Human Language—Human Consciousness

The languaged human mind begins its development prenatally, forming a sensitivity to the sounds and rhythms of the human mother even as it constructs itself. As soon as the infant emerges, it is imperative that the mother, or another, support it and keep it warm or it will die quickly. Unlike all other primates, it cannot cling. This means that a human mother cannot treat the infant as an extension of her own body as she moves through space. From its very entrance into the world, the human infant undergoes a radically different self-other embodied experience than a bonobo infant. Bonobo infants are extensions of the mother’s body within minutes after birth. The human baby is, instead, an object that is swaddled, carried, passed around, rocked, spoken to, picked up, put down, attended to — and sometimes not attended to, but left to cry. It has little input to, or responsibility for, any of the adult activities directed toward it. Unlike ape infants, it does not need to attend to the maintaining contact with the mother by holding tight and paying attention to her slightest movement to prepare itself for travel. Indeed, the only means available to it for maintaining parental contact is to signal by crying. Human parents are far more likely to ignore loudly crying babies than bonobo mothers are to ignore the soft whimpers of their babies. If a human baby is inside a hut and relatives are watching outside, predators who hear the cry will not gain access to the baby. Human babies can afford to cry; their survival does not depend upon being quiet. Bonobo babies cannot. They must adopt a kind of stoicism from birth.

Infant clinging has been characterized as a simple ‘innate response,’ but this is far from the truth. It requires very skillful co-ordination with the bonobo mother, and the baby must make constant adjustments to the mother’s postural changes. Once a bonobo infant learns to cling and to anticipate what the mother will do, very little conscious monitoring of the baby is necessary, on the part of the mother. Human infants are too heavy to cling at birth and, unlike bonobos infants, they continue to gain weight quickly, making it impossible for them to cling even if they were to become strong enough to do so. In addition, their feet and the size of their heads relative to the strength of their necks renders clinging hopeless. Human mothers must support the large head, weak neck and mobile spine in careful manner.

When a human infant is placed on a substrate other than the mother’s body, a very different visual, auditory, tactile and kinesthetic world impinges upon it. The bonobo baby is, during this same time, looking at a sea of hair and trying to stay with the mother. The human baby is looking at the mother’s face while observing her mouth move in synchrony with the sound that is coming to its ears as she speaks. Babies neurologically entrain to the mouth-movements, facial expressions and sounds of others; and they engage in facial-vocal dialogues with caretakers by two to three months (Stern, 1971, 1977; Trevarthen, 1978, 1998). These entrainment periods evoke rhythmic similarities to later ‘true conversations’. New born infants also ‘imitate’ facial expressions, such as tongue protrusion and smiles, from a few days after birth (Meltzoff, 1996; Myowa-Yamakosi, Tomonaga, Tanaka and Matsuzawa, 2004). The early onset of such behaviors suggests that they may be guided by the mirror neuron system, and that the inter-individual neurological mirroring is intimately tying babies to their caregivers (Rizzolatti, Fogassi, and Gallese, 2001). Thus the emotional and attentional world of the caregiver is patterning that of the infant, in both apes and humans, but in very different ways because the human infant does not cling. The bonobo baby is designed to quickly become a direct extension of the mother and to link tightly into her physical movements through space. The human baby is designed to be sat down from time to time and to be swaddled; and thus its attention focuses on the objects around it, and what others are doing with them.

When a human mother sets the baby down, even for a moment, if she is attending only to her own tasks, something can happen to the child. The bonobo mother, by contrast, needs only to watch out for her own welfare, as by so doing she assures also the welfare of the baby holding tightly to her. As a human baby is put down, it is exposed to the sun, rain, smoke, insects, wind or predators that could harm it quickly. The human mother must create in her mind a constant scenario of the baby’s needs, which differ from her own. This is demanding for most new human mothers. The mother’s emotional reaction to this constant need affects the content and style of the linguistic dialogue which begins to accompany her actions toward the baby (Oh, you’re all wet, You are getting sleepy, Now smile, don’t be fussy, What is bothering you?, Stop crying, Why don’t you smile at me? — etc.). Her dialogue carries a quality of aboutness with it, toward the baby. The degree of and tone of the aboutness expressed corresponds directly to the degree of differentiation taking place between mother and infant (Brigaudiot, Morgenstern, and Nicolas, 1996; Savage-Rumbaugh, 1990). Any sounds made by bonobo mothers are typically about what others in the group are doing, not about what the baby is doing.

