SEVEN

Toward a Natural History
of Language

Two coeds met a frog on the campus. “Kiss me,” the frog said, “and I’ll turn into a psychologist.” One of the girls picked him up and tucked him in her jeans. “Aren’t you going to kiss it?” the other asked. “No. A talking animal is really worth something.”

—Anonymous

1. SETTING THE SCENE

Konrad Lorenz, who, in 1973, was awarded the Nobel Prize for his studies of animal behavior, concluded his book The Foundations of Ethology, published five years later and translated into English in 1981, with an appendix concerning Homo sapiens. He professes there to “share Noam Chomsky’s opinion that syntactic language is based on a phylogenetic program evolved exclusively by humans.”¹ Some of his terms require notice.

First of all, by “humans” Lorenz does not mean only members of the species Homo sapiens sapiens, to which we—the writer and his readers, and Professor Lorenz too—belong. He intends to include all our predecessors in the genus Homo, beginning—so far as paleontologists now know—with a species the late Louis Leakey had dubbed Homo habilis, a creature which seems to have emerged from the australopithecine line, in East Africa, about 2 million years ago, and rather abruptly at that. This advanced ape was endowed with a substantial increment in brain weight over its ancestors, having a skull capacity of about 750 cc. The number of tool types that these archaic hominids employed in this period, which is called the Oldowan, amounted to perhaps a half-dozen or so. A recent close look at their stone artefacts reveals that these hominids “habitually transported materials, presumably for future use,”² a circumstance which implies that they operated with considerable—and ever increasing—advance planning. The point here is that long-range foresight presupposes language, so that, judging by the technological complexity of Homo habilis, we can safely assume that language must have evolved prior to two million years ago.

This assumption is reinforced by the fact that a critical lateralization of the hominid brain had occurred by 1 .9 to 1.4 million years ago, and that the unique hominid preference for right-handedness must have been allied to language, “since both traits appear to be strongly controlled by the left hemisphere of the brain in most modern humans.”³ Language, moreover, is believed to have been selected for first, followed by a neurological “field effect” in the left hemisphere, and that it was in consequence of this that right-handedness developed in most modern humans.⁴

The appearance, in an era called the Acheulian, a mere half-million years later—which is amazingly swift in evolutionary terms—of another hominid form, Homo erectus, with a skull capacity of up to 1,300 cc, hence endowed with far more advanced cognitive skills, argues strongly for the survival value of this new aptitude we call language. Then, in an even more impressive evolutionary acceleration, an early form of Homo sapiens emerged, a mere 200,000 to 300,000 years ago. This fellow, of the Mousterian era, had a 1,400 cc brain capacity. And, by 40,000 years back, in an age known as the Upper Paleolithic, the modern human being, Homo sapiens sapiens (you and me, in short) materialized, with our brains averaging 1,500 cc. It was during this period of, roughly, 2 million years of overall hominid development that, as I shall argue, yet another unique human propensity—which must by no means be confused with language, although it habitually is—ultimately evolved. We call it speech.

Further questions raised by Lorenz’s observation are these: Just what does he mean by a “syntactic language”? And, more generally, what does the term “language” denote—or, more accurately, does not convey—in this context? Most laymen (and perhaps some scientists, too) might argue that the essential purpose of language is to enable people to communicate with one another. However, the same Chomsky, this eminent contemporary linguist Lorenz opted to associate his opinion with in his cited statement, has also pointed out “that either we must deprive the notion ‘communication’ of all significance, or else we must reject the view that the purpose of language is communication.”⁵ This provocative statement needs to be considered from several points of view: how did ancient men and women communicate among themselves; and, if language did not emerge as an adaptation for purposes of communication, what could its true function have been to merit such undeniable evolutionary success?

Ancient men and women communicated—that is, exchanged messages—within and between their societies very much as all other species of animals do and have always done: by means of nonverbal signs. Each and every member of each and every species alive or ever existing has come into the world with its unique repertoire of nonverbal signs which decidedly promotes its survival.⁶ This holds for modern human beings no less than for our ancestors: that is, we ourselves communicate with other human beings, as well as with the other forms of animal life we share an ecological niche and regularly interact with, most of the time by a means of a large variety of nonverbal messages.

