Kant, Hegel and the puzzles of McDowell's philosophy

In his famous book, Mind and World [1994/2003], John McDowell offers an original description of the relation between mind and world by focusing on the human experience of the external world. Trying to avoid an “oscillation” between opposite theories of human experience of the world, the fundationalist (the “Myth of Given”) and the coherentist, he argues that we should describe experience as conceptualized in terms of the Kantian categories of receptivity and spontaneity. These Kantian categories leave no gap between mind (thought) and world (experience) as human conceptual capacities (spontaneity) become involved in all experience of the external world (receptivity). McDowell seems to strengthen his notion of conceptualization of experience by explicit radicalization of this Kantian framework. He explicitly admits his commitment to Hegel’s view on the unboundedness of the conceptual content.

The paper seeks to understand a proper motivation for McDowell’s interest both in Kant and Hegel. It reconstructs his arguments in favour of the Hegelian notion of conceptualized experience, and shows how it affects his reading of Kant’s transcendental philosophy. It concludes with a draft comparison of McDowell’s position on experience with the Hegelian one by pointing out the most important difference as regards the notion of factivity.

I. Kant’s Problem

With recourse to Sellars [1956/1997], McDowell locates his interest in Hegel in the context of an attempt of avoidance of the “Myth of the Given”, which contains a proper motivation for the Kantian notion of conceptualized experience. McDowell explicates:

Sellars represents “Empiricism and the Philosophy of Mind” as one battle in a general campaign against what he calls “the framework of givenness” […]. He mentions Hegel in the first sentence of his paper, offering “immediacy” as a Hegelian equivalent for “givenness”, in the sense he wants to expose as mythical […]. In this connection he describes Hegel as “that great foe of ‘immediacy’” — though he qualifies this by saying that he thinks not even Hegel is completely emancipated from the framework of givenness. (I do not know what he means by this hint at a criticism.) Later he imagines an interlocutor who describes what Sellars is engaged in as “incipient Méditations hégéliennes” […]. It is clear that though he does not appeal to specific texts, Sellars takes his campaign against the Myth of the Given to be Hegelian in spirit (McDowell [2003] p. 2).

Sellars indeed stresses the scope and variety of conceptions of “givenness”. He notes: “Many things have been said to be ‘given’: sense contents, material objects, universals, propositions, real connections, first principles, even givenness itself” (Sellars [1956/1997] p. 14). As rightly diagnosed by McDowell, Sellars constructs a position according to which even the most basic perceptual knowledge requires conceptual capacities. “Even for knowledge directly acquired in perception, [Sellars] insists on conceptual mediation, with substantive knowledge already in place” (McDowell [2003] p. 3).

McDowell’s Mind and World observes Kant as proposing that there is no empirical immediacy in a proper understanding of spontaneity and receptivity. In other words, McDowell portrays Kant as claiming that there is no experiential intake without conceptual mediation. McDowell makes the substantial objection that Kant’s theory does not realize the “full promise” of its conception of experience.

He argues that Kant, by situating his conception of experience within the framework of a “transcendental story” or “transcendental idealism”, collapses into a kind of “Myth of the Given”. While Kant avoids empirical immediacy, he still offers a “transcendental givenness” in his theory. McDowell criticizes Kant on the grounds that “[f]or a whole-hearted exorcism of the framework, we need to eliminate the transcendental givenness too”, and furthermore confesses that

[t]his is the context in which I found myself saying things that sounded Hegelian. It might seem that absolute idealism, a more whole-hearted idealism than transcendental idealism, would be an unpromising refuge for a conception of perceptual experience whose point is to exemplify how subjective states can make gap-free contact with genuinely objective reality. But I suggested that Hegelian rhetoric can be domesticated. It can be interpreted as expressive of a philosophical outlook that is precisely protective of the ordinary realism of common sense (McDowell [2003] p. 5; see also [2001] p. 529-530)^[1].

