This paper draws a limited parallel between Dewey's naturalistic approach to habits, conduct and environment in Human Nature and Conduct (1922), on the one hand, and Merleau-Ponty's phenomenological descriptions of “milieu”, behavior and habit, on the other. Without minimizing the irreducible distance between the two philosophical approaches, I will focus on habit-forming within perception, arguing that a cross-reading of the two authors from that specific angle might be fruitful.

Dewey's anthropology attributes a pervasive role to habits, due to their dynamic, functional, social and integrative meaning – which I shall briefly recall. The author also emphasizes the reciprocity and complementarity of the organism-environment transaction; as they “incorporate” environmental conditions, habits would also modify them (“they are adjustments of the environment, not merely to it”, 1922: 38). Moreover, he underlines the variability and plurality of both the moving organism and its changing environment (1922: 104-105).

Yet the openness and creative tendency of acquired habits seems at odds with their primary conditions of emergence. Habits appear to stem up, mainly, from the social shaping of “impulses” or “instincts”; but “impulses” (the same, or other ones?) also combine with habits in order to modify and enrich them. Hence the duplicity of habits, surrounded by different sorts of practical solicitations and possibilities of innovation that might not be perceived: since “individuals form their personal habits under conditions set by prior customs” (1922: 43), and “all habit involves mechanization” (1922: 50), Dewey observes the dual (though not dualistic) nature of “intelligent” and “routine” habit.

Thus, locating our acquisition of habits within experienced, built and inhabited environments, several parts of the description are missing. Human dependency upon social and spatial contexts, inherited from infancy and childhood, calls for a more detailed appreciation of the process of habit-forming. How do we acquire them exactly, and how are we to understand Dewey's elliptic sentences on “incorporation” and “adjustment”? If “the medium of habit filters all the material that reaches our perception and thought” (1922: 26), we may ask what perceptual habits are exactly, and how they are formed.

I will first argue that, highly concerned with refuting behaviorism and dualism, Dewey neglects the ecological complexity of his own claims. Obsessed with separating habit from mere repetition, his writing, in 1922, displays hesitations. The author is led to rather abstract considerations on the unity of experience, relating “impulse”, “habit” and “conduct” without enough clarification of their interdependence, and lets the reader without a satisfactory picture of situated habit-forming.

In a parallel series of objections to psychological dualism or “atomism”, Merleau-Ponty's thoughts on the durable unity of behavior over time, the living body in motion, and the perceptual field as a pre-reflective structuring of the world, provide resourceful arguments (Merleau-Ponty, 1942; 1945). In my second point, I claim that such phenomenological formulations might throw some light on Dewey's own pragmatist program. For both authors, indeed, conduct is a temporal variation of the organism's behavior, deeply rooted in a social and historical environment, and practically engaged in situations. Therefore, to apprehend the emergence of perceptual habits, the contextualizing of habituation – a biology-based concept Dewey tends to criticize – should be more fine-grained, down to bodily processes. Acquiring a habit is a modification of body postures involved in “motor tasks”: it transforms the “body schema”, which means, according to psychologists, the practical relating of my body and the world. Elaborating on Merleau-Ponty's remarks, I will emphasize that a perceptual “milieu” is not only achieved from socialization and history, but also matches behavioral and functional properties of human bodies. Its very structuring makes the perceived environment, as a concrete physical place, a suitable location for acquiring and transmitting habits, while renewing them (the “reprise”). This is the reason why no dualistic gap separates the natural “sensible world” from the perceived world of human culture and history.

Thus, following Dewey's naturalistic attempt to conceptualize the environment-organism transactions, one may notice that Merleau-Pontian phenomenological anthropology provides concepts which can throw light on the unsatisfactory phrases of “adjustment” and “incorporation”: “sedimentation”, “bodily anchoring”, “grasping”, “motor physiognomy”, “horizon” or “collusion with the world”, for instance, all aim to describe the open unity of historic, social and spatial backgrounds of perceptual habituation. Although Dewey's ecological claims may receive some clarification by this parallel, I will conclude by emphasizing one major limit – among possible others – to my parallel reading: beyond obvious methodological discrepancies, the previous analysis indirectly shows at least one main incompatibility. Merleau-Ponty's idea, as he characterizes perception, is to bring further description of perceived “structures” to bear, eventually, upon a naturally structured objective “sensible world”; an orientation that a pragmatist epistemology is definitely not likely to accept.