It was Gregory Bateson, one of the great thinkers of our age, who pointed out that the coding devices characteristic of language differ profoundly from those of the many nonverbal devices prevalent throughout the animal world (ourselves, of course, included) and that the general belief that “language replaced the cruder systems of the other animals” is totally wrong. Were language in any sense an evolutionary replacement for nonverbal communication, we would expect the older devices “to have undergone conspicuous decay.” Clearly, they not only have not done that, but also have in history become even more intricately elaborated in humans, so much so, indeed, that our skills in this respect exceed by far anything that any other animal is capable of displaying. This surely means, Bateson then goes on to argue, that language, on the one hand, and our repertoire of nonverbal communicative contrivances, on the other, must serve totally different functions.⁷

To appreciate the true function of language at this early age, we must sharply distinguish between two pivotal concepts of evolutionary theory: historical genesis and current utility.⁸ As was emphasized by Gould and Vrba in their attempt to clarify the confusion surrounding this issue, “Most of what the [human] brain now does to enhance our survival lies in the domain of exaptation” (a term they coined to account for characters that have evolved for other usages but were later “co-opted” for their current role).⁹

In that sense, it can be argued that, whereas language was a primary evolutionary adaptation, speech—which appeared, with Homo sapiens, not more than about 300,000 years ago—is but a recent secondary exaptation. This means that this vocal-auditory, temporal (hence linear) expression of language has acquired, in its manifestation as speech, an important incremental function, namely, to serve the current utility of a communicative function, thus supplementing, in a subtle and intricate fashion, the entire human repertoire of nonverbal devices inherited from our primate ancestry.

It is reasonable to suppose that the adjustment, or fine-tuning, of the encoding capacity required by speaking to the decoding capacity required to understand speech, and vice versa, took about two million years to achieve at least partially. (Full understanding is a rare commodity; most of the time most of us don’t quite grasp what another human being is trying to tell us.) Even today, humans have no special organ for speech, which is formed by a tract orginally designed for two entirely different biological functions: the alimentary and the respiratory. Speech is then received, as any other sound, by the ear, which has still another phylogenetic source and is a rather newly acquired sensory receptor.

But if language was not, in its origins, a character selected for its current role to supplement, or to enhance, communication, what did it evolve for?

The answer to this cardinal question is inherent in the life’s work of one of the most original—although, for reasons that can’t be discussed here, persistently misunderstood—geniuses of theoretical biology, Jakob von Uexküll.¹⁰ Briefly, Uexküll founded a special method of inquiry he called Umwelt-Forschung, a term notoriously difficult to render into English. It means research into the worlds surrounding all animals, as they themselves perceive their own subjective universes.

These phenomenal worlds, which diverge from species to species, can be visualized as if they were bubbles within which each creature is imprisoned, as it were, by virtue of its total and unique stock of particular sensory instruments. (Phenomenal worlds differ not only among species but even among members of the same species. An elementary example: most of us humans see our surroundings in color; however, an occasional individual who chances to be totally color-blind sees the identical surroundings in black and white.)

Every animal comes alive equipped with its distinctive Umwelt, which some can modify to a degree by learning. Let’s call this self-world a model. Obviously, a reasonably accurate correspondence must obtain between reality and the model of it in every animal’s mind, or else the species will become extinct. The reasons for this were vividly protrayed by another Nobel Laureate (1965), the French geneticist François Jacob: “No matter how an organism investigates its environment, the perception it gets must necessarily reflect so-called ‘reality’ and, more specifically, those aspects of reality which are directly related to its own behavior. If the image that a bird gets of the insects it needs to feed its progeny does not reflect at least some aspects of reality, then there are no more progeny. If the representation that a monkey builds of the branch it wants to leap to has nothing to do with reality, then there is no more monkey. And if this did not apply to ourselves, we would not be here to discuss this point.”¹¹

The term “model” is here intended to convey (1) an intellective construct, together with (2) rules for logical operations. The models at the disposal of other animals are relatively rigid, with only a few simple rules necessary to ensure, or at least to tend to improve, their chances for survival. The human model, by contrast, includes not only the objects of everyday life (“reality”) but also an extremely elaborate and powerful set of rules capable of working radical transformations to suit human purposes and goals. Such a set of rules is called a syntax.

Fasten, if you will, on Jacob’s last phrase, where he switches to a human example, and let us at last return to Lorenz’s expression, “syntactic language.” A language, as the term is used here, refers, in its evolutionary beginnings, not to an external communication system at all, but rather to a modeling system. (In passing, it might be worth noting that Soviet scholars, independently, came to this same conclusion in the early 1970s.)¹²

While all speechless creatures, as Jacob emphasizes, must model their universe, in order to survive, in fair conformity to “what is really out there”—this is, in the end, the deep meaning of the notion of an Umwelt—humans have evolved a way of modeling their universe in a way that not merely echoes “what is out there” but can, additionally, dream up a potentially infinite number of possible worlds (the phrase and idea hark back to Leibniz).

