What Might Be Learned from Studying Language in the Chimpanzee? the Importance O F Symbolizing Oneself

Ten years ago, there was little reason to believe that much could be learned from studying language in the chimpanzee. Earlier reports (e.g. Ref. I , 2) of a chimpanzee's inability to use language seemed to demonstrate that biological factors limited the extent to which the chimpanzee could learn to use abstract symbols. Thanks to the work of the Gardners,'-4 Premack,'J Fouts,9.10 and Rumbaugh and Von Glaserfeld," we now know that a chimpanzee has a much larger linguistic potential than was ever imagined. Once allowance was made for the vocal limitations of a chimpanzee (Lieberman, 1968; Lieberman et al., 1972), it became possible to teach it to use arbitrary gestures, plastic tokens, or the lexigrams of a computer console in a manner that parallels the human use of single symbols. Most provocative have been demonstrations that chimpanzees can use sequences of symbols. This suggests a syntactic potential as well as a symbolic one. In the eagerness to establish whether a chimpanzee can acquire syntactic competence, certain more basic functions of language appear to have been bypassed. Our main purpose in this paper is not only to assess the syntactic accomplishments of Washoe, Sarah, and Lana, but also to delineate other functions of human language that can be studied in the chimpanzee-functions that d o not require syntactic competence . Before considering nonsyntactic functions of human language, it is of interest to digress briefly to compare studies of language in a chimpanzee with studies of language in another currently popular subject, the computer. At present, there is nothing to suggest that computers can simulate the acquisition of language. However, language acquisition by another being may illuminate parallel processes in humans. As compared with the computer, the chimpanzee has many obvious advantages. First, because of its biological and social similarity to man, a chimpanzee is a much more likely source of useful contrastive information on human language. Second, a computer can d o nothing more than execute the instructions of its programmer. We cannot be sure what a chimpanzee will d o with symbols once it has learned their meanings. Human language has two important functions: ( I ) it facilitates communication (particularly about events and objects displaced in time and/or space) and (2) it structures how we perceive ourselves and the world. The child's tendency to symbolize interpersonal relations between himself and others is obviously a fundamental ingredient of personality development in all human societies-in particular, the concept of seffbounded on one side by desires and on the other by social patterns. A child's ability to refer to itself, its desires, and the social pressures of its