Home » Early Modern Philosophy » Kant’s Transcendental Idealism: A Hopeless Case?

Kant’s Transcendental Idealism: A Hopeless Case?

George P. Simmonds
Oxford Brookes University


The interpretive mayhem engendered by Immanuel Kant’s Critique has, in the space of two centuries, yet to provide a standard or altogether satisfactory exegesis of transcendental idealism, a theory which on all counts lies at the very heart of Kantian philosophy. This paper aims to delineate two of transcendental idealism’s most salient readings in hope of proffering a well-considered comparison and, ultimately, a proposal that neither interpretation provides an account which conforms unerringly to Kant’s own promulgations.

Part I: Kant’s Transcendental Idealism

The Kantian doctrine of transcendental idealism concerns itself with the distinction between appearances and things-in-themselves, i.e.. objects as they appear to us and objects as they are in and of themselves. Kant’s finishing thesis on the matter posits the human mind as an active contributor to the objects of its perception and thus, in some way, a direct authority upon the nature of reality as we know it (McCormick, 2012, §4).

An exhaustive exposition of transcendental idealism demands a full consideration of Kant’s Critique of Pure Reason (1998), an enterprise well beyond the scope of this paper. Thankfully, in the ‘Transcendental Aesthetic,’ however, Kant assures us that his views on space and time are of particular relevance here, and it is upon these views that the present section will aim to focus (Janiak, 2012, §6). It may, for the sake of clarification, be useful to juxtapose the Kantian notion of space and time with that of Newton (see Newton, 1990, pp.85-109, 823-60), whose transcendental realism epitomises the notion of external reality Kant aimed to oppose. With an eye to Newtonianism he writes:

Those […] who assert the absolute reality of space and time, whether they be subsisting or only inhering, must themselves come into conflict with the principles of experience. For if they decide in favour of the first […], then they must assume two eternal and infinite self-subsisting non-entities (space and time), which are there (yet without there being anything actual) only in order to comprehend everything actual within themselves (Kant, 1998, pp.166-7).

Here Kant presents the transcendental realist position as one which posits space and time as a pair of quasi-objects which exist independently of the human intuition. Without attending to Kant’s direct objections to this concept, it should suffice to say that he does not conceive of space and time as objects, quasi-objects, or indeed anything to be considered independent of human intuition. For him, they are to be conceived neither as things-in-themselves nor properties that can be perceived or verified empirically; they are rather ‘forms of intuition,’ that is, ‘a priori elements of sensible perception’ which would not ‘subsist in themselves’ if one were to contemplate them in abstraction from the minds of those to whose perception they are essential (Guyer, 2006, p.53). It is in this that Kant proffers the notion of the synthetic a priori proposition: observations on these necessary forms are synthetic, since ‘the predicate […] is not logically or analytically contained in the subject,’ but simultaneously a priori because they are ‘verifiable independently of experience,’ since they essentially constitute it (Encyclopædia Britannica Online, 2015). These forms are not, in other words, borne out of the objects themselves but imposed upon them as necessary conditions of the mind’s ‘receiving’ the external world (Schulting and Verburgt, 2011, p.5). When we look at a tulip as it is situated within spatiotemporal reality, then, we are not seeing it as it is, but as it appears to us following the intuition’s attempt to sort it into forms more easily digested by the understanding (van Cleve, 1999, p.134).

But what is the nature of an object beyond the veil of the mind? What are objects like when we are not considering them? It is when we ask questions like these that we stumble into Kant’s controversial distinction between appearances and things-in-themselves or, more concisely, phenomena and noumena. While on the one hand we have the phenomenon, an object as we perceive it through the prism of our intuition and understanding, on the other we have the noumenon, an object ‘unembellished’ by the mind and thus in possession only of those properties that are intrinsic to it (Walsh, 1901, pp.464-5). Things-in-themselves are on all counts considered inscrutable: while Kant claims we are able to perceive phenomena and acquire empirical truths regarding their relations, he insists that we will never apprehend the ‘unknown somethings’ of which appearances are mere representations (Kant, 1998, pp.276-96). This, he says, is impossible from the perspective of the finite mind.