This kind of capability is achievable solely by means of a language, such that sentences in that language can be decomposed and recomposed in an indefinite number of ways (given a sufficient number of pieces to start out with); we can do this because all our natural languages possess (as do the genetic code, which is the language of life, and the immune code, which we now know constitutes the biological language of the self) a syntactic component.

By means of such a syntax, we can construct numberless novel narratives, imagine many versions of the past and construct as many future scenarios (including those of our death and afterlife), lie, frame scientific hypotheses (such are Newton’s and Einstein’s great models of the universe), including hypotheses about language itself, create poetry—in short, build a civilization. (It was Niels Bohr who first emphasized the doctrine that scientists have no concern with “reality”; their job has to do with model building.) Putting the matter somewhat more technically, in addition to making declarative sentences, we can construct as many conditionals (“what ifs”) as may be called for by an infinite variety of circumstances. No other animal, so far as we know, has language in this sense.

There are two points in the foregoing that are not currently controversial (save perhaps at the lunatic fringes of science), and then there is a third one that is passionately so. The first two are that (1) all living creatures, therefore all animals, can and do communicate; and that (2) no animal other than Homo sapiens is endowed with speech. (Although everybody agrees now that the latter is so, such was not the case earlier in the century: in 1916 William H. Furness III of Philadelphia tried, but dismally failed, to teach an orangutan to speak; and even as late as the 1950s Keith and Cathy Hayes of Florida lavished six fruitless years trying to coax a chimp to speak.)¹³

The hotly contested issue—the one which each of the four books under review must come to grips with and take a stand on—is the question whether language (in the sense used above, not, of course, metaphorically) does or does not occur in any species outside the genus Homo.

2. THE CAST OF CHARACTERS

The search for talking animals is ancient and ubiquitous. It pervades our myths and our literature, down to Dr. Dolittle and one of the late Bernard Malamud’s last fantasies, God’s Grace (1982). As Calvin Cohn, the sole human survivor of a thermonuclear war, trying to fashion the island on which he has survived only in the company of baboons, chimps, and one gorilla into a functioning social community, with himself as advisor and protector, daydreams in this novel: “You are not the chimps your fathers were—you can talk. Yours, therefore, is the obligation to communicate, speak as equals, work and together build, evolve into concerned, altruistic living beings.”¹⁴

Allegedly language-endowed or calculating animals and the like have also long figured as living props in exhibitions by fairground entertainers and miscellaneous con artists, demonstrating over and over again—in the words of Joseph Jastrow (who, by the way, seems to have held the first Ph.D. granted in psychology in America)—“how a simple humanizing error in observation under a prepossession can compromise rationality.”¹⁵

Before the 1950s, this sort of generally amiable con was played out typically using domestic animals, such as dogs and cats, horses and pigs, geese or other birds, and sometimes even fleas, as stage props.¹⁶ The famous case of Clever Hans—solidly enshrined in standard textbooks of psychology—was perhaps the most notable episode in the annals of such illusions, or errors (technically ascribed to “experimenter bias”), that offered an explanation of the phenomenon, not as one caused by deliberate deception, but rather as one due to mere innocent self-deception.

Hans was a stallion (in Berlin, around the turn of the century) that answered by stomping the correct number of times when asked even very complicated arithmetical, spelling, or other kinds of questions. The investigator of the phenomenon, a farseeing German psychologist named Oskar Pfungst, found that Hans responded to subtle, unwitting visual cues provided unknowingly by his owner or by the audience participants present at the demonstrations.¹⁷ Alas, a confession by Hans’s groom has just been uncovered; this rascal is quoted in an interview as having blurted out: “Clever Hans, he is really myself. When I lower my eyes, then the horse stomps until I raise my eyes again.”¹⁸

Beginning about 1960, the playing stakes were significantly raised, as a result of the abrupt infusion of government funding required to support a new habit. This unprecedented financial support, largely on the part of the taxpayers, accompanied a shift from domestic animals as targets of language-instillation experiments to those involving remote, hitherto alien—therefore extremely expensive—creatures in our midst. At first, these were marine mammals, notably the bottlenosed dolphin, Tursiops truncatus, soon followed by other kinds of whales (and, eventually, by the California sea lion, which is a pinniped).

