In response to:
Gorilla Talk from the October 9, 1980 issue
Gorilla Talk from the October 9, 1980 issue
To the Editors:
Francine Patterson’s letter, inherently chaotic enough, complicates matters in that it seems to be addressed, pell-mell, to Herbert Terrace and the two undersigned, whose books are being reviewed, and to Martin Gardner, who reviewed them. Her letter contains scattered quotations, as in her opening paragraph, ascribed to “the Sebeoks,” but some of these are not, in fact, from our book; they are the words of Mr. Gardner. Her complaint also features a number of bizarre denials of dramatic assertions which, to our knowledge, no one has ever made—certainly not the two of us. An example of the latter is Patterson’s indignant rejection: “To say that the gorilla’s use of sign language is virtually identical to that of the human child is wrong….” The contrary has never been asserted by any scholar, not even the most enthusiastic scribbler.
Patterson’s repeated whimper, “They have not examined my data,” is counter-factual. Indeed, we have checked over every scrap of information—such as it is—that she has disclosed through normal scientific outlets, and have also surveyed all available vulgarizations of her data, presuming that it was she who authorized their public circulation. Her dissertation, long delayed, became accessible to us only after Speaking of Apes was typeset; it will, accordingly, be critically dealt with in our forthcoming article, “Clever Hans and Smart Simians, The Self-Fulfilling Prophecy and Kindred Methodological Pitfalls,” now in press and due to appear, early in 1981, in a leading anthropological journal. To anticipate, however, we must point out here that there are basic and very disturbing discrepancies between her data as reported in her thesis and as published in her scattered articles.
Patterson argues in her letter that “Non-verbal cues are omnipresent in human communication as well as in ape-human communication…. Contrary to the Sebeoks’ assertion, it is easy to control for cues such as eye-pupil size…by wearing sunglasses.” In fact, Sebeok says virtually the same thing, on p. 420 of Speaking of Apes. Patterson has not, however, previously troubled to report this form of control and still neglects scores of other sources of leakage, many of which we enumerated in our study. In Patterson’s reports, the precise training methods or test situations she employed are not usually described, and we are asked to accept her assertions about her apes’ performances on faith. When experimental conditions are specified, controls are often so feeble as to defy belief. Her facile statement about the importance of nonverbal cues belies her continual treatment of the Clever Hans effect as a minor methodological irritation instead of recognizing it—in the face of the vast amount of scientific evidence attesting to its pervasive influence—as a global application of the self-fulfilling prophecy, even in situations where experimenters are not as emotionally committed to their experimental subjects as Patterson is, by all accounts, to her gorillas.
Patterson censures us for our “ignorance of sign language structure,” but the shoe is on the other foot. She has never produced a shred of evidence that her apes’ gestures are, in fact, signs, in the technical acceptation of this basic semiotic unit, as tellingly put forth by Petitto and Seidenberg (Brain & Language 8:162-83, 1979). Terrace’s findings, which have now been supplemented by a discourse analysis of the data presented in Patterson’s dissertation, fully confirm our long-held suspicion that the roughly duplicative gorilla gestures that she persists in calling signs are scarcely more than “signifiers” without any “signification” in the human sense.
Patterson’s repeated overinterpretations of her subjects’ behavior as jokes, apologies, puns, and now English rhymes (!) are clear examples of the Pathetic Fallacy, with which we have dealt at length in our study. In the case of the sign for drink, which she focuses on in her letter, we would still like to know what other locations were used by Koko with the hand configuration in question. Assuming that, as Patterson reports, the trainer had been trying for some time to persuade Koko to make this sign, we may guess that the ape in fact moved her hands in various directions during the session, and with various accompanying facial expressions. How were all of these other “signs” and expressions interpreted? The possibilities, given the lack of the kind of information any thoughtful person would demand, are limitless. Patterson would profit immensely from mastering the principles of biology—notably the writings of Jacob von Uexküll and his Umweltlehre—and the best of contemporary linguistic theory—e.g., Noam Chomsky’s Rules and Representations, ch. 6—not to mention a number of classic circus training manuals.
Patterson complains that there are “numerous” erroneous remarks in our introductory chapter. In her letter, she offers but one example, which happens to be false. We cannot repeat here our detailed critique of the cautionary method of the so-called “double-blind” test, a magic device in which Patterson seems to place touching faith, but which, as we and a number of others have demonstrated, is all too often embarrassingly inadequate. To take a single example from Patterson’s work, descriptions and illustrations in her published articles show that the box she used for double-blind testing was sufficiently small so that Koko could have moved it around at will. Patterson’s experimental design for such tests by no means rules out certain guessing strategies on the part of ape and the “blind” experimenter, given the small universe of stimuli which was available for use and the familiarity of both ape and human with these materials, as well as with one another’s facial expressions, body movements, and the like. It should be said, finally, that these double-blind tests have been used only rarely by Patterson. We have found published reports of only one series, administered to Koko in September, 1975. Koko, Patterson has admitted in print, was extremely reluctant to perform under these conditions.
Patterson claims that “One cannot trace the evolution of language from an armchair in Indiana.” Our view is that, to the contrary, a Gedankenexperiment can never be separated from its technical realizations in any field of science, since an understanding of nature can only be obtained by the informed and careful application of both. Patterson’s lack of methodological sophistication is precisely traceable to her innocence of fundamental theoretical advances in fields adjacent to her own. Her fellow psychologists who are knowledgeable about such issues will have to judge the extent to which she is a victim of self-deception and why, when some of the more prominent “pongists” have publicly renounced this line of investigation, she persists, against the weight of versant opinion and the laws of probability, in pursuing the will-‘o-the-wisp that apes are capable of language-like performances. Those who are ignorant of the millennial history of the Clever Hans Phenomenon are doomed to replicate it endlessly with one animal form or another, whether embodied in birds, horses, pigs, porpoises, the great apes, or, most recently, the wondrous tortoises of Milwaukee.
Thomas A. Sebeok
National Humanities Center
Research Triangle Park, North Carolina