phenomenological critique of representationalism
Rejection of the notion that representational states define and explain the most basic kind of human interaction with the environment. See also representation, phenomenology, intention-in-action, Background, Hubert Dreyfus.

Although the phenomenological critique of representationalism has its roots in the writings of the Continental philosophers Heidegger and Merleau-Ponty, its importation into contemporary philosophy of mind has been accomplished largely by Hubert Dreyfus. The critique's basic position involves a rejection of the proposition that the fundamental relation of a person to the world consists in the relation of the content of an individual mind to the world of objects, events, and states of affairs as represented by that content. Instead, it is held that the most fundamental variety of human action consists in the apparently unthinking, skilled action that makes up much of our everyday activities, and that does not require mental guidance or intervention for its successful accomplishment (Dreyfus 1991, p. 52).
The evidence on which the phenomenological critique is built derives largely from the first-person experiences associated with nondeliberate action. Dreyfus, writing in collaboration with Jerome Wakefield, argues that when we engage in much of the day-to-day activity at which we are routinely skilled, we just do whatever it is we are doing, in unthinking response to the "moment-to-moment local forces acting upon" us (Wakefield and Dreyfus 1991, p. 263). As Dreyfus puts it elsewhere, our first-person experience of such action is of a "steady flow of skillful activity" in which we are "simply solicited by the situation to get into equilibrium with it" (1996, paragraph 34).
Although Wakefield and Dreyfus do ascribe a variety of intentionality to the person engaged in nondeliberate action, they label it "G-Intentionality," where the "G" stands for gestalt. ("G-Intentionality" is contrasted with "R-Intentionality," which is described as a representationally-mediated form of intentionality (Wakefield and Dreyfus 1991, p. 267).) The claim is that G-Intentionality involves the representationally-unmediated "direct responses [of the agent's body] to familiar perceptual gestalts" (1991, p. 265). The first-person experience associated with G-Intentionality is thus described as consisting in both "an experienced causal connection" between the environment and the agent's activity, and in "the feeling [that] our motion `fits' or is `appropriate'" to the situation (Wakefield and Dreyfus 1991, p. 269). Wakefield and Dreyfus assert that despite the richness and evident complexity of the skilled activities associated with G-Intentionality, they apparently are executed and maintained without benefit of "representational monitoring" (1991, p. 264). Rather, such maintenance is attributed to the "tendency of the muscular-perceptual system to promote certain _gestalts_ (or the analogous processes in intellectual functioning)" (1991, pp. 267-268).
According to Dreyfus, representational states do come into play in everyday activities when our skillful coping somehow breaks down due to unexpected obstacles, malfunctioning equipment, or any of a number of other circumstances which might intervene to prevent the smooth execution of our task. In that case we have a problem to confront and solve, and it is then that "a conscious subject with self-referential mental states directed toward determinate objects with properties gradually emerges" (Dreyfus 1991, p. 71).
The phenomenological description of what it is like to engage in everyday skillful action rings true. In terms of our actual experience of what it is like to do something we are skilled at doing, the appeal to a mediating representational mental content does seem superfluous. It is worth noting, however, that Searle (1991), drawing largely on the same first-person experiential base as Wakefield and Dreyfus, comes to the very different conclusion that a variety of representational intentionality is involved, even if we factor in the importance of presumably nonrepresentational background skills and capacities (see intention-in-action, Background). And once we attempt to explain, rather than simply describe, G-Intentionality, we do seem to require recourse to some sort of representationalism.
For it would seem that the closely related capacities of pattern recognition and perceptual discrimination, both of which figure prominently in G-Intentionality, far from involving "direct responses [of the agent's body] to familiar perceptual gestalts," rely instead on a high degree of abstraction. Studies in artificial grammar learning and the related task of anagram problem solving (in which the anagrams are based on artificial grammars) indicate that pattern-detection relies on abstract knowledge independent of, even when induced from, the particular stimuli to which subjects are exposed (Reber 1993, pp. 54-55). By the same token, perceptual discrimination, which involves the ability to detect similarities and differences among entities, seems to require access to a related variety of abstract information. As recent studies of how people construct analogies show, similarity judgments are not a simple matter of comparing entities' surface features, but instead involve the recognition of correspondences in the systematic or structural relations of those features (Kitcher 1996, p. 62).
The kind of abstraction involved in these capacities captures structural relationships independent of any given embodiment in particular instances. It is this very independence that strongly suggests a representational explanation, for such independence raises what might be thought of as a problem of location. If the relevant abstract structural relationships are not located in the particular instances exemplifying them then they must reside instead within the person as an induced, and necessarily internal, state or structure of some sort. Abstraction is an artifact of the human cognitive capacity, in other words, and as such can be expected to reside in the cognitive architecture. And by at least one viable definition, that is precisely what a mental representation is: an artifact of the human cognitive capacity resident in the cognitive architecture. Thus describing as representational in nature the embodiment and availability of the type of information implicated in G-Intentionality simply follows from what abstraction and mental representation are.
The ascription of representational status to G-Intentionality is an inference from the relevant facts, phenomenological and otherwise. However, it seems a plausible inference to make. In saying this, we are offering what is in effect a variety of the "what else" argument -- that is, what else can account for the sophisticated capacities involved in absorbed coping if not a cognitive system dealing in representations of abstract information? But the case for doing so is compelling because it takes up the burden of explaining how nondeliberate action can succeed. For even if, with Dreyfus, we reject the proposition that all successful action requires deliberatively attended-to mental direction (in the form of, e.g., rules recited or a meditation on the specific conditions the agent's behavior must meet), we still need to explain how it is that the capacities associated with G-Intentionality are afforded. Such an explanation would seem to require that our action, no matter how absorbed, be underwritten by some kind of mechanism that -- because it deals in highly abstract content -- ultimately is representational.
Daniel Barbiero