In the next heady decade and a half, four species of apes became the objects of such studies: two sorts of chimpanzees, the gorilla, and the orangutan.¹⁹ Here only ape language games can be discussed, as these form the subject matter of the books under review. However, while the tacit claim which underlies all of the latter is the truism that efforts to inculcate quasi-linguistic skills in the great apes, being “Our Closest Living Kindred Species” (Savage-Rumbaugh, p. 4), somehow sanctifies the lavish funding required to assist in the discovery of human origins, no such profession will hold water (as it were) when it comes to studies of off-shore and pelagic animals distantly related to us, that is, related only by virtue of their being fellow mammals. Consequently, a sharp shift in the justification for working with such creatures becomes necessary; now attention is focused on the sensory modalities rather than on the cognitive preadaptations, as, for instance, in the following: “The human and the dolphin share the capability of communication by means of sounds. The chimpanzee and the gorilla cannot do this with any degree of complexity whatsoever.”²⁰ This is also a truism, of course, but two truisms don’t add up to a single truth.

As Linden plaintively remarks (p. 86), “from roughly 1976 onward, the ‘market,’ in the sense of financial and academic support for the work [with primates] began to contract.” By now, it has all but dried up. The reasons for this decline (apart from the scarcity of federal research funding overall) are twofold: first, the investigations were producing virtually nothing but negative results; and, second, all the best talent, having sensed this total lack of success before others did, nimbly faced up to it and wisely turned to more amenable scientific problems.

It might be worth noting, by the way, that ape language games were, and are, played out all but exclusively in the United States, presumably the only nation whose citizens could afford the price tag and thus tolerate them. One such project was begun at the University of Kyoto, but rapidly came to naught. An ape is, after all, the Cadillac among experimental animals; the French, on the other hand, opted for a deux-chevaux model, incarnated as a woodpecker, yet whose trainer claimed results “analogous to that found by the Gardners and Premack” with their exorbitant chimpanzees.²¹

Of the authors represented here, three are professional psychologists, and the fourth, Eugene Linden, is a journalist. After his two earlier efforts in this area, Apes, Men and Language (1974, 1981) and The Education of Koko (with Francine Patterson, 1981),²² Linden’s new book—which might perhaps have been titled more appropriately The Education of Eugene—comes as a distinct and pleasant surprise. In his first publication, written in his early twenties, he displayed a deplorable gullibility, one that seems to seize not only the media-enthralled public at large, but also most scribblers when they contemplate and then are moved to write about this subject.

It is easy to see why this should be so. As a well-known publicist once explained (speaking of another sensational topic), when a UFO descends in your backyard, that’s news; but when a UFO doesn’t land there, that’s no news at all. Similarly, when Koko the gorilla is reported “to tell stories, escape blame, make jokes, tell fibs,” and the like (p. 121), that’s news; but when Koko merely imitates the gestures of her keeper, Ms. Patterson—“all of Koko’s . . . utterances . . . were prompted,” Terrace repeatedly tells us (for example, p. 221)—that’s simply another illustration of the adage “Monkey see, monkey do.” It is hardly worth sounding off about so commonplace an event. Notwithstanding these obvious facts, Linden had collaborated with Patterson on what is surely one of the silliest books in the age-old literature about “talking” animals.

It is to Linden’s credit, however, that in this new book, which “involved a painful reexamination of events” he had written about before (p. xi), he retreats, on a broad front, from his previous positions. He now confesses (pp. 124-127) that Patterson’s (and, of course, his own previous) claims were “catastrophic in terms of credibility . . . because the context of the scientific debate is so far removed from the sophisticated behaviors” attributed to this pathetically sequestered pet gorilla; and then he goes on to admit (doubtless still too charitably) that this “impression is compounded by the fact” that Koko’s trainer “has on occasion stretched her interpretations of Koko’s gestures.”

Little wonder, then, that Patterson “has had a procession of assistants who have arrived starry-eyed with anticipation and have left more or less disillusioned.” This last point must, alas, be generalized: the field is littered with disenchanted “former assistants” from this and several of the other major projects. A most reprehensible consequence seems to have escaped Linden: namely, that the senior investigators supposed to provide for the training of their successors have in fact abdicated this common academic responsibility. This explains why this game is played by fading stars: there are no promising starlets, with novel ideas or improved research designs, anywhere in the visible firmament.

