Provided by Cognitive Sciences ePrint Archive
How to do things without words:
Infants, utterance-activity and distributed
cognition.
David Spurrett and Stephen Cowley
Note
This is a preprint version of a paper subsequently published in Language Sciences.
(Spurrett, D. and Cowley, C. (2004) ‘How to Do Things Without Words: Infants,
utterance-activity and distributed cognition’ Language Sciences, 26(5), pp 443-466.) The
final published version is available from ScienceDirect here:
http://dx.doi.Org/10.1016/j.langsci.2003.09.008
Abstract
Clark and Chalmers (1998) defend the hypothesis of an ‘Extended Mind’, maintaining
that beliefs and other paradigmatic mental states can be implemented outside the central
nervous system or body. Aspects of the problem of ‘language acquisition’ are considered
in the light of the extended mind hypothesis. Rather than ‘language’ as typically
understood, the object of study is something called ‘utterance-activity’, a term of art
intended to refer to the full range of kinetic and prosodic features of the on-line
behaviour of interacting humans. It is argued that utterance activity is plausibly regarded
as jointly controlled by the embodied activity of interacting people, and that it
contributes to the control of their behaviour. By means of specific examples it is
suggested that this complex joint control facilitates easier learning of at least some
features of language. This in turn suggests a striking form of the extended mind, in which
infants’ cognitive powers are augmented by those of the people with whom they interact.
Introduction
In ‘The Extended Mind’ Clark and Chalmers (1998) argue for ‘active externalism’ - the
view that the mind, or what realises it, need not be confined within either the brain, or
the body, of the minded individual. We’re sympathetic to their position, and line of
argument. Among the many things outside the brain and body of any particular individual
are, of course, other brains and bodies. This paper is a preliminary sketch of what might
happen when minds extend into one another. The paper is in two parts - the first
establishing some theoretical points of reference, the second being largely descriptive.
We note at the outset that what we have written here is speculative and sometimes loose.
It is also, hopefully, suggestive of fruitful lines of further reflection and investigation.
Our sub-title refers to ‘utterance-activity’. This is a term of art, used, here, to refer to the
full range of kinetic and prosodic features of the on-line behaviour of interacting
humans. Utterance-activity sometimes includes what are usually regarded as words and
strings of words, but need not. We regard utterance-activity as at least as good an object
of scientific interest in its own right as ‘language’ traditionally conceived. Further, we
regard it as continuous with, and inextricable from, (non-written) language. We combine
this continuity thesis with the developmental claim that language, as usually understood,
1 / 23