Social and socio-
Ontogeny and symbolism
Animal language and cognition projects
Interpreting the data on the cognitive and linguistic abilities of non-human species needs to be done with care. The results of particular studies need to be understood in the light of (1) the ontogeny of the ability in question and the specific training procedures involved in demonstrating it; (2) the settings within which abilities are observed þ laboratory versus field; (3) a precise description of and the limits of the abilities claimed to have been demonstrated; (4) the number of individual animals that are claimed to show the abilities in question; and (5) possible problems in experimental designs. Given all these constraints, summary statements must be somewhat provisional.
Some of the Great Apes, in some situations, have achieved the use of rudimentary symbols. This ability can be used to support symbolic forms of communication, especially requests, between apes and their human 'carers'. This symbolic communication ability does not show the multiplicity of functions that humans employ. Particularly in the past, the training methods used to establish these abilities did not bear much resemblance to the procedures by which human children develop language skills; more recent methods do (Savage-Rumbaugh et al 1993; Boysen 1993b). There is little evidence that most apes use grammar in their communications, though one bonobo has attained at least a rudimentary grammar. Furthermore, in some highly specific situations, apes, sea-lions (Zalophus californianus), and dolphins (Tursiops truncatus) are able to comprehend the order of lexcical items, and some apes and pigeons can reproduce certain simple series reliably.
The relation between comprehension and the production of symbols in non-human species is unclear, though the two systems appear more independent of each other than in humans. With specific training to do so, highly 'language'-trained chimpanzees (Pan troglodytes) can transfer symbols learned in one mode to the other, and then generalize this ability to new symbols. The bonobo (Pan paniscus) seems able to make such transfers far more readily. Some enhanced quantitative and reasoning abilities can be demonstrated for 'language'-trained apes as compared to apes not so trained. In some cases these findings depend on the performance of one particularly apt ape; and it is not yet clear that it is only 'language' training that is responsible for the differences found: this research should be extended.
The demonstration of symbol use and category formation in non-primate species implies that it is not the unique organization of the primate brain, nor any special property of their social and physical environments alone, that is responsible for the possession of some symbolic ability. Furthermore, since the apes are our contemporaries and not our ancestors, the question remains open as to the evolutionary significance of those abilities for human abilities. It is clear, however, that humans have a far greater facility for acquiring and using symbolic systems than any other species.