In writing this book, Linden has exhibited courage of a sort Kipling called “grace under pressure.” But while it is encouraging to be able to report that his education has become more sophisticated, it is, all the same, troubling to reflect on the length of the maturation process. After all, the strictures he recognizes only now—not just about his former coauthor’s labors, but also about many other aspects of the ape-language illusion—were foreseen at least six years ago by others.²³ (Of course, in the interval, he did busy himself with the preparation of two well-selling books.)

Unsaying is the hallmark which stamps not only Eugene Linden but also the scientists whose work he recounts. Take only the three psychologists under review here:

David Premack: “As early as 1970, I essentially quit concentrating on the attempt to operationally analyze some aspects of human language, develop training procedures for them and instill them in the ape, because it was clear to me that the accomplishments of which the ape was capable with regard to human-type language were very slight. . . .”

E. Sue Savage-Rumbaugh: “Frankly, we are not interested in whether or not language is the exclusive domain of man. That question leads all who address it into a quagmire of confusion, despair, and impatience. We want none of that!”

Herbert S. Terrace: “When I began my study with a male chimp called Nim Chimpsky, I hoped to demonstrate that apes can, indeed, form sentences. I wanted to . . . show that grammatical rules are needed to describe many of an ape’s utterances. . . . I discovered that the sequences of words that looked like sentences were subtle imitations of teacher’s sequences. I could find no evidence confirming an ape’s grammatical competence, either in my own data or those of others, that could not be explained by simpler processes.”²⁴

Terrace has been severely chastised by the faithful for his recantation, as Linden, for one, reports (pp. 65-69, 236). Nevertheless, the proper method of science is to formulate falsifiable hypotheses, and then to search vigorously for negative instances. This is precisely what Terrace has done: he has posited a hypothesis—one that, to be sure, was truly preposterous, albeit not wholly impossible (to wit, that apes are capable of forming sentences, that is, that they do possess language propensity); found, in the course of his empirical researches (described in Nim), essentially only negative instances (data from his own subject as well as those of his from fellow investigators, including B. T. Gardner and R. A. Gardner, Patterson, Lyn Miles, and Savage-Rumbaugh), thereby refuting his prediction; and finally proposed a much simpler explanation. As Popper emphasized, “It is through the falsification of our suppositions that we actually get in touch with ‘reality.’ ”²⁵

So far, so good. But now the plot takes on an unexpected, not to say Byzantine, twist. The selfsame Terrace, the man who had insisted, as late as 1981, that the study of syntax in apes be abandoned on the grounds that “multiple sign combinations are unlikely to provide meaningful evidence of syntax,” since the signs made by chimpanzees known to him “were simply repetitions made shortly before by the animal’s trainer,”²⁶ turns up reincarnated as a scientific consultant for the selfsame “we-want-none-of-that” Rumbaughs, who, with renewed and generous public funding, are happily back in the language business. (Even the innocent Linden sniffs at this turn of events [pp. 70-71]!)

With this news, we have, however, reached merely the threshold of a brand new chapter in the seemingly never ending saga about animal language. In 1986, the Columbia University Press announced the start of a new series, edited by Herbert S. Terrace, devoted to the subject of “animal intelligence.” Yet the first volume in this series is on the subject of ape language and written by—you guessed it—E. Sue Savage-Rumbaugh, whose “discoveries,” Terrace confidently trumpets as he relentlessly draws us back to the very “quagmire” out of which we have just barely escaped, “should add significantly to our understanding of the origin and nature of language” (p. xix). The question therefore arises: what is meant by intelligence, especially in a zoological context, and what is the relation of that to language?

According to the classic accounts of Lenneberg, “The definition and measurement of intelligence is difficult enough in our own species. When it comes to comparing different species, it is no longer permissible to talk about intelligence as if it were a single, clear-cut property that can be measured by a single objective instrument so as to yield quantities that are commensurable across species. . . . I would like to propose . . . that the ability to acquire language is a biological development that is relatively independent of that elusive property called intelligence.”²⁷

In 1983, Howard Gardner went much further still, suggesting that “intelligence” is but an indeterminate concept of common sense which may well not refer to “a genuine tangible, measurable entity,” but rather be “a convenient way of labeling some phenomena that may (but may well not) exist.” In fact, Gardner, in his tour de force synthesis, posits a set of up to seven human intellectual potentials, “of which all individuals are capable by virtue of their membership in the human species.”²⁸

As if this kind of vagueness weren’t damaging enough, Terrace, in introducing his new series, scarcely mentions “intelligence,” dwelling instead on the currently faddish topic of animal thinking, and the continuity or discontinuity thereof with human thinking, and on the putative “linguistic competence of animals with highly developed brains.” He thus so thoroughly confounds apples (“intelligence”) and oranges (“thinking”) and bananas (“language”) with one another that his entire discussion becomes impossibly vacuous.