Though it appears uncharacteristic for somebody who claims to avoid an idealist charge, McDowell seems to claim that transcendental idealism undermines Kant’s aims, and that Kant’s view of experience requires a reworking in Hegelian terms. Moreover, according to McDowell, Hegel is “not hostile to the very idea of sensory consciousness of the objective”, and any contradictory claim is just “a misreading of Hegel, in particular of the structure of the Phenomenology of Spirit” (McDowell [2003] p. 5; see [2001] p. 548). He maintains that:

Consciousness, which is as such other than its object, does not disappear from the Phenomenology after the section explicitly devoted to it. Hegel’s aim is to “sublate”, aufheben, that otherness — to do away with the gap it can seem to open, while preserving the otherness as an element in a more comprehensive picture, in which it no longer has philosophically damaging consequences. The Phenomenology does not advocate discarding the concept of consciousness of objects. It undertakes, among much else, to rehabilitate the concept, a project Kant undertakes with only partial success (McDowell [2003] p. 5).

As aptly observed by Haddock, McDowell explicitly claims in his recent writings that “transcendental idealism usurps the prospect of a realism/idealism coincidence, which we need to go the way of ‘Kant’s successors’ to secure” (Haddock [2008] p. 90). Furthermore, McDowell indicates that seeing Kant “in this light is a useful way to make less programmatic sense of one aspect, at least, of Hegelian thinking” (McDowell [2003] p. 6). McDowell reads Kant through Hegel, and Hegel through Kant.

Interestingly, McDowell’s central self-criticism of Mind and World is that he failed to take into full consideration Kant’s view of “things as objects of experience, and those same things as things in themselves” (Kant [1781/1929] p. Bxxvii; see also Allison [1983] p. 290). In his most recent writings [2001, 2003, 2008], he insists that Kant accepts things as they are given to our senses as things in themselves. At the same time, McDowell makes a caveat that to describe such things as they are given to our senses is to describe them in “terms of the relation they bear to our sensibility” (Haddock [2008] p. 90). Eventually, he concludes that it is transcendental idealism which caused Kant to abandon this conception in favour of the notion of unknown things in themselves.

McDowell quotes Kant’s observation that

[t]he same function which gives unity to the various representations in a judgment also gives unity to the mere synthesis of various representations in an intuition: and this unity, in its most general expression, we entitle the pure concept of the understanding” (Kant [1781/1929] p. A79/B104-105).

In McDowell’s reading of this passage, Kant claims that intuitions (as “cases of sensory consciousness of objects”) have logical structures identical with logical structures of judgments:

We get Sellars’ version of the thought, which he put with the metaphor of experiences containing propositional claims, if we replace judgments, in this Kantian identification of functions, with claims – as befits the fact that between Kant and Sellars comes ‘the linguistic turn’ (McDowell [2003] p. 6).

This remark is designed by McDowell to understand the Transcendental Deduction of the Categories. Namely,

[h]is aim there is to demonstrate the ‘objective validity’ of the categories, the pure concepts of understanding, by showing how they figure in a conception of experience on which it comprises intuitions, cases of sensory consciousness of objective reality (ibidem).

In McDowell’s reading of Kant, conceptual capacities have their primary actualization in judging. “Judging is making up one’s mind about something” (ibidem, p. 7). In other words, to judge is to engage in free cognitive activity, and due to that engagement, freedom is central to Kant’s picture of conceptual capacities. Human beings are both experientially in touch with “objective reality” and they are able to make judgements about it. This dualistic understanding about experientiality and judgement is inherent in the structure of Kant’s Critique of Pure Reason. Beginning with the Transcendental Aesthetic which concerns sensible intuitions (and their pure sensible forms), it subsequently proceeds to the Transcendental Analytic which in turn concerns the understanding in general (and its pure concepts in detail). McDowell proposes that Kant’s expositional order makes

[...] it look as if there are two independent sets of conditions [i.e., those of our sensibility (intuitions) and those of the understanding], as if the forms [of our sensible intuitions] are independent of the synthetic powers of the understanding (ibidem; see McDowell [2001] p. 531-533).

