Davidson on Conceptual Schemes
In “On the Very Idea of a Conceptual Scheme,” Davidson attacks the scheme/content division by way of questioning conceptual relativism. As he says, conceptual relativism is a “heady” notion wrought by an underlying paradox: “different points of view make sense, but only if there is a common coordinate system on which to plot them; yet the existence of a common system bellies the claim of dramatic incomparability.” His goal: tease out the sense in which this paradox is fatal versus the sense in which it is just interesting, and his crucial step is to ‘equate’ languages and conceptual schemes. In other words, Davidson turns the problem of relative conceptual schemes into a question of relative translation — or alternatively, into the question of translatability between languages, where partial failure versus complete failure dictates the sense in which schemes can be relative, or if even “relativity” can be said of them at all [1]. In the final analysis, his conclusion is pragmatic and agnostic: based on the question of translation, he finds “no intelligible basis on which it can be said that schemes are different,” but “neither can we intelligibly say that they are one.” As such, we should give up the very idea of the scheme/content division in order to “reestablish unmediated touch with the familiar objects’” of daily life that makes our “sentences and opinions true or false.”
Now the Anus likes unmediated touch with familiar objects, and it’s never been fond of “paradigms” and “schemes” either, except in a fast and loose approximation that dissolves on a closer look. Thus, the Anus likes seeing Kuhn — and by extension Rorty, though not mentioned here — taken to task by showing not just that but also how ostensibly ‘incommensurable’ notions in their own work, or in the work of others, are in principle declared “incommensurable” but in practice are taken as comparable, using independent but unacknowledged reference points, based on something — on some kind of common ground — even if a ‘view from nowhere’ or a ‘permanent neutral framework’ is rejected.
In other words, the Anus thinks Davidson is right to put some meat on the bones of the all-too-often invoked formal inconsistency of conceptual relativists and schemers, but it wonders how his anal-lysis fares with perhaps the greatest conceptual schemer of all — Kant. Specifically, does Davidson’s analysis of the scheme/content distinction, with its subsequent agnosticism, apply to Kant’s transcendental account of the subsumption of intuitions (the content) under the concepts of the understanding (the scheme), as this merger makes up experience? In another essay, “The Myth of the Subjective,” he indicates that it does, but that application bears examining.
On first blush, it seems unlikely the argument against conceptual schemes via the paradox of conceptual relativism will apply to Kant because Kant was not a conceptual relativist. In fact, his transcendental analysis is principally a response to conceptual relativism, and in his mind, the non-relativistic elements added to experience via the categories of the understanding is only antidote for it. But in actuality, like most, this first blush conceals a hidden intent as much as it reveals a denial, for it would apply to Kant if for him, though unstated, the “conceptual scheme” is all one — that is, if the constituting activities of the mind are universal to all rational creatures. On the flip side, though, it would not apply in the sense that for him, the conceptual scheme itself is decidable via the problem of translation. In other words, for Kant, all experience — for everyone — is lawful and orderly in the same sense, therefore all empirical claims are in the end ‘translatable,’ according to the same rules, i.e., they are for all intents and purposes completely translatable via transcendental rules.
So, it’s clear that for Kant there is a sense in which there is a universal, single conceptual scheme, but it remains unclear if the consequences of the dual arguments from partial or complete failure of translation would apply to him, because for Kant all relative empirical claims are completely ‘translatable’ into a transcendental grounding, as it were — or to put the matter more precisely, they are interchangeable and/or arbitratable against a neutral matrix, through a single set of rules, and that common ground is constituted experience, via transcendental operations of the mind. So, again, the question of whether Davidson’s attack on this “third dogma” of empiricism applies to Kant boils down to the question of experience, and more specifically, to how the “scheme/content” distinction in Kant relates to that experience.