If this kind of hodge-podge is what Linden refers to in his remark (p. 236) that today “there is a veritable renaissance in the study of cognitive behavior of all types of animals” and what Terrace has in mind when he predicts that “the coming decades may prove to be the beginning of an era of psychology that will parallel the era of the blossoming of biology” (“Foreword,” in Savage-Rumbaugh 1986:xii), both writers are bound to be gravely disappointed. It would be far wiser to heed John Stuart Mill’s admonition (in his 1867 inaugural address at the University of St. Andrews) “To question all things . . . , letting no fallacy, or incoherence, or confusion of thought, step by unperceived; above all, to insist upon having the meaning of a word clearly understood before using it, and the meaning of a proposition before assenting to it.”

But there is more. Although the Savage-Rumbaugh volume inaugurates a series about “animal intelligence” and features the noun “language” in its title, it compounds the terminological inexactness already alluded to by throwing in—although never defining—an arguably even more heavily freighted word: “symbol” (throughout, but esp. Ch. 11). Savage-Rumbaugh seems to think that the teaching of symbols to an ape constitutes some sort of a breakthrough. In point of fact, instances of the sign category that philosophers (especially since Peirce) routinely call “symbols” occur naturally not only throughout the animal kingdom,²⁹ but, as the late Gordon M. Tomkins brilliantly revealed, unique states of the environment can also be represented by symbols in the metabolic code that governs processes inside animal bodies.³⁰

A desperate hunger for priority, coupled with acute anxiety over research funding and a generally insatiable craving for favorable publicity (Terrace cites “public opinion,” p. xv), is doubtless responsible for the exceptionally venomous and unforgiving nature of the discourse among the handful of psychologists still extant in the field of animal-language research. Threats of legal action resound. These men and women can’t quite succeed, it seems, without also constantly castigating, or putting down, their peers; even the (early) Terrace is ritually criticized, until his “colleagueship”—presumably coincidentally with his consultantship—is welcomed (pp. 375-379) by Savage-Rumbaugh and other members of her team. (Likewise, by the way, Premack excoriated what I take to be Savage-Rumbaugh’s core project, which claims to have established “symbolic communication” between two her subjects and to which Chs. 7-11 of her book are mainly devoted.)

This palpably rancorous, politically charged atmosphere “has taught us,” the Rumbaugh team importunes (p. 383), “that scientists need to work harder both to establish and to retain a sense of community,” for “efforts on the part of some researchers to preempt or to preclude the entry of others into the field of ape-language research was misguided and unfortunate in that they put at severe risk the probability that ‘community’ and the exchange of data between laboratories might ever be attained.”

It is certainly true, as Gerald Holton reminds us, that “the first principle of integrity is that you must submit yourself to the dialogue with others to find out whether you are right. New science starts in the head of an individual, but it does not survive if it does not become part of the consensus of the community.”³¹ Unfortunately, as became apparent, for instance, to the organizers who tried to assemble a reasonably representative cast of characters for a conference held under the auspices of the New York Academy of Sciences (see note 23), most of the invited ape-language investigators either refused to confront their colleagues or accepted an invitation but failed to show; and those who did show (for example, Duane and Sue Savage-Rumbaugh) had their say but left before their colleagues had an opportunity to respond.

It should further be pointed out that the criticisms of this or that project are typically directed at the underlying logical design of the experiments and the results announced. Personal inspections in working laboratories—except by spectators known in advance to be sympathetic—are not tolerated, and the most outlandish excuses for such exclusions are invented. (One famous animal psychologist once reported, with much amusement, that he was forbidden to visit one of the best-known ongoing experiments on the grounds that he might infect the chimpanzee with viral hepatitis, which the man had never had and which is not, in any case, aerobically transmissible.)

The title Gavagai! echoes a coinage by Willard Van Orman Quine,³² which has to do with subtle arguments by that eminent philosopher about the indeterminacy of translation from one language into another. Given the insuperable problems raised by such interlingual operations, Premack asks, what problems might be raised by (to use a more sophisticated terminology) intersemiotic transmutations, that is, when decoding “the chimpanzee’s would-be words or sentences”? His answer: none other than “we do with our language peers” (p. 8).