According to McDowell’s delineation, Kant conceives sensibility as entering into the constitution of our experience and generating a serious threat. That is to say, conditions for objects to conform to the understanding are subjective conditions, “in the sense that they are requirements of the thinking subject” (Haddock [2008] p. 90). Yet Kant seems to McDowell to overstate this point. One of these conditions is that objects must figure in items with a certain kind of unity; the unity of thinkable contents to the effect that things are thus and so. However, “conditions for objects to be conformable to the understanding – to be thinkable – are not as such conditions for objects to be able to be given to our senses” (McDowell [2003] p. 9). As Haddock ([2008] p. 90) astutely remarks: “given how Kant conceives of objects as they are so given, it looks as if getting objects themselves to figure in thought involves the imposition of this subjective condition onto reality that does not conform to it”.

Yet it looks like this condition of unity is something “over and above” the conditions for objects to be given to our senses. Moreover, insofar as it can be given to our senses, the reality of objects cannot be something “super-sensible”, but, at the same time, it is outside the “space” of conceptually organized items (subjective objects). The threat becomes that “the very conceptual unity that makes thought thought ensures that no thought can capture reality »as it really is, without distortion«” (Child [1994] p. 56).

According to McDowell, Kant is aware of this danger of “a sideways-on view” that places empirical reality beyond the space of concepts. This awareness likely prompted Kant to seek to establish subjective conditions that grant unity of thought, which might be both “conditions for objects to be in conformity to the understanding”, and “conditions for objects to be given to our senses”. McDowell writes:

Kant rewrites the Transcendental Deduction in a way that seems designed to avert this threat. The essential move is to correct the impression that the Aesthetic offers independent conditions for objects to be able to be given to our senses ([2003] p. 9-10, see also [2001] p. 527-531).

Kant was ultimately aiming to “establish an equipoise between subjective and objective [through an] idea of conditions that are subjective and objective together” (McDowell [2003] p. 10). Conditions of the understanding are not only primarily subjective due to the fact that they are requirements of the subject, but they are also objective conditions as they are requirements for objects “to be given to our senses”. In McDowell’s view, the latter requirements, presumably, are also requirements that hold of things in themselves in Kant.

It should be emphasised here that the latter requirements seem to follow the general framework of the position that McDowell advocates in Mind and World. In his articulation of conceptualized experience, he insists that locution in items with conceptual content is both a condition for objects to conform to the understanding, and a condition for those same objects to be given to our senses (see McDowell [1994/2003] p. 3-23, [2003]).

II. The Hegelian Solution

In recent papers [2003, 2008], McDowell seems to be more and more worried about the danger of “a sideways-on viewpoint”. In fact, he was already aware of it in Mind and World, as we can see by his invocation of the “Hegelian spirit” as regards the notion of conceptualized experience (see McDowell [1994/2003] p. 44-45). Nevertheless, he seems to be more preoccupied with the danger in his later writings, seeking to remove this by means of the modified Kantian equipoise. In my opinion, his strategy offers a suitable opportunity for understanding why McDowell regards Hegel’s solution to Kant’s problem as the only plausible manoeuvre within the framework of the conceptualisation of experience. The extent to which his own position is a Hegelian one (also as regards the possible idealistic implications), we will see in the next section. At present we must turn to McDowell’s conceptual machinery which justifies the “Hegelian turn” of his project.

McDowell reads Kant’s position on sensibility as if it is constrained or shaped in specific kind of forms. Indeed, Kant offers a description of sensibility that is constructed “spatially” and “temporally”, and as such, it correlates with the “formal intuitions” of space and time (see Kant [1781/1929] p. B34/A20–B73; McDowell [2001, 2003]). As objects, these formal intuitions conform to the requirement of the understanding. But, as McDowell puts it, “sensibility as such does not have to be spatially and temporally formed” (ibidem, p. 9), for “[t]here may be differently formed sensibilities, which unite their manifolds into different formal unities” (Haddock [2008] p. 91). McDowell further stresses:

Kant contrives to represent the combination of manifolds into the formal intuitions, space and time, as a case of the kind of unity that is not intelligible except in the context of the freedom of judgement. But he depicts the fact that it is space and time in particular that are the formal intuitions answering to the form of our sensibility as a mere peculiarity of our sensibility, not an attunement of it to the way things anyway are ([2003] p. 12).