As noted in Braiche (2008), a great deal of tension exists among Kantian scholars where transcendental idealism is concerned. While some interpret it as a doctrine interested in making ontological claims, others read the Critique as propounding an epistemic thesis (pp.2-3). The question of whether Kant intends to suggest that reality consists in two ontologically distinct worlds (one phenomenal, the other noumenal) is pivotal here; and it is around this question that the following sections will work.

Part II: Strawson and the Noumenal World

Needless to say, it is usually the thing-in-itself that provokes interpretive issues. Even Jacobi, one of the Critique’s earliest commentators, famously claims that ‘the “thing-in-itself” is the kind of concept without which it is impossible to enter Kant’s system, but with which it is impossible to get out of the system’ (Jacobi, 1912, p.304). Strawson’s The Bounds of Sense (1996) takes similar issue with Kant’s reliance on noumena and attempts to release transcendental idealism from its inconsistencies by attributing its metaphysical system to Berkeleyan idealism; that is, the notion that external reality is but a phenomenal illusion. ‘The only element in transcendental idealism which has any significant part to play in those structures,’ he writes, ‘is the phenomenalistic idealism according to which the physical world is nothing apart from perceptions’ (p.246). Despite the extremity of this deduction, Strawson’s ‘sortings of wheat from chaff’ are broadly acknowledged to stand among transcendental idealism’s most canonical interpretations (Bennett, 1986, p.340).

In reaching these conclusions Strawson focuses on the troublesome relationship between phenomena, noumena, and our cognitive faculties. He begins from what has become known as the ‘two-worlds’ reading of transcendental idealism, a view from which things-in-themselves and their appearances occupy two distinct realities, only the latter of which being comprehensible from the human perspective. The former, that noumenal ‘sphere of supersensible reality,’ must on Kant’s view transcend our intuitive notions of space and time as well as those ‘pure concepts’ which follow from them (such as that of causation) (Strawson, 1996, p.236). This interpretation is not without textual evidence. In Kant’s own words:

We should consider that bodies are not objects in themselves that are present to us, but rather a mere appearance of who knows what unknown object; that motion is not the effect of this unknown cause, but merely the appearance of its influence on our sense (Kant, 1998, p.435).

On this reading, then, we are to consider phenomena and noumena as ontologically distinct objects, one inhering within space and time and the other in some sort of transcendental realm of aspatiotemporal things-in-themselves. This is not to say that these worlds do not interact, however. Strawson insists that human experience must be the result of some ‘complex quasi-causal relation’ between phenomena and noumena, a connection he terms the ‘A-relation’ (Strawson, 1996, p.236). It is by way of this ostensible quasi-causality that noumena and human minds are able to ‘collaborate’ in their formation of the phenomenal world (Braiche, 2008, p.9).

But Strawson does not believe this relationship comports with Kant’s earlier conception of things-in-themselves as unknowable objects that do not conform to the modes of experience central to phenomenal nature. In support of this thesis he questions two aspects of the transcendental-idealist system in hope of bringing the notion of the thing-in-itself into question (Matthews, 1969, pp.206-7).

First, if noumena are unknowable and cannot be cognised, how is that we are able to know that they cause phenomena, or that they are in fact there at all? On Strawson’s reading of Kantian epistemology, things-in-themselves do not fall within the category of ‘possible human experience’ and thus possess neither the capacity to be verified nor any significant meaning as theoretical concepts. To insist that noumena exist despite this would be to approach transcendental idealism as a rationalist, where to claim that they do not would be to fall worryingly close to the extreme idealism of Berkeley (Strawson, 1996, pp.237-40). Kant here faces a dilemma, for he fits comfortably into neither camp. Second, if such notions as space, time, and causation do not exist beyond the realm of appearances, how is it that the A-relation is possible? How is it, in other words, that noumena are able to provide us with the material from which our cognitive faculties are able to construct phenomena? This, too, is a problem for Kant since it is not clear how this might occur without presupposing concepts of causation and, by extension, the forms of space and time (ibid., pp.246-8). While his epistemology is challenged on the first confutation, his ontological account of things-in-themselves is undermined in the second.

The Critique provides no easy way out of these difficulties and this, for Strawson, ‘tolls the death knell’ for transcendental idealism (Braiche, 2008, p.2). If we recall his statement that the doctrine’s only remaining foothold is the ‘phenomenalistic idealism according to which the physical world is nothing apart from perceptions,’ we see that Strawson chooses to equate Kant with Berkeley, both of whom deny the external existence of phenomena yet fail to affirm the things-in-themselves that would otherwise ground them in reality (Strawson, 1996, p.260). And thus Kant is, where his idealism is concerned, considered nothing more than an ‘inconsistent Berkeley’ (Allison, 2004, p.4).