Premack’s own “linguistic” work with Sarah and his other chimpanzees has been severely censured. As noted above, and as he himself announced, he had abandoned his initial goals sixteen years earlier. One of the points at issue had been his lack of proper measures against social cueing, that is, the intrusion of the Clever Hans phenomenon, but which he considers “a red herring,” not “a legitimate scientific issue.” He feels that his critics should have pointed out “what a proper clever-Hans control would look like” (p. 12).

Unfortunately, Premack either does not know the huge literature on the subject or he chooses to ignore it. Admittedly, the pioneering work of Oskar Pfungst, in the first decade of this century, was flawed, but Premack never mentions the subsequent voluminous writings of Heini Hediger, or the book of Ernst Timaeus; worst of all, he disregards the far-reaching researches of Harvard’s Robert Rosenthal on the very nature of experimentation itself.³³

Premack concludes his essay with the finding—and, as to this, one must surely concur with him, for it brings us round precisely to where we started our own protracted journey—of “not only a discontinuity between human and nonhuman” but also “the lack of any degree of language among nonhumans” (p. 149). The last sentence of his paper then reads, in part: “To get on with an understanding of our species, we shall have to relinquish our infatuation with language” (p. 155). Ape-language researchers, above all others, please take his admonition to heart! And please remember one other, the sage advice of P. B. Medawar to scientists of any age: “The intensity of the conviction that a hypothesis is true has no bearing on whether it is true or not.”³⁴

Before ending this perhaps overlong chapter, a possible misunderstanding needs to be anticipated: the more or less simultaneous appearance of several books devoted to a similar topic must not be taken to mark a resurgence of animal “language” studies. Terrace’s Nim is a reprint of a book already seven years old; Savage-Rumbaugh’s Ape Language mostly brings together materials already published elsewhere, although here rhetorically refurbished; Linden’s apostasy tidies up others’ works; and Premack’s piece is but a slightly altered (and, in the book, unacknowledged) version of an earlier magazine piece bearing the same title.³⁵

Notes

1. Konrad Z. Lorenz, The Foundations of Ethology (New York: Springer, 1981), 342.

2. Nicholas Toth, “The Oldowan Reassessed: A Close Look at Early Stone Artefacts,” Journal of Archaeological Science 12(1985):114.

3. Nicholas Toth, “Archaelogical Evidence for Preferential Right-handedness in the Lower and Middle Pleistocene, and Its Possible Implications,” Journal of Human Evolution 14 (1985):612.

4. Dean Falk, “Language, Handedness, and Primitive Brains: Did the Australopithecines Sign?” American Anthropologist 82(1980):72-78.

5. Noam Chomsky, Rules and Representations (New York: Columbia University Press, 1980), p. 230. The philosopher Karl K. Popper makes a similar point when he speaks of the “futility of all theories of human language that focus on . . . communication ”; see his Objective Knowledge: An Evolutionary Approach (Oxford: Clarendon Press, 1972), 121.

6. A large variety of such repertoires is described in, for instance, Thomas A. Sebeok, How Animals Communicate (Bloomington: Indiana University Press, 1977).

7. Gregory Bateson, “Redundancy and Coding,” in Thomas A. Sebeok, Animal Communication (Bloomington: Indiana University Press, 1968), 614.

8. These are discussed by G. C. Williams, in his classic book, Adaptation and Natural Selection (Princeton: Princeton University Press, 1966).

9. Stephen Jay Gould and Elisabeth S. Vrba, “Exaptation—A Missing Term in the Science of Form,” Paleobiology 8(1982):13.

10. See especially his The Theory of Meaning, republished from the German original (1940) in Semiotica 42:1.1-87 (Special Issue).

11. François Jacob, The Possible and the Actual (Seattle: University of Washington Press, 1982), 56.

12. This concept is discussed, for instance, by Ju. M. Lotman, in his “Primary and Secondary Communication-Modeling Systems,” originally published in Russian in 1974, but available in English in Daniel P. Lucid’s anthology, Soviet Semiotics (Baltimore: The Johns Hopkins University Press, 1977, 1988), 95-98.

13. For details, see the entertaining book by Adrian J. Desmond, The Ape’s Reflexion (New York: Dial Press/James Wade, 1979).

14. God’s Grace (New York: Farrar Straus Giroux, 1982), 127.

15. Wish and Wisdom: Episodes in the Vagaries of Belief (New York: D. Appleton-Century, 1935), 213.

16. Such scams are well documented since Elizabethan times. See the excellent account by the outstanding magician, Milbourne Christopher, ESP, Seers & Psychics (New York: Thomas Y. Crowell, 1970), 39-54.