Within the proposed Kantian equipoise, “conditions of the understanding just are conditions of our sensibility” (ibidem).

This picture is problematic insofar as it supposes one set of subjective and equally objective conditions, that is, the conditions that are viewed as both conditions of sensibility and understanding. Incongruously, McDowell states in his proposal that “the fact that our human sensibility has a spatial and temporal form is not a condition laid down by the understanding; it is simply a condition of our human sensibility” (Haddock [2008] p. 91). In other words, it is not a condition for objects “to be thinkable per se”, but rather a condition for objects “to be given to us”, and, presumably as such, “to be thinkable by us”. I suppose that it is a main reason why McDowell thinks that Kant’s “transcendental story” or “transcendental idealism” makes the Kantian equipoise “unbalanced”: there is some or at least one condition of our sensibility that is not a condition of the understanding. In the Kantian picture there is something on the side of our sensibility (intuitions), which is not present on the side of the understanding (spontaneity).

Perceiving Hegel’s role in this particular context for McDowell’s project is crucial. McDowell himself enjoins us to do so:

[T]here is an incipiently Hegelian flavour to Kant’s idea of conditions that are indifferently subjective and objective. And Kant’s thinking embodies a partial counterpart to Hegel’s focus on freedom, in the central position Kant attributes to the unifying powers of the spontaneous understanding. What spoils Kant’s approximation to a Hegelian equipoise between subjective and objective is that he sees our cognitive freedom as constrained, from outside, by the specific forms of our sensibility, which he leaves looking like a mere brute fact about the shape of our subjectivity. As I have urged, the result is that the objectivity he contrives to credit to the requirements of the understanding looks like an impostor. […] It seems reasonable to conclude that in order to get genuine objectivity into a descendant of Kant’s picture, we need to leave nothing outside the scope of our cognitive freedom. We need to stop seeing spatiality and temporality as immediate givens. What figures in Kant as a constraint on the understanding from outside needs to be reconceived as no more than an element within the free self-development of reason ([2003] p. 14).

Subsequently he explicitly adds that “[t]hat is the sort of Hegelian language that, as I put it before, needs to be domesticated” (ibidem; emphasis added). Interestingly, he seems to think that this “domestication” can be achieved by the Kantian reading of Hegel, which is supposed to suggest, “how we might begin to make sober sense” of Hegel.

McDowell suggests that “[t]he real trouble is the way Kant conceives the forms of our sensibility as mere givenness” (ibidem). It leads to a situation that the elements of “objective reality” – that we have contrived to see experience as taking in – are “mere reflections of an aspect of our subjectivity that we cannot understand”. Cognitive freedom (conceptual capacities) therefore becomes constrained. According to McDowell, a place of need for a Hegelian solution is opened up by way of talking about “the free self-development of reason”, that is, by granting its conceptual freedom.

Surprisingly, McDowell does not think that this Hegelian manoeuvre can lead to another danger, the threat of reconstructing objective reality as “a mere projection from utterly unconstrained movements of the mind” (ibidem). It is rather more natural to think that Kantian pure forms of sensibility might be seen as external to the spontaneity of the understanding, and in that way protect Kant from “deprecation of the world’s independence” (Friedman [1996/2002] p. 439-444). McDowell apprehends this difficulty, but dismisses it, reasoning that

[s]o far from ensuring a common-sense realism about objective reality, Kant’s framing his attempt to vindicate objective validity for the categories within the doctrine that space and time are transcendentally ideal is just what ensures that what the categories are shown to have is not recognizable as genuinely objective validity (McDowell [2003] p. 15).

What Hegel rejects is a conception of empirical knowledge as the outcome of understanding constrained from outside by pure forms of sensibility. He instead develops for McDowell a “new approach” to the problem, in which “empirical thinking is responsive to empirically accessible reality” by “the free self-development of reason”, which is something more than Kant’s understanding. “We might think of the independent layout of the world, rather, as the medium in which the freedom of reason is exercised” (ibidem). Paradoxically, McDowell finds that the presence of things in themselves in Kant’s proposal “spoils its claim to credit our empirical thinking with genuine objectivity” (ibidem).