Now Davidson distinguishes two ways in which a conceptual scheme relates to experience: it can organize experience or it can fit experience, the difference being the former systematizes and divides up “the stream of experience” while the latter predicts and accounts for particulars or events against the “tribunal of experience.” He then goes on to distinguish within experience between ‘organizing reality’ (the world, nature) and ‘organizing things within experience’ (such as surface irritations, sense data, etc.). Eventually, he rejects as problematic the two sense of senses of “organize” he describes, as well as the single sense of “fit.” Neither organize nor fit accounts for how a conceptual scheme relates to experience.
The details of his argument seem weak in some respects, especially regarding organizing experience, but their force can be granted without affecting an underlying point critical to his argument, or the argument here. Specifically, in both cases, Davidson is presuming that there is such a thing as experience to be organized or fitted in the first place.
This is crucial: in his analysis of how the scheme applies to contents given in experience, either as a single entity like “the world” or a plurality of objects like “objects in the world,” experience itself is taken for granted as always already there to be fitted or organized. In neither case does he go any distance toward making the possibility of experience as such problematic, and this failure bears directly on the applicability of his analysis to Kant because that is precisely what Kant did. His version of the “scheme/content” distinction works at that transcendental level; in other words, Kant was concerned with the possibility of experience as such, not per se with the ‘organization’ or ‘fitting’ of experience as first presumed, and this casts serious doubt on the applicability of Davidson’s analysis.
To put the matter more directly in the context of Davidson’s argument, Kant never argued that the categories (read “conceptual scheme”) apply to (read “organize” or “fit”) experience per se without simultaneously saying that the application of the categories to intuition is experience itself. Davidson misses this duality of application/constitution in his exploration of the scheme/content duality. As Kant states repeatedly, the ‘application’ of the categories ‘to’ intuition anticipates not just the organization of a given experience but rather represents “the form of possible experience in general,” such that “the principles of the pure understanding are nothing more than the principles apriori of the possibility of experience” itself. Or to put it in terms of Davidson’s closet illustration, the “conceptual scheme” for Kant doesn’t organize the closet taken as such, or even the closet taken as full of objects, as much as it makes possible the experience of a closet with objects in it that is as such organizable or fittable at all. Under this use, the “conceptual scheme,” as it were, is making coherent experience possible, and that logically precedes the application of a conceptual scheme in Davidson’s more restricted sense of application to actual experience.
This ‘making possible a closet with contents to be organized’ (in Davidson’s more limited senses) is clear in Kant’s summary statement, again repeated several times in the Critique as its principle conclusion: “the conditions of the possibility of experience in general are likewise the conditions of the possibility of objects of experience.” In this transcendental respect, Davidson’s arguments against scheme and content organizing or fitting experience overlooks the constitutive role Kant gives his “conceptual scheme,” and this omission renders his “On the very idea” analysis of the scheme/content dogma inapplicable to Kant, absent further argumentation. In other words, Davidson may have compelling objections to a “third dogma” of empiricism; it’s just that the ‘empiricist’ application is not Kant’s empirical realism, subtended by a transcendental idealism [2].
This rejection of Davidson’s application, however, is incomplete, for in one respect his argument both has a deeper force than so far indicated, even as in that respect it actually loses all of its force, as it might apply to any transcendentalist like Kant.
In the first place, under the stipulation that psychological nominalism is true, Davidson’s rejection of the scheme/content distinction via the route of failure or translatability is convincing. It would even extend, for example, to Sellars’ attempt to re-establish the distinction within psychological nominalism after purging it of the Myth of the Given. This applicability, then, increases its scope in a way not mentioned, even by Davidson.
In the second place, however, under the stipulation that psychological nominalism is false, which it is, Davidson’s argument loses all force against anyone who proposes, like Kant, that the inherent traits of a so-called “conceptual scheme” are already active in pre-linguistic experience — in other words, active in an experience rich in cognition that gets taken up and magnified into more directly “conceptual” schemes inherent ‘in’ language, after the fact. In Kant, this activity is implied in how the categories of the understanding structure experience itself, making that experience possible, which though not stated as such must occur prior to language. In this respect, elements of Kant’s conceptual scheme must be pre-linguistic.
