Gupta’s non-propositional conception of experience

Last week, I attend an excellent talk and seminar from Anil Gupta, author of Empiricism and Experience, in which he presented his non-propositional conception of experience and its contribution to rational thought, along with his latest thoughts on the metaphysics of experience.

Gupta’s central thesis can be simply stated as the claim that rational entitlement to knowledge is not generated solely by present experience, but by experience in conjunction with the prior metaphysical and experiential standing of the subject. It is this pre-existing ‘view’ that the subject brings to experience that yields entitlement to knowledge, thereby enabling them to make rational judgements even in cases where their ontology is radically mistaken; e.g. they are a ‘brain in a vat’ (BIV). This contrasts with views, such as Pryor’s dogmatism, that take experience itself to confer prima facie rational entitlement, and Wright’s entitlement of cognitive project, which flows from the logical structure of rational enquiry.

Gupta’s conception of experience has some novel and interesting features. Firstly, its contribution to rationality—what Gupta calls the ‘given’—is non-propositional in nature. In contrast to content or representation based views, experience is seen as analogous to a function or ‘argument schema’, such as modus ponens, which maps subjective views onto judgements. Thus experience dictates which judgements are warranted by a given view, and is in turn conditioned by past experience and beliefs, thereby creating a cycle of mutual interdependence. This logic of interdependence is described by Gupta’s work with Belnap, which is supposed to avoid the obvious problems with circularity, although not having read this work I can’t really comment on the details.

The resulting theory aims to remain entirely neutral on ontological issues whilst enabling experience to make a rational contribution to knowledge despite its non-propositional structure. Thus, a sense datum theorist, who takes experience to consist only of private sense impressions rather than public objects, and a direct realist can be having the same experience, irrespective of which ontology is correct. Indeed, Gupta considers it an advantage of his account that it does not mandate or presuppose the correctness of a common sense ontology of objects, persons, and so on, on the basis that the epistemology of experience shouldn’t constrain one’s choice of metaphysics. The sense datum theorist’s ontology may even turn out to be right, and the issue of what things are to be found in the world is taken to be a purely empirical matter.

Up to this point, I find Gupta’s view to be extremely interesting and appealing. It avoids the need to assign particular propositional content to experience—something that Travis, for example, regards as highly problematic—whilst retaining its essential role in rational enquiry. I also like the way in which it takes account of the mutual interaction between present experience and the pre-existing ‘view’ that a subject’s brings to bear upon the world in a way that respects the temporally extended nature of experience. Gupta’s logic of interdependence seems to offer a powerful way of explaining how experience both informs judgement and our overall world view, which in turn entitles us to make certain judgements, rather than presenting a purely synchronic picture of entitlement, as some other theories (e.g. dogmatism) attempt to do.

What I find less convincing, however, is Gupta’s characterisation of the nature of experience. In the first place, it seems counterintuitive to describe the solipsist or BIV as having the same—i.e. ‘subjectively identical’—experiences as the direct realist who has a radically different ontology. Experience is surely theory laden in the sense that it can be affected by the concepts that one possesses, and not just in terms of the judgements that it entitles one to make.

Gupta also differentiates between subjective identity, which is determined by the things (real and unreal) that experience makes present to the subject, and indistinguishability, which is limited by the subject’s powers of discrimination. This introduces an externalist element to experience that sits uncomfortably with the kind of empiricism that Gupta wants to defend. Gupta’s principle of ‘same phenomenology, same rational contribution’ seems clear enough, but if the phenomenology of experience is not dictated by an experience’s subjective character but by the objects which it makes present to the subject (i.e. subjective identity), then two subjectively indistinguishable experiences can still make different rational contributions to knowledge—the very point that an empiricist might want to deny.

Another worry concerns Gupta’s argument against the propositional given. He argues that the propositional given commits one to a Cartesian view of experience, which consists only of sense data, and not of objects in the world, and so is unable to yield entitlement to knowledge. However, the premises upon which this argument—and indeed, the whole book—is based seem question begging. A epistemological disjunctivist, such as McDowell, would simply reject the premise that phenomenologically identical experiences always yield the same rational contribution to knowledge, Gupta’s so-called equivalence principle. The notion that experience is always reliable and can never mislead also seems poorly motivated, not to mention counterintuitive, although in conversation Gupta attributed it to the empiricist tradition, and so worthy of defending. Furthermore, it seems that Gupta’s own account could itself be formulated in terms of a propositional given consisting of as a (potentially infinite) series of conditionals connecting views and rational entitlements to judgements without being committed to the Cartesian ontology. Although Gupta himself would no doubt eschew such a formulation, it is unclear why the notion of a propositional given is necessarily problematic, thereby undermining the motivation for Gupta’s non-propositional account.

On the whole, I find Gupta’s approach to be a refreshing and novel solution to a variety of problems in epistemology and the philosophy of perception. But his unwillingness to countenance the notion of direct acquaintance with objects (as opposed to sense data) and adherence to questionable methodological principles seem to be motivated more by a desire for logical tidiness and consistency than phenomenological accuracy. Perhaps this is no bad thing, and the resulting approach is certainly more elegant than some of the alternatives, but I find myself unconvinced as to its merits, despite being intrigued by the possibility of this kind of account of perceptual knowledge.

About these ads