it to be real as a bundle of sensations appearing in space. In its particularity, it allows for various conceptual determinations via the understanding: it is one in quantity, it shares the same basic quality of blueness with the book on the shelf, it admits for certain possible interactions and transformations, and so on. As I look around at the other objects in the room I find they too have received these similar conceptual determinations allowing me to compare and contrast them with the lamp and unify them along certain lines in thought (such as adding up the quantities). This then refers me to categories simply, for example the category of quantity, through the abstraction of each particular set of determinations of each individual object given in line with sensation. I arrive at the concept of an object in general as a pure form of “objecthood” that is not itself an empirical object in space (as all particular sensations have been abstracted from it) but an object of thought that organizes the categories. It’s this organization that has us apprehend particular objects as being fully, and not just partially, conceptually determined.
Kant underscores that “this object cannot be called the noumenon” (A253, In Kitcher, 2011, p.206). Why not? It seems to be a pure intellectual object of the understanding (the definition of noumenon, as in nous). However, this transcendental object is what appearance is referred to: not only does it have no legitimate use beyond its application to sensibility (as with the categories), it has no meaning beyond this application. Unlike a noumenon that the understanding may entertain beyond sensibility, the concept of the transcendental object wholly concerns things that may appear in sensibility. So, though I can conceive of a purely intellectual noumena (the number 4,382.7 existing as a platonic object, for example) I cannot conceive of an “object in general” in isolation of the possibility of real sensations (empirical intuitions). However, as it is an object in general it has no determinate intuition corresponding to it, only the possibility of the intuitions of countless “objects in particular”.
“The pure concept of this transcendental object (which really is always the same = X in all our knowledge) is that which alone can provide for all our empirical concepts in general a reference to an object, or objective reality. This concept cannot contain any determinate intuition, and can therefore only apply to that unity which must be found in any manifold of knowledge insofar as this manifold has reference to an object” (Kant, 2004, A109).
To give a simplified analogy: In regards to a single object, I take the five simple orders of sense (sight, sound, touch, etc) to be coordinated regarding it. It’s this coordination of these different sensations that allows me to understand that each of them are issuing from this single object. If I “taste” the grape before I see it go into my mouth and after I cease to feel it there, I would have no grounds to believe that the different orders of sense related to the same object. They would form no “object” at all. In this simplified example the “transcendental object” is the organization of the senses in general. It makes no sense to think of this organization purely intellectually (as a noumenon) and abstract it from sensation, because it wholly concerns sensation despite not being given directly in sensation (there is no additional sensation of the coordination of sight and hearing, for example, just the sight and the sound). Merely speaking of the five sensory orders, we already see how important this conceptual coordination of sensation is for our being able to see the world as populated by objects (rather than just a flux of sensations), as well how important it is in our use of sensation to attribute reality to objects in our experience (checking that the object coordinates all five senses, as with trying to touch a hologram).
Kant is not referring to the mere coordination of the five senses, but to the coordination (unity) of the entire manifold of knowledge concerning an object: apprehension of representations, reproduction of them over time, and recognition under a concept (Kant, 2004, A98). The transcendental object is that by which our multitudinous knowledge can converge onto an object, and not be fragmented and “out of joint”. Without this unity, objects themselves wouldn’t be known to us, and through this unity the manifold knowledge of an object points to its absolute unity in a “completed” knowledge. It’s in reference to this “completed knowledge” that Strawson et al’s Humility Argument enters, with its reference to the further potentials of the object beyond what is given to our mode of cognition. But we needn’t postulate “possible beings” who cognize differently (As Strawson, and Bird do, for example). However it is they cognize (these beings with faculties unlike our own), if they possess knowledge of an object, then their knowledge admits of some degree of coordination towards it. On this point we and them can share an understanding of the transcendental object as ‘that which’ coordinates that knowledge (be that within us, separately and perhaps differently, or belonging to the thing-in-itself that appears to each of us).
That is, there is some X that causes our knowledge (including our representations and sensations) to converge into stable, non-contradictory unities. It is this that is the foundation and ground of appearances because it determines the laws by which they appear as they appear, providing the bounds on possible appearance.