These different styles of caregiving are molded within the first two years of life and give rise to adult social structure. Infants raised by either pattern eventually develop some degree of self-awareness and reflective thought. But the objectification of the self is fostered far earlier by human rearing and human language patterns. The bifurcation of self into “I” and “me” becomes elaborate in the human adult as it is embedded in language directed to the child from birth. However, by the time the bonobo infant has begun to develop its sense of self-agency, its primary focus of attention has already been unalterably fixated upon group cohesion and group movement through the terrain. It has formed a formidable spatial/temporal map of its world and it has become inextricably embedded within its landscape and its social group. Time spent in self-reflection is minimized by bonobo culture, as this culture places a higher priority on group survival and shared knowledge than on individual survival or private knowledge. Human cultures value individual knowledge, individual competency, self-narrative and self-justification.

These differences in maternal care patterns, which human and bonobo infants experience from the moment of birth forward, are responsible for the many of the behavioral distinctions that later emerge between the species. They are culturally instantiated and are open to change at any time. From these caregiver patterns emerge the different styles of human/ape consciousness which have captivated philosophers and psychologists (Kitcher, 2006; Locke, 1690/1959; Savage-Rumbaugh, Fields and Segerdahl, 2005). However, we should recognize first, that it is possible for ape mothers to exhibit more human-like patterns (which they do when they have an infant that cannot cling) and second, that it is possible for human mothers to rear ape infants who do not cling because they are carried and treated precisely as human infants would be treated. Therefore the apparent differences between humans and apes are not biologically fixed, but they are biologically and culturally instantiated.

When apes experience human-rearing and are exposed to a human language they begin to display the human patterns of self-awareness and self-reflection by 6 months of age. An obvious index of self-awareness is the use of a mirror to view the self as the self is being intentionally altered (or immediately after it has been altered). Many apes explore their image by seeking out a mirror to look at their teeth, their tongue, their ears, their eyes and other portions of their body that could be observed only in a reflected image. Linguistically competent apes expand this awareness by beginning earlier and by elaborating. They paint their faces, put on wigs, shawls and monster masks, and rush to the mirror to see how the look. They try to blow bubbles with bubble gum while using mirrors to watch their cheeks. They practice displays by adding fur capes as they swagger in front the mirror. They seek out live video images to see things that even a mirror would not reveal. Only a live camera image can reveal their epiglottis and allow them to learn to vibrate it in real time (Menzel, Savage-Rumbaugh, and Lawson, 1985; Savage-Rumbaugh, 1986).

Their concern with understanding the appearance of the self from the perspective of another arises from the bifurcated, or dualistic, view of the self, whose roots lie in the I/Me distinctions embedded in the structure of the human language which they are acquiring. The doer/viewer duality of consciousness enables the youngster to think about what it is doing, the appearance of its action, and/or how the action will be perceived by others — all at the same time. When this dualistic process begins to operate, there emerges, within a single brain and body, the capacity to consciously separate the imaged self into that of the doer of one’s actions and the viewer of those same actions (Bates, 1990). The viewer begins to sometimes hold an action by the doer in abeyance, or sometimes even to reflect upon the past actions of the self as doer with a certain amount of chagrin and dismay. This is the formative basis of mental time travel and the mental construction of alternative world views (Suddendorf and Corballis, 2009).

Every act of the viewer is like a tiny seed that grows and expands very slowly at first, but with increasing maturity it takes over more and more of the body’s actions until the adult individual emerges with a highly self-reflective consciousness, capable of moral agency. This capacity for dual consciousness, or bifurcated thought, lies at the root of Cartesianism, but there is no need to reduce the bifurcations of consciousness to mind/body dualism. When consciousness, as an emergent property of the neural system, reaches a certain level of self-awareness, or awareness of self as causal agent, it becomes able to bifurcate. As it does so, it gains the ability to self-reflect. Metaphorically speaking, consciousness splits into two parts, like a soap bubble that becomes two halves, by the appearance of film in the middle. One half of the ‘consciousness bubble’ specializes in guiding the immediate actions of the organism and the other half specializes in reflecting upon those actions. Through this lens of reflected consciousness, one sees one’s self as a causal agent outside one’s immediate tendency to react. Human language, coupled with human maternal care, enables the consciousness to bifurcate very early and extensively. Without the self-reflective properties inherent in a reflexive agent- recipient language, and without the objectification of the human infant — a very different kind of humanity would arise.