Current candidates for this pre-linguistic ‘scheme’ include many — if not most — of Kant’s categories (though to be absent the epistemological function that Kant intended them). Causality and dependence, quantity, negation, inherence and subsistence, quality — these cognitions, along with many others have been empirically demonstrated not just in pre-linguistic infants but also in non-linguistic primates. Again, these ‘categories’ are non-epistemological, in Kant’s sense, but they nevertheless ‘constitute’ experience in a sense prior to a conceptual scheme that can be equated with language. Keeping with Kant’s terms, they are “transcendental,” at least with respect to language (and hence to the question of translation) because they are inherent to experience prior to its acquisition, even if for different reasons they don’t play the epistemologically constitutive role Kant assigns to them, in determining the possibility of synthetic apriori judgments.
Either way, ‘concepts’ like object permanence, inherence in substance, reciprocity, causality, negation, and such still ‘structure’ pre-linguistic experience, in that they are the organization of that experience; they make it what is it. In this sense, a ‘conceptual scheme/content’ dichotomy is active in Kant, but one immune to Davidson’s criticism of this third dogma of empiricism.
At the end of the day, then, Davidson’s arguments against the scheme/content dogma via arguments against conceptual relativism via the problems of translatability don’t apply to Kant because first, they only apply under psychological nominalism, and Kant is no psychological nominalist, and second, because ‘conceptually schematic’ elements like causality, negation, object permanence, quantity, etc. are empirically demonstrable in pre- and non-linguistic subjects; therefore the issue of translatability is completely impotent to arbitrate their relationship to experience, when experience is taken problematically, as it is in Kant.
In short, the “third dogma” as Davidson conceives it is specific to an empiricism that no more applies to Kant than it does to current empirical knowledge of what non-linguistic experience is actually like — in other words, to forms of modern empiricism or transcendentalism still in use. As such, Davidson’s rejection of the scheme/content distinction has no independent force outside a rather limited context of distinctions and arguments anal-lytic philosophers are so fond of making (which really is of no concern to anyone else) [3].
As a thorough pragmatist, the Anus applauds Davidson’s rejection of the very dilemma of an “interpreted” versus an “uninterpreted” reality in a return to an “unmediated touch” of familiar things, but he fails to see how it applies to an empiricism that conceives experience without the conceptual hobble-crutch that is psychological nominalism. Like Davidson, the Anus prefers its familiar objects neat, even if it is forced to drink from a different cup in a toast to a thinker like Kant.
[1] It remains unclear in just what sense conceptual schemes and language are ‘equated,’ beyond the stipulation that “associates having a language with having a conceptual scheme,” and presumably vice versa. The Anus happens to think this doctrine is problematic, though not necessarily fatally so, so the sense in which it is will not be raised yet and the stipulation will be granted on its face. Nevertheless, it seems possible that important elements of a “conceptual scheme” could be pre-linguistic, and that language magnifies this pre-linguistic cognition through the action of language-specific, cognitive mechanisms. This possibility would only seem impossible to a psychological nominalist, but that reluctance should be neither here nor there because psychological nominalism is false. The details of this pre-linguistic cognition have been explored in extensive empirical research on primates, infants, and young children, as well as the development of these forms in language acquisition and use during later development. This idea will be taken up shortly, as it pertains to Davidson’s point about scheme and content.
[2] This is not to say that Davidson’s argument couldn’t be extended to encompass a transcendental argument, but how it fails to do so as currently stated is developed next.
[3] Another way to contextualize the limits of Davidson’s critique: the criterion of translatability would apply only to an empiricism under which psychological nominalism is true, but since psychological nominalism is in fact false, his critique has little real independent applicability to empiricism as such, much less any force for exposing a “third dogma” within it.