“-in the manifold of their representations an accompaniment or succession takes place according to certain rules; for without that our empirical imagination would never get to do anything suitable to its capacity … If cinnabar were now red, now black, now light, now heavy, if a human being were now changed into this animal shape, now into that one, if on the longest day the land were covered now with fruits, now with ice and snow, then my empirical imagination would never even get the opportunity to think of heavy cinnabar on the occasion of the representation of the color red” (Kant, 2004, A100–101).
Note that Kant here sees that the regularities in the order of appearance that allow us to form knowledge at all are just that: regularities within the order of appearance. The transcendental object (the concept of an object in general) is both the ground of appearance (the ‘that which…’ of its unity) and the ground of our knowledge towards the object of appearance. Now it is the question of whether this ‘that which’ (causes the manifolds of sensation and knowledge to converge and coordinate into unities) is a transcendental a priori feature of appearance “in us”, or something independent of us, that cannot be worked out. This is the necessary agnosticism of the thing-in-itself. Not something beyond appearance, “out there” in the world independent of us, but an a priori feature of appearance that cannot be determined to be independent of or dependent on us.
“By warning sensibility that it must never claim to apply to things in themselves, but only to appearances, (the understanding) forms the thought of an object in itself, but only as a transcendental object. This object is the cause of appearance (therefore not itself appearance) and cannot be thought as magnitude, or as reality, or as substance, etc. (because these concepts require sensible forms in which to determine an object). Of this object, therefore, it must always remain unknown whether it is to be found only within us, or also without us; and whether, if sensibility were removed, it would vanish or remain.” (B344–345|A288)
So, to briefly summarize, if by “thing-in-itself” it is meant the object correlate existing “out there” whose reality causes the object of experience, then this “thing” is meaningless within Transcendental Idealism: the object of experience is the real object, for to be “real” means to appear in space. Nothing in space is real except that which is represented within it. Now beyond this representation of an object appearing in the representation of a space, we may inquire what the grounds of all of this representation is. That is, the “thing-in-itself” as that which is wholly independent of us. To this Kant’s answer is that which provides unity to appearance. Now whether this is “in us” (and thus not truly independent) or independent of us cannot be worked out. An assertion one way or the other is of the precise type that Kant legislates against. This means that Kant has not contradicted himself in depending on a metaphysical interactionism the argument for which is inadmissible by the very bounds he places on thought, but rather has precisely identified what role a “thing-in-itself” might play, how it is to be thought, but remains agnostic whether this role is truly played by a thing-in-itself, a transcendental object, or is an a priori condition of appearance, with there being nothing beyond it. However, this is not an ambiguity between realism and solipsism: Kant can give an account of the reality of objects in experience, as well as the universality and applicability of knowledge, that is, of objectivity. What remains murky is a confirmation on a point that is beyond knowledge. That is, even if we received confirmation, we would not know what to do with it and it would make little difference to us.
Kant, in the introduction of the Second Edition, when he is explaining why these arguments have been deleted, refers to wide-spread misunderstanding as the cause for the revisions. This might be more or less just a way of saying “I have been conclusively refuted on these points”. Perhaps there are arguments contemporaneous to Kant that successfully deconstructed his view as it has been put here. However, I feel the argument of the Critique is stronger overall before the cuts, after them it falls into a far simpler contradiction on this thing in itself point, with its emphasis on the noumenon and that appearance has to be appearance of something. Perhaps, rather, the kick-back Kant got was just simply because his peers were not ready for something so radically antagonistic to the common sense…
Bennett, J. (2016). Kant’s Analytic. Cambridge: Cambridge University Press.
Bird, G. (2006). The Revolutionary Kant. Chicago: Open Court.
Collins, A. (1999). Possible Experience: Understanding Kant’s Critique of Pure Reason. Los Angeles: University of California Press.
Kant, I., & Weigelt, M. (2007). Critique of Pure Reason. London: Penguin Books.
Strawson, P. (2007). The Bounds of Sense. Abingdon: Routledge.