Human consciousness, as constructed by human language, becomes the vehicle through which the self-reflective human mind envisions time. Language enables the viewer to reflect upon the actions of the doer (and the actions of one’s internal body), while projecting forward and backward — other possible bodily actions — into imagined space/time. Thus the projected and imagined space/time increasingly becomes the conscious world and reality of the viewer who imagines or remembers actions mapped onto that projected plan. The body thus becomes a physical entity progressing through the imaged world of the viewer. As the body progresses through this imaged world, the viewer also constructs a way to mark progress from one imagined event to another. Having once marked this imagined time into units, the conscious viewer begins to order the anticipated actions of the body into a linear progression of events.

A personal narrative then arises through the vehicle of language. Indeed a personal narrative is required, expected and placed upon every human being, by the very nature of human language. This personal narrative becomes organized around the anticipated bodily changes that it is imagined will take place from birth to old age. The power of the bifurcated mind, through linguistically encoded expectancies, shapes and molds all of human behavior. When these capacities are jointly executed by other similar minds — the substrate of human culture is manufactured. Human culture, because it rides upon a manufactured space/time self-reflective substrate, is unique. Though it shares some properties with animal culture, it is not merely a natural Darwinian extension of animal culture. It is based on constructed time/space, constructed mental relationships, constructed moral responsibilities, and constructed personal narratives — and individuals, must, at all times, justify their actions toward another on the basis of their co-constructed expectancies.

Human Consciousness seems to burst upon the evolutionary scene in something of an explosion between 40,000 and 90,000 years ago. Trading emerges, art emerges, and symboling ability emerges with a kind of intensity not noted for any previous time in the archeological record. But if Darwin was correct, humanity must have descended from ape-like beings. The problem is that there seem to have arisen a few inadequacies in the Darwinian account. The first is that apparently — if the current fossil record is to be believed — there are no recent fossil apes, only fossil humans that look a lot like apes. Everyone assumes that recent fossil apes will eventually be found, but no one ever finds them. The second problem is that living apes appear, at first blush, to lack any semblance of what we believe to be important human attributes: language and rational thought. No wild apes wear clothes, construct dwellings, paint their bodies, sing, count or tell stories, as far as we know. Yet our anatomy, as compared with that of living apes, leaves no doubt as to the closeness of our kinship — a fact confirmed with the sequencing of the chimpanzee genome in 2005. The closeness of human and chimpanzee DNA varies depending on how you measure it, but regardless of the percent of relatedness one assigns, it clearly reveals that humans and apes are far closer than anyone had dared to speculate on the basis of anatomical characteristics alone. No one understood, from the standpoint of natural selection, how human beings could evolve so quickly from an apish mentality to a world of symbols, syntax, reasoning, and complex planning. These essential human characteristics were not simply an elaboration of the ape mode of life — as Darwinian theory would predict. They represented a fundamentally new form of self-conscious life, empowered by a reflexive language that separated me from thee, past from now, and him from her.

Humans came with a propensity to alter the world around them wherever they went. We were into object manipulation in all aspects of our existence, and wherever we went we altered the landscape. We did not accept the natural world as we found it — we set about refashioning our worlds according to our own needs and desires. From the simple act of intentionally setting fires to eliminate underbrush, to the exploration of outer space, humanity manifested the view that it was here to control its own destiny, by changing the world around it, as well as by individuals’ changing their own appearances. We put on masks and masqueraded about the world, seeking to make the world conform to our own desires, in a way no other species emulated. In brief, the kind of language that emerged between 40,000 and 90,000 years ago, riding upon the human anatomical form, changed us forever, and we began to pass that change along to future generations.

