In Cowley and Spurrett (2003) we criticise Taylor (in Savage-Rumbaugh, Shanker and Taylor 1998) for
reacting to what he sees as the failure of traditional linguistics by suggesting that we relax our demands
for (scientific) knowledge, partly by means of some Wittgensteinian therapy.
See, e.g., Chomsky (1965, 1967). Laurence and Margolis (2001) is a useful recent and philosophical
review of the poverty of the stimulus argument.
This work tested language discrimination (in this case the ability to distinguish Dutch from Japanese) in
both human newborns and cotton-top tamarins. Both types of subject show significant powers of
discrimination depending on fairly abstract equivalences rather than simply prosodic features. The
authors conclude that ‘Since tamarins have not evolved to process speech, we in turn infer that at least
some aspects of human speech perception may have built upon pre-existing sensitivities of the primate
auditory system.’
The work (see also Nazzi, Bertoncini and Mehler 1998) indicates that rather than distinguishing
languages per se, infants distinguish between stress-timed, syllable-timed and mora-timed languages.
We note that Savage-Rumbaugh herself accepts the poverty of the stimulus argument and then arggues
that the genetic similarity between chimpanzees and humans suggests that chimpanzees are likely to
have at least some of the same adaptations for language. We prefer the line suggested here, and in
Cowley and Spurrett (2003).
A more general form of our question, without the developmental spin of the version in the main text, is:
How do the apparently symbolic aspects of talk relate to wider utterance-activity?
One of us (Cowley 2002) has critically engaged with aspects of Deacon’s account elsewhere, and
accused Deacon of ‘token realism’ about the neural counterparts of apparently symbolic behaviour.
There is evidence (see Scheper-Hughes 1985) that under conditions of severe scarcity a combination of
factors relating to the apparent physical health of an infant and its patterns of interaction (including
levels of crying) are significant factors in determining levels of care and feeding, possibly determining
which offspring will survive. Mann (1992) found that in the absence of serious scarcity, maternal
attention tended to focus on the more healthy of two pre-term twins, whether or not the less healthy
infant was more responsive, and smiled more.
A parent may have other children to which to allocate resources, or may bet on their chances of success
with future offspring, whereas the developing infant has no such options. Haig (1993) documents the
ways in which, during pregnancy, the foetus (which has less interest than the mother in her own other
and possible future offspring than it does in its own life) can operate more like a parasite than an ally,
competing, inter alia, over blood supply, and levels of blood sugar. See also Trivers (1974) on some
aspects of parent-infant conflict.
This research, with 28 hour old infants, showed an appreciable preference for a static and schematic
smile over a frown and a bulls-eye figure. The infants showed slightly greater interest in a 6 by 6
checkerboard pattern.
Fernald (1992) documents, inter alia, prosodic patterns (found across multiple cultures) indicating
approval, prohibition, comforting, and engaging attention. It is important to note one way in which the
approach we favour departs from hers. We are interested not only in the ‘internal’ prosodic properties of
utterances, but also in relational properties discernible in ongoing utterance interactions. Our third
example below (‘Oeu!’) is an illustration.
The contingent patterns need not be cultural: It is well documented that, for example, levels of maternal
depression make specific and measurable differences to patterns of affective display and behaviour in
infants and children (Lundy et al 1997).
Papousek (1969) showed that by creating environments in which specific movements by an infant could
make things happen in those environments, that the infants smiled when they did ‘work out’ how to
exercise control. This suggests that infants are disposed to derive satisfaction from such discoveries.
22 / 23