17. See Dodge Fernald, The Hans Legacy: A Story of Science (Hilldale: Lawrence Erlbaum, 1984).

18. See Thomas A. Sebeok, “A Scientific Quibble,” Semiotica 57(1985): 123.

19. Known in the zoological literature, respectively, as Pan troglodytes, Pan paniscus (the pigmy chimp), Gorilla gorilla, and Pongo pongo.

20. So says John C. Lilly, in his Communication between Man and Dolphin: The Possibilities of Talking with Other Species (New York: Crown, 1978), 78.

21. On the Gallic Greater Spotted Woodpecker, see Bernadette Chauvin-Muckensturm, “Y a-t-il utilisation de signaux appris comme moyen de communication chez le pic epeiche?” Revue du comportment animal 9:185-207. There are also reports of experiments, in this country, with parrots and pigeons, showing, by and large, embarrassingly comparable results to those accomplished with the primates.

22. Reviewed by Thomas A. Sebeok under the title “The Not So Sedulous Ape,” Times Literary Supplement (September 10, 1982). The fuller, original version is to be found in Thomas A. Sebeok, I Think I Am a Verb (New York: Plenum Press, 1986), 205-208.

23. For example, see Thomas A. Sebeok and Jean Umiker-Sebeok, Speaking of Apes: A Critical Anthology of Two-Way Communication with Man (New York: Plenum Press, 1980), esp. 1-59; and, afterward, many of Terrace’s publications, such as “A Report to an Academy, 1980,” in Thomas A. Sebeok and Robert Rosenthal, The Clever Hans Phenomenon: Communication with Horses, Whales, Apes, and People (New York: The New York Academy of Sciences, 1981), vol. 364, pp. 94-114.

24. Premack, in a debate with Noam Chomsky, “Species of Intelligence,” The Sciences 19(9):8. Savage-Rumbaugh, with Duane Rumbaugh, writing in 1979, quoted by Linden, p. 58. Terrace, in a characteristically titled piece, “How Nim Chimpsky Changed My Mind,” Psychology Today 13(6):65, 76.

25. Popper, Objective Knowledge: An Evolutionary Approach, 360.

26. In D. R. Griffin, ed., Animal Mind—Human Mind (Berlin: Springer Verlag, 1982), 402.

27. Eric H. Lenneberg, ed., New Directions in the Study of Language (Cambridge: MIT Press, 1964), 77-78. Lenneberg greatly elaborated on these statements in his very influential book, Biological Foundations of Language (New York: Wiley, 1967), esp. 228-230.

28. Frames of Mind: The Theory of Multiple Intelligences (New York: Basic Books, 1983), 69, 278.

29. For further discussion, definitions, and examples, see Thomas A. Sebeok, Contribution to the Doctrine of Signs (Lanham: University Press of America, 1985), 134-138.

30. “The Metabolic Code,” Science 189:760-763.

31. “Niels Bohr and the Integrity of Science,” American Scientist 74(3):240.

32. Word & Object (Cambridge: MIT Press, 1960), Ch. 2.

33. Among the extensive writings of Heini Hediger on this general subject, but including ape language research as well, see, most recently, his Tiere Verstehen: Erkenntnisse eines Tierpsychologen (Munich: Kindler, 1980), esp. the chapter “The Return of Clever Hans,” 112-160. One of the few psychologists who tried to replicate (partially) the original experiments with Clever Hans was Ernst Timaeus; see Experiment und Psychologie: Zur Sozialpsychologie psychologischen Experimentierens (Gottingen: Hogrefe, 1974). Robert Rosenthal’s books and papers are too numerous to list here, but see esp. his Experimenter Effects in Behavioral Research (New York: Irvington, 1976). For one review of the Premacks’ The Mind of an Ape, see the Times Literary Supplement, June 29, 1984.

34. Advice to a Young Scientist (New York: Harper & Row, 1979), p. 39.

35. This will be found in Cognition 19:207-296. At least three errors need to be corrected in the 1986 version: the article by Miles, referred to on p. 29, was published already in 1983, by Springer. “Petitio,” named on p. 33, is actually Laura A. Petitto. And the “Menzle,” who appears twice on p. 141, is in fact the primatologist E. W. Menzel.

________

This chapter first appeared as an illustrated review article commissioned by the editors of The World & I (October 1986, pp. 462-469). It was loosely constructed around four books: Linden (1986), Premack, (1986), Savage-Rumbaugh (1986), and Terrace (1986 [1979]). An expanded version then appeared in Semiotica 65 (1987):343-358.