Part III: Allison’s Two Aspects

How might transcendental idealism be navigated from Strawson’s impasse? According to the account proffered in the Bounds, Kant is describing two different classes of objects: the tulip as we see it, and the tulip as it is in and of itself. This is a metaphysical interpretation, and from this approach arises Strawson’s refutation. Allison does not read transcendental idealism this way; rather he views it as an appendage of Kant’s epistemic ideas. For him, the Critique does not intentionally discuss the ontologies of two distinct-yet-somehow-interactive tulips, but a single tulip considered in two different ways (Allison, 2004, pp.229-35). This, at least theoretically, diverts transcendental idealism away from the ambush Strawson prepares in his own exegesis.

Allison does not contend that there is nothing to be considered beyond phenomena: while he does not award things-in-themselves their own ontological status in the way of Strawson, he nonetheless acknowledges them as an important aspect to Kantian philosophy. On his view, sometimes linked with the ‘two-aspects’ position, what distinguishes a thing-in-itself from its appearance is not the domain of existence it occupies but the way in which the human mind considers it. Given that our cognitive faculties actively process and order the external world, thus giving rise to phenomena, it follows that an object of this reality may retain its own sort of existence where these devices are not present. This does not, however, entail the treatment of this existence as a separate, ontologically distinct entity, for it is simply an object of the phenomenal world considered in abstraction from the conditions under which we perceive it (Allison, 2004, pp.33-6). This interpretation is no more lacking in textual evidence than Strawson’s:

We can have cognition of no object as a thing in itself, but only insofar as it is an object of sensible intuition, i.e. as an appearance […] We [presume] the distinction between things as objects of experience and the very same things as things in themselves (Kant, 1998, p.116).

As seen above, phenomena and noumena are ‘the very same things’ considered in different contexts. We cognise phenomena as they appear to us within space and time, adorned with all the concepts part and parcel to human experience; while noumena are these same manifestations considered (via transcendental reflection) in the notional absence of such conditions. In this they retain a sort of methodological or formal status, but by no means an ontological one. This kind of formal significance is in no way peculiar to human thought: in theoretical physics we often consider objects abstracted from their necessary properties, but we do not insodoing commit ourselves to the belief that these abstracted entities exist in any real sense of the word (Allison, 1978, pp.53-4).

To avoid Strawson’s critique, Allison emphasises noumena’s negative role as more a description of what phenomena are not than an account of what might exist beyond the realms of possible experience. This account gives things-in-themselves a viable position within Kantian epistemology without force-feeding them the unwieldy metaphysical significance found in two-worlds interpretations. It shifts, in other words, the axis of the phenomena-noumena distinction from the way things are to the way (or whether) our cognitive faculties respond to them (Braiche, 2008, p.14).

Part IV: A Hopeless Case?

It seems fair to say that the accounts of both Strawson and Allison more-or-less conform to Kant’s original proposition; they would not, otherwise, be so widely discussed as valid interpretations. It is worth considering, however, that throughout the Critique Kant himself appears to oscillate between a two-worlds and a two-aspects position. Transcendental idealism is by no means a straightforward discipline to comprehend, and it could be that our failure to reach a univocal reading of its postulations owes to the irresolution of its author (Matthews, 1969, p.204). To end on a quotation from Wood (2005):

I think much of the puzzlement about transcendental idealism arises from the fact that Kant himself formulates [it] in a variety of ways and it is not at all clear how, or whether, his statements […] can be reconciled or taken as statements of a single, self-consistent doctrine. I think Kant’s central formulations suggest two quite distinct and mutually incompatible doctrines (pp.63-4).

Works Cited

Allison, H.E. (1978). Things in Themselves. Dialectica. 32 (1).

Allison, H.E. (2004). Kant’s Transcendental Idealism: An Interpretation and Defence. New Haven: Yale University Press.

Bennett, J. (1968). Strawson on Kant. The Philosophical Review. 77 (3).

Braiche, M. (2008). Strawson and Allison on Transcendental Idealism. Unpublished undergraduate dissertation. Lewis & Clark College.

