Simulating others: the basis of human cognition?

The paper critiques the argument of Michael Tomasello's Cultural Origins of Human Cognition (1999). This culture-first theory is judged to be a good sketch of how nature predisposes humans for talk. Above all, this is because if language mediated 'perspective-taking' depends on cultural process, no innate linguistic representations are necessary in learning to talk. Unfortunately, the model is flawed by Tomasello's claims for a putative species-specific competency. Rather than posit a simulation mechanism to link orthodox views of language with Gricean models of communication, I follow Dennett in treating 'intentions' as folk constructs. Talking, on this view, arises from encultured contextualizing. Situated, embodied activity turns infants into perspective-takers who, far from learning or acquiring 'forms', slowly become persons. Gradually, the infant's developing social capacities produce activity that invites others to attribute linguistic knowledge to the child. (C) 2003 Elsevier Ltd. All rights reserved.