Following Allison ([1983] p. 13), McDowell perceives Kant’s transcendental idealism as a doctrine which distinguishes “conditions of the possibility of knowledge of things” from “conditions of the possibility of the things themselves”. Regarding the need for an alternative to such a position, Allison and McDowell differ.^[2] In McDowell’s view [2001, 2003], only Hegel offers a satisfying alternative account to the Kantian picture of experience, successfully providing a proper understanding of the conceptualization of experience based on only one set of conditions: “[t]he conditions [which] are inseparably both conditions on knowledge and conditions on objects, not primarily either the one or the other” ([2003] p. 17).

While Kant’s conception of experience “as actualisation of conceptual capacities in sensory receptivity” moves in the right direction in thinking about our way of experiencing the world, it ultimately fails according to McDowell, by not enabling us “not to find philosophical mystery in the idea that conceptual capacities reach all the way the world” (ibidem, p. 18), though originally it aims to do so. It is the case as the Kantian things in themselves are eventually nothing more than further objects above those that figure in our world view, and that the Kantian experience, “in so far as it presents us with a world that is spatially and temporally ordered, falls short of disclosing things in themselves to us” (ibidem). For this reason McDowell interprets his own proposal in the “Hegelian spirit” in the respect that Hegel furnishes a suitable correction to the Kantian picture. For McDowell, Hegel eliminates the “mere givenness of spatiality and temporality” from the Kantian view, in order to “yield an unqualified form of the thought that conceptual capacities make gap-free contact with objective reality” (ibidem). The Hegelian conception extends beyond the Kantian “in exorcising the framework of givenness” (ibidem).

After presenting McDowell’s interpretation of Kant and Hegel, and his motivation for claiming to be a Hegelian as regards the notion of conceptualized experience, it is time to examine their relation in more detail. Let’s turn in the following section to a comparison of McDowell’s and Hegel’s proposals on the conceptualization of the experience of the world.

III. McDowell and Hegel in Contrast

In order to avoid any undesirable affiliation with the foundationalist “Myth of the Given”, McDowell circumspectly claims that conceptual capacities which belong to “spontaneity are already operative in receptivity”. Experience is not a result of our minds (thoughts) acting on something given. He sustains the view that “states or occurrences of sheer passivity” (i.e. receptivity) are states or occurrences in which conceptual capacities are operative (ibidem: 26, 29, see also [2002] p. 291). In consequence, conceptual capacities, as Houlgate enjoins it, “are not put to work ‘on something independently supplied to them by receptivity’” ([2006] p. 252; see McDowell [1994/2003] p. 61), but rather, “conceptual content is already borne by impressions that independent reality makes on one’s senses” (ibidem, p. 67). Due to the fact that conceptual capacities are exercised in receptivity, experience does not in any way shape or transform pre-conceptual deliverances of sensibility.

Ontologically it might be described in the following way. What we take in from the world in experience are perceptible facts, such as the fact that things are thus and so (see ibidem, p. 26). It implies, then, a conception of perception as a factive state, not mental state. (It seems to have an application to his disjunctive theory of perception elaborated in other places).^[3] Such a factive state is constituent to a conceptual content of experience, and as such, it also becomes a content of normative judgments. In virtue of the above proposed operation of conceptual capacities in receptivity, McDowell aims to secure the openness of experience “to the layout of reality”, for “in enjoying an experience one is open to manifest facts, facts that obtain anyway and impress themselves on one’s sensibility” (ibidem, 26, 29). “[E]xperience at its best [is] openness to how things are” (McDowell [2002] p. 291). Therefore,

[t]hese facts that are taken in in experience as “thinkable contents” constitute “norms for belief”. That is to say, “the fact that things are thus and so […] equip[s] one with a warrant for believing that things are thus and so” (Houlgate [2006] p. 252).

As Bernstein notes,

[t]his is precisely the reading that Hegel gives to the Kantian distinction of intuitions and concepts, receptivity and spontaneity. McDowell’s statement of the relation of receptivity to spontaneity helps us to understand what Hegel means when he speaks of a distinction that is no distinction. Experience […] is of such character that it is “always already” constituted by conceptual capacities. There is no receptivity where spontaneity is not already at work ([2002] p. 13; see also Brandom [2002] p. 182-187).

