Why Radical Enactivism is not Radical Enough: A Case for Really Radical Enactivism (original) (raw)

This is nominally a book review of Hutto and Myin’s Radicalizing Enactivism: Basic Minds Without Content (The MIT Press, 2013). But it is a narrowly focused and highly prejudicial review, which presents an analysis of a contradiction at the heart of the book. Radicalizing Enactivism is a powerful and original philosophical argument against representations in cognition, but it repeatedly endorses an old-fashioned representationalism about language. I show that this contradiction arises from the authors’ unexamined, reflexive adoption of traditional linguistic concepts and terminology, which presuppose a representational interpretation of linguistic capacities and phenomena. The key piece of evidence for this analyses is the separability of Hutto and Myin’s substantive remarks on the ontogeny of language-dependent cognitive capacities, which they explain in terms of scaffolding and decoupling, from the representational gloss on those remarks that they present as if it were simply identical with observed empirical matters of fact. They follow a model laid out in Hutto’s earlier work, in which everyday linguistic activity is understood as instantiating abstract public vehicles with representational content (i.e., sentences which express propositions). I argue that this model is susceptible both to pre-existing arguments against representational theories of language and to a variant of their own ‘Hard Problem of Content’. The take-away lesson from Radicalizing Enactivism is that anti-representationalist accounts of language remain unconvincing - even to radicals like Hutto and Myin - because they have no way of explaining the phenomenal experience of literate speakers, wherein words really do feel like instantiations of abstract forms with determinable semantics. I suggest that anti-representationalists can address this by focusing on the ways in which patterns of attention become stabilized and interpersonally regularized as we learn language.

In this paper we defend a position we call radical connectionism. Radical connectionism claims that cognition never implicates an internal symbolic medium, not even when natural language plays a part in our thought processes. On the face of it, such a position renders the human capacity for abstract thought quite mysterious. However, we argue that connectionism is committed to an analog conception of neural computation, and that representation of the abstract is no more problematic for a system of analog vehicles than for a symbol system. Natural language is therefore not required as a representational medium for abstract thought. Since natural language is arguably not a representational medium at all, but a conventionally governed scheme of communicative signals, we suggest that the role of internalised (i.e., self-directed) language is best conceived in terms of the coordination and control of cognitive activities within the brain.

Davidson holds that thinkers cannot employ radically different conceptual schemes, but he he does not deny the (seemingly obvious) fact that small-scale conceptual divergences are possible. He defends the former claim against Quine by appealing to interpretivism, the idea that ascriptions of intensional states to a speaker do no more than systematically record facts about the speaker’s behavior. From interpretivism it follows that it is theoretically (though not practically) irrelevant which set of concepts an interpreter uses to state her theory of meaning. This is what allows Davidson to avoid conceptual relativism while accepting Quine’s thesis that behavioral evidence necessarily underdetermines assignments of meaning to speakers. However, as Davidson seems not to have realized, interpretivism also rules out the possibility of uncontroversial, small-scale conceptual differences. This fact, I argue, should lead us to reject interpretivism.

Attempts at combining Uexküll’s ideas with those of Peirce within a single quasi-discipline called ‘biosemiotics’ are ill-founded. Peirce’s ‘interpretant’ sensu lato refers to two qualitatively different mental states, one relating to indexes and icons (INT 1) and the other to symbols (INT 2). Animal communication is dyadic – the referent is a directly induced mental state (INT 1). Glottocentric communication is triadic because the connection between symbol and INT 1 is mediated by INT 2. Whereas the gradualist view of glottogenesis is erroneous, Müller’s and Chomsky’s saltationist theories may imply that the idea of language Rubicon is antievolutionary. However, the views of Pavlov and Vygotsky and of their modern followers, Deacon and Tomasello, while being Darwinian, support the saltationist scenario. The emergence of the second signal system, of symbols, and of INT 2 was a psychological leap. In human communication, apart from the semiotic triangle (INT 1 – INT 2 – symbol), the dyadic relation between non-symbolic signs and INT 1 still holds.

In this paper, we sketch a conception of the theory of meaning that focuses on political speech as a core example, which has different goals than standard practice—such as capturing unintended communicative effects—and different key concepts, such as that of a speech practice. The book that this paper begins is devoted to adjusting, reinterpreting, and modifying tools, concepts, and understandings thereof to incorporate the resources we need to understand speech in a non-ideal world.