Encyclopædia Britannica Online. (2015). Synthetic a priori proposition. Available: http://www.britannica.com/EBchecked/topic/578646/synthetic-a-priori-proposition. Last accessed: 02/04/15.

Guyer, P. (1987). Kant and the Claims of Knowledge. Cambridge: Cambridge University Press.

Guyer, P. (2006). Kant. London: Routledge.

Jacobi, F.H. (1912). David Hume über den Glauben, oder Idealismus und Realismus. Ein Gespräch. London: Garland.

Janiak, A. (2012). Kant’s Views on Space and Time. Available: http://plato.stanford.edu/archives/win2012/entries/kant-spacetime/. Last accessed: 17/04/15.

Kant, I. (1998). Critique of Pure Reason. 2nd ed. Cambridge: Cambridge University Press.

Langton, R. (1998). Kantian Humility: Our Ignorance of Things in Themselves. Oxford: Clarendon Press.

Matthews, H.E. (1969). Strawson on Transcendental Idealism. The Philosophical Quarterly. 19 (76).

McCormick, M. (2012). Immanuel Kant: Metaphysics. Available: http://www.iep.utm.edu/kantmeta/#H4. Last accessed: 14/04/15.

Michael, R. (2014). Immanuel Kant. Available: http://plato.stanford.edu/archives/sum2014/entries/kant/. Last accessed: 17/04/15.

Newton, I. (1999). The Principia: Mathematical Principles of Natural Philosophy. Oakland: California University Press.

Palmquist, S. (1986). Six Perspectives on the Object in Kant’s Theory of Knowledge. Dialectica. 40 (2).

Schulting, D. & Verburgt, J. (2001). Kant’s Idealism: New Interpretations of a Controversial Doctrine. New York: Springer.

Strawson, P.F. (1996). The Bounds of Sense. London: Routledge.

Van Cleve, J. (1999). Problems from Kant. Oxford: Oxford University Press.

Walsh, C.M. (1901). Kant’s Transcendental Idealism and Empirical Realism. Mind. 12 (48).

Wood, A. (2005). Kant. Massachusetts: Blackwell Publishing.



  1. Omar Lughod says:

    Suppose the thing in itself to be a mere “thought object”, “presupposed” by the appearance, hence only analytically thought therein. Then the empirical reality of the appearance takes on the entire burden of reality. The thing-in-itself becomes, in Kant’s terms, a mere negative, critical conception-what we cannot know. While in positive terms, it functions as a regulative/reflective notion- a placeholder, into which, in the practical sphere, we posit, justified by an unconditional (and hence unjustifiable) moral law – moral objects, such as God, freedom, the soul, the highest good. The thing in itself is here viewed as a mere critical object that closes down the traditional metaphysical routes while enjoining the farthest reaches of our moral will.

  2. danlanglois says:

    I’m familiar with the notion that a textual stalemate between two camps has evolved.  It has long been disputed whether Kant’s transcendental idealism requires two worlds – one of appearances and one of things in themselves – or only one. However, you geniuses are thinking about two numerically distinct entities, else two aspects of the same thing, when there is no real difference between the two views. They’re both wrong as Kant interpretation, because undertaken in this realist spirit where there are supposed to be possibilities for discerning the different categories of ‘reality itself’. Don’t look for this sort of ontological import, for your system of ontological categories. I can’t help but repeat something like ‘duh’, when I consider this as Kant exposition. The debate is about what sort of uniquely true and complete system of ontological categories Kant was offering, like he was that kind of comedian. Duh.

  3. I’m familiar with the notion that a textual stalemate between two camps has evolved.  It has long been disputed whether Kant’s transcendental idealism requires two worlds – one of appearances and one of things in themselves – or only one. However, you geniuses are thinking about two numerically distinct entities, else two aspects of the same thing, when there is no real difference between the two views. They’re both wrong as Kant interpretation, because undertaken in this realist spirit where there are supposed to be possibilities for discerning the different categories of ‘reality itself’. Don’t look for this sort of ontological import, for your system of ontological categories. I can’t help but repeat something like ‘duh’, when I consider this as Kant exposition. The debate is about what sort of uniquely true and complete system of ontological categories Kant was offering, like he was that kind of comedian. Duh.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s