Yet it seems to me implausible to present McDowell’s project in these terms. McDowell strives to distance himself from idealism, but if no escape from the conceptual is possible, what constitutes the difference between what McDowell is affirming and the Davidsonian coherentism (idealism) to which he objects? McDowell in Mind and World (Lecture III) is perfectly aware that this sort of objection might be directed against him, causing him to qualify his claim in a way which eventually differentiates him from Hegel’s notion of conceptualized experience.

In my opinion there is a pivotal disjunction between two pictures that might be observed when we look in more detail at McDowell’s and Hegel’s frameworks. I follow here some general considerations of a Hegel scholar, Houlgate (2006), on the topic. In short, we may say that for McDowell, the world “exercises authority over thought through perceptual experience”, while for Hegel, the world “exercises authority over our perceptual experience through thought” (see Hegel [1801/1977] p. 93, [1807/1988] §36/28; [1817/1991] §41). “Thought is the authority that ensures that our perceptual experience is of the world, not the other way round” (Houlgate [2006] p. 254).

How can we understand this difference? Hegel’s picture appears a much broader project than that described by McDowell (see Hegel [1825-6/1985] p. 69). Hegel seems to sustain simultaneously the two views that McDowell regards as incompatible. McDowell emphasises that “in judgments of experience, conceptual capacities are not exercised on non-conceptual deliverances of sensibility. Conceptual capacities are already operative in the deliverances of sensibility themselves” (McDowell [1994/2003] p. 39). Yet Hegel, holding both view-points, as far as I can see it, determined that conceptual capacities are operative in receptivity (see Hegel [1830/1970] p. 100; see also Craig [1987/1996] p. 207). At the same time he assumed that our understanding works on the non-conceptual deliverances of sensibility (see Houlgate [2006] p. 242-243; see also Hegel [1830/1970] p. 104). However, he qualifies this latter assumption by stating that human beings do not perceive bare sensations: “for Hegel, we see nothing without understanding it to be some determinate (or indeterminate) thing” (Houlgate [2006] p. 252). On one hand, then, Hegel seems to accept that understanding is “irreducibly operative” in receptivity, but on the other, in being so operative, “it make[s] experience of an objective world out of items that are in themselves less than that” (ibidem; see Hegel [1830/1970] p. 81; Craig [1987/1996] p. 189).

It is worth adducing Houlgate’s rationale on this point:

In Hegel’s view, our conceptual capacities are drawn into operations in receptivity in the sense that nothing is received into the conscious mind without their operation. Through their operation, however, we actively posit what we see as something objective. When drawn into operation, therefore, our conceptual capacities do not themselves become ways of being receptive: they do not let us take in more than is supplied by sensations, let us take in facts as thinkable contents” ([2006] p. 252-253).

In other words, for Hegel, we can never “take in” facts or conceptual content, because human beings simply do not take in anything other than sensory content. This sensory content we actively understand (and judge) to be something objective, which is there before us.

In my opinion, the difference relies on the divergent conceptions of the explanatory role of experience and the role of judgments within it. According to McDowell, judgment is the activity of “freely making up one’s mind that things are thus and so” ([1998] p. 439). This freedom is the reason “judgements […] [are] actively exercising control over one’s cognitive life” (ibidem, p. 434). “[I]t simply endorses the conceptual content […] that is already possessed by the experience on which it is grounded” (McDowell [1994/2003] p. 49), while at the same time, “there is a disconnection between perceptual experience and judging” (McDowell [1998] p. 439). For Hegel, “active judgement and understanding are constitutive of our experience, for without them we would not experience a world of objects at all” (Houlgate [2006] p. 253). In the final analysis McDowell modifies his picture, though possibly inspired by Hegel, rather in terms of Wittgensteinian facts than world objects, by which Hegel modifies in turn. For McDowell, perception is simply factive at face value; it is not a kind of relation to sensible objects. For Hegel, we do not take in facts, we actively judge objects to be thus and so. Factivity is secondary.^[4]