Commentary on James W. Hurford

Abstract: 56 words

Main Text: 991 words

References: 45 words

Total Text: 1092 words

phone +64 3 364 2211

andrew.carstairs-mccarthy@canterbury.ac.nz

http://www.ling.canterbury.ac.nz/

Abstract

Hurford claims that empty variables antedated proper names in linguistic (not merely logical) predicate-argument structure, and this had an effect on visual perception. But his evidence, drawn from proper names and the supposed inability of nonhumans to recognise individual conspecifics, is weak. So visual perception seems less relevant to the evolution of grammar than Hurford thinks.

Hurford draws attention to a parallel between, on the one hand, the roles of the ventral and dorsal pathways in vision, and, on the other, the roles of predicates and variables in predicate calculus. Just as the variable in predicate calculus has no role other than a deictic or indexical one, of locating an individual to which certain predicates belong, so the dorsal pathway (it seems) has scarcely any role other than to locate an object in space, nearly all its other characteristics being processed via the ventral pathway. How significant is this for language, either today or at an earlier evolutionary stage? Hurford does not claim that the correlation is today very close, and I agree with him. One cannot identify the dorsal-ventral contrast with the noun-verb contrast, for example. But he alleges a reflection of the dorsal-ventral contrast in the mental representations of all animals except modern humans, inasmuch as (he claims) only modern humans have a concept of individuals that are in principle proper-namable—that is, individuals associated with more semantic content than mere indexical place-holders. On that I find what he says unpersuasive. So I suspect that the parallel that he adduces has even less significance for language than he suggests. If so, then visual perception sheds little or no light, unfortunately, on the puzzle of why language (particularly syntax) is as it is.

‘Protothought had no equivalent of proper names’, says Hurford, and that is why it is easy to fool tern chicks about their parents: visually they are so easily fooled that they will react towards a loudspeaker as it were a parent tern. Hurford concludes from this that tern chicks have no mental representation of their parents as individuals. (Hurford would presumably interpret in the same way the apparently sophisticated social awareness of vervet monkeys (Cheney & Seyfarth 1990).) But that seems an overambitious conclusion. Terns may be easier to trick than humans are, but that proves nothing relevant to this issue. Let us suppose that, unbeknownst to me, Jim Hurford has an identical twin brother, Tim Hurford. I know Jim Hurford slightly from occasional encounters at conferences, and I meet at one conference a person who looks very much like Jim (similar height, hair colour, voice quality and so on). However, this person is Tim, who has agreed to impersonate Jim in order to bamboozle unwitting colleagues. I may well be taken in for a few minutes, or indeed for the whole duration of the conference. Does this mean that I have no concept of Jim Hurford as an individual with spatio-temporal continuity and a unique life-history—a proper-namable individual, in other words? Clearly it does not. Likewise, the terns’ befuddlement by the loudspeaker does not show that they lack any concept of at least some other terns as individuals, and hence it does not show that they lack the mental underpinning to use proper names, supposing that they were linguistically equipped to do so.

In support of his view that proper names are linguistically ‘late’, Hurford points out that some proper names are derived from definite descriptions, such as Baker, Wheeler, Newcastle. But one might just as well argue that the fact that signing chimps can use ASL signs that are classified as proper names (‘ROGER’, ‘WASHOE’, etc.), even if such signs are not derived from definite descriptions, establishes that our protolinguistic ancestors—indeed, the common ancestors of chimps and humans—must have been mentally equipped to use proper names too. So it is risky to derive conclusions about proto-cognition from proper name etymologies and usage.

Hurford mentions the language Matsigenka. This is spoken in a community in which there are genuinely no personal names, the few individuals that a person interacts with regularly being identified solely by kinship terms: ‘father’, ‘patrilineal same-sex cousin’ and so on. But this surely demonstrate the very opposite of what Hurford thinks. I would be surprised if a Matsigenka speaker treats as a single individual all his or her relatives to which the same kinship term applies. That is, I assume that, if a Matsigenka speaker has two relatives both of whom fall under the term glossed ‘patrilineal same-sex cousin’ (for example), he or she will nevertheless be aware that they are different individuals, and will treat them as such. If I am wrong in this assumption, then Hurford’s case is supported. On the other hand, if I am right, what it shows is that the unavailability of proper names has no bearing on the ability to recognise entities to which proper names could appropriately be applied, if the necessary framework (social as well as linguistic) permitted that.

The reason why Hurford is so keen to establish this point, it seems, is an assumption that there must be some stage of linguistic evolution at which (proto)syntax behaved in a fashion that reflected more closely than it does now the way in which predicate-argument structure works in logic. This assumption, if correct, opens up the possibility that, as Hurford puts it, ‘the dorsal/ventral separation in higher mammals is ... an evolved hardware implementation of predicate-argument structure’. However, I do not share his assumption. Even if it is true that proper names are complex to handle in first-order predicate logic, there is no reason to suppose that what came early in the mental representations of (proto)humans is the same as what is basic in logical terms. Whatever one thinks of evolutionary psychology, it has drawn attention to the fact that mental tasks that are simple, in some logical sense, are not necessarily done well by most humans, while there are ‘harder’ tasks that we handle with ease. That is likely to be true of how language arose too. I have expressed elsewhere scepticism about whether the neurophysiology of vision can explain much in language (Carstairs-McCarthy 1999: 90-1). Hurford’s exploration is interesting, and he is evidently well-informed on brain neurophysiology, but he has not made me any less sceptical.

References

Carstairs-McCarthy, A. (1999) The origins of complex language: an inquiry into the evolutionary beginnings of sentences, syllables and truth. Oxford University Press.

Cheney, D. L. & Seyfarth, R. M. (1990) How monkeys see the world: inside the mind of another species. University of Chicago Press.