Our suggestion is that a similar function is served by emotional signalling in the
epistemic,19 rather than primarily strategic, interactions between infants and their
caregivers, and in adult conversation. Our descriptions, unlike many accounts of
linguistic and some of strategic phenomena, have not been limited to turn-taking
interactions, and instead have emphasised the ways in which roughly simultaneous co-
ordination of prosodic and affective display takes place, and how such co-ordinated
display can convey significant information about relationships. Such display must convey
social information in animals without language, and we contend that it continues to do so
in humans. If this speculation isn’t obviously wrong, then it suggests two lines of
development of the notion of the extended mind.
First, especially considering the ‘Oeu!’ example, it seems unquestionable that sources of
feedback relevant to both Aldo’s and Monica’s control of their own vocal production,
during the period in which they are so strikingly co-ordinated, come from both their own
vocal production, and that of the other. More generally, all of the types of affective co-
ordination we have described involve integration of inputs from each participant’s own
behaviour and that of others. This is a striking set of examples of embodied cognition of
the sort Clark refers to in the work we have grouped under the ‘robots’ category. We
hope to have shown something of how this type of embodied control could be crucial to
the functioning of utterance-activity, and why it merits further empirical investigation.
Second, considering the epistemic pay-offs of the types of embodied co-ordination we
have described, it is clear that the model of the solitary infant epistemologist upon which
much of the poverty of the stimulus debate is based, is seriously in need of revision.
Infants are, in virtue of affective co-ordination, able to function as a kind of cognitive
extension of their own caregivers, who focus their attention, regulate their levels of
arousal, reinforce and retard patterns in their behaviour, and provide all manner of
sources of environmental regularity amenable for infant exploitation. This type of
interaction environment permits the construction of socially indexical relationships, and
the disembedding of labels and relationships in ways amenable to being recognised as
symbolic. The types of embodied co-ordination noted immediately above, that is, permit
a particular type of extended mind, in which infants’ cognitive powers are augmented by
those of the people with whom they interact.
Acknowledgements
Earlier versions of this paper were presented to the mind AND world working group at
the University of Natal, Durban in April 2001, and (under the title ‘Minded Apes,
Talking Infants and the Distribution of Language’) at ‘The Extended Mind’ conference in
Hertfordshire, in June 2001. The present version of the paper makes considerably less
reference to ape language research than earlier incarnations, although see Cowley and
Spurrett (2003). This paper has benefited from comments from and discussions with
Andy Clark, Anita Craig, Andrew Dellis, Dan Hutto, Denis McManus, Richard Menary,
Mark Rowlands, Fran Saunders, Leslie Stephenson, Susan Stuart, and Michael Wheeler.
Author affiliations:
(1) David Spurrett
Philosophy
University of Natal
Durban
16 / 23