The World of SPLectrum
Home > Positioning > Subjects > Enactivism
Enactivism
Enactivism is the position in the philosophy of mind and cognitive science that cognition is the enactment of a world through an agent’s embodied activity, rather than the representation of a pre-given world by an inner model. Its home question is what cognition is: not the processing of representations that mirror an external reality, but the sense-making of a living, embodied agent coupled to its surroundings, bringing forth the world it is sensitive to through its own history of action. It is one of the “4E” approaches — cognition as embodied, embedded, enacted, extended — and the most philosophically developed challenge to the computational and representational theory of mind. It shares its anti-representational premise with constructivism, and reaches it from the philosophy-of-mind side rather than the epistemological one.
The founding: The Embodied Mind
The approach was founded in The Embodied Mind (1991) by Francisco Varela, Evan Thompson, and Eleanor Rosch, building on the earlier theory of autopoiesis that Maturana and Varela had developed. Its founding thesis is anti-representational: cognition is not the mirroring of an outside world by an inner model but the bringing-forth of a world through embodied action — a living being maintaining its coherence in the face of what perturbs it, and thereby constituting a domain of significance, an Umwelt, that is neither simply given nor merely projected. The book drew together three sources that remain the tradition’s reference points: the biology of autopoiesis, the phenomenology of Husserl and Merleau-Ponty, and Buddhist meditative psychology, whose analysis of a self without a fixed ground it used against the idea of a central representing subject.
The three strands
What looks from outside like one movement is internally a rivalry between three strands that disagree about how far to push the anti-representational claim and where to ground it — the tradition’s central structure, and the thing any honest account has to hold open rather than resolve.
Autopoietic / life–mind continuity. The strand closest to the founding, developed by Varela, Evan Thompson (in Mind in Life), and Ezequiel Di Paolo. It grounds cognition in the self-production of life: a living system, by maintaining itself, already institutes a perspective and a concern, so that mind is continuous with life and sense-making reaches all the way down to the cell. Di Paolo’s adaptivity and his and Hanne De Jaegher’s participatory sense-making — social understanding generated in the interaction between agents, not inside either head — extend this strand into the social and the linguistic.
Sensorimotor. Associated with J. Kevin O’Regan and Alva Noë, this strand locates perception in the agent’s practical mastery of sensorimotor contingencies — the lawful ways sensory input changes as one moves and acts. To see is not to build an internal picture but to exercise a skilful know-how of how appearances shift with activity. It is enactive about perception without necessarily committing to the life–mind continuity the autopoietic strand insists on.
Radical. The strand of Daniel Hutto and Erik Myin, which presses the anti-representational claim hardest: basic minds, they argue, involve no mental content at all — no representations, contentful or otherwise — and content enters only later, with language and shared practices. Radical enactivism is thus a claim about what to deny, and it sits in tension with the autopoietic strand’s talk of sense-making and the sensorimotor strand’s know-how, both of which it presses to say whether they smuggle content back in.
Reach and reception
Enactivism’s influence has spread well beyond its founding questions — into social cognition, psychiatry and autism research, education, robotics, and the study of language, where Di Paolo, De Jaegher, and Elena Cuffari’s linguistic bodies carry the enactive account up into speech. The approach is a standing challenge to the representational and computational orthodoxy in cognitive science, and its practitioners are active in resisting attempts to absorb it into neighbouring frameworks — notably the argument over whether enactivism can be folded into the predictive processing and free-energy picture, which several enactivists reject as resting on incompatible foundations.
Where enactivism stops
The recurring challenge is the scaling-up problem: enactivism gives a compelling account of basic, embodied, sensorimotor cognition, but critics doubt it can reach abstract thought, offline reasoning, and language without reintroducing the representations it set out to do without. The autopoietic strand’s answer — a graded continuity from biological organisation through sensorimotor engagement and participatory sense-making up to linguistic agency — is a proposal under debate, its dialectical method itself contested, not a settled result. The radical strand’s alternative answer, that content simply arrives with language, faces the mirror-image charge of leaving the transition unexplained.
The three strands are also a boundary of a different kind: because they disagree about the anti-representational thesis and its grounding, “enactivism” names a family whose members are not always compatible, and an objection that lands on one strand may leave the others untouched. Whether the family shares enough to count as a single position, or is held together mainly by a common opponent, is a question the field has not closed.
Persons
See also: Autopoiesis · Phenomenology · Constructivism