While Kanzi and family are bonobos, the kind of language they have acquired — even if they have not manifested all major components yet — is human language as you and I speak it and know it. Therefore, although their biology remains that of apes, their consciousness has begun to change as a function of the language, the marks it leaves on their minds and the epigenetic marks it leaves on the next generation. (Epigenetic: chemical markers which become attached to segments of genes during the lifetime of an individual are passed along to future generations, affecting which genes will be expressed in succeeding generations.) They explore art, they explore music, they explore creative linguistic negotiation, they have an autobiographical past and they think about the future. They don’t do all these things with human-like proficiency at this point, but they attempt them if given opportunity. Apes not so reared do not attempt to do these things.

What kind of power exists within the kind of language we humans have perfected? Does it have the power to change biology across time, if it impacts the biological form upon conception? Science has now become aware of the power of initial conditions, through chaos theory, the work of Mandelbrot with fractal geometric forms, and the work of Wolfram and the patterns that can be produced by digital reiterations of simple and only slightly different starting conditions. Within the fertilized egg lie the initial starting conditions of every human. We also now realize that epigenetic markers from parental experience can set these initial starting conditions, determining such things as the order, timing, and patterning of gene expression profiles in the developing organism. Thus while the precise experience and learning of the parents is not passed along, the effects of those experiences, in the form of genetic markers that have the power to affect the developmental plan of the next generation during the extraordinarily sensitive conditions of embryonic development, are transmitted. Since language is the most powerful experience encountered by the human being and since those individuals who fail to acquire human language are inevitably excluded from (or somehow set apart in) the human community, it is reasonable to surmise that language will, in some form, transmit itself through epigenetic mechanisms.

When a human being enters into a group of apes and begins to participate in the rearing of offspring, different epigenetic markers have the potential to become activated. We already know, for example, that in human beings, expectancies or beliefs can affect gene activity. The most potent of the epigenetic markers would most probably arise from the major difference between human and ape infants. Human infants do not cling, ape infants do. When ape infants are carried like human infants, they begin to development eye/hand coordination from birth. This sets the developmental trajectory of the ape infant in a decidedly human direction — that of manipulating the world around it. Human mothers, unlike ape mothers, also communicate their intentions linguistically to the infant. Once an intention is communicated linguistically, it can be negotiated, so there arises an intrinsic motivation to tune into and understand such communications on the part of the ape infant. The ‘debate’ in ape language, which has centered around do they have or don’t they — has missed the point. This debate has ignored the key rearing variables that differ dramatically across the studies. Apart from Kanzi and family, all other apes in these studies are left alone at night and drilled on associative pairings during the day.

The question to be addressed is whether or not they can exhibit components of it when the care-giving patterns foster its emergence; and whether or not, as a result, epigenetic events take place which will enable future generations to begin to follow a developmental trajectory that is more human-like as a result. With the birth of Kanzi’s first infant, Teco, unusual effects are appearing. Teco exhibits clinging patterns that are distinctly human-like rather than ape-like. Consequently, his development is being affected not only by being carried so that he will not have to cling, but by being carried because he will not cling.


Bates, E. (1990). Language about me and you: pronominal reference and the emerging concept of self. In C. Cicchetti and M. Beeghly (Ed.), The Self in Transition (pp. 165-182).

Breidert, Matthias, and Karl Hofbauer. (2009). Placebo: Misunderstandings and Prejudices. Deutsches Aerzteblatt International, 106(4), 751-755.

Brigaudiot. M., Morgenstern, A., and Nicolas, C. (1996). Guillaume i va pas gagner, c’et d’abord maman: Genesis of the first person pronoun. In C. E. Johnson and J.H.V. Gilbert (eds.), Children’s Language (Vol. 9, pp. 105-116). Mahwah, NJ: Lawrence Erlbaum Associates.

Kitcher, P. (2006). Two normative roles of consciousness. In H. S. Terrace and J. Metcalfe (eds), The Missing Link in Cognition (pp. 174-187). Oxford: Oxford University Press.

Locke, J. (1690). An Essay Concerning Human Understanding. New York: Dover.

Meltzoff, A. N. (1996). The human infant as imitative generalist: a 20 year progress report on infant imitation with implications for comparative psychology. In J. C. M. Heyes and B. Galef (Eds.), Social Learning in Animals (pp. 347-370).

Menzel, E., Savage-Rumbaugh, E. S. and Lawson, J. (1985). Chimpanzee (Pan troglodytes) spatial problem solving with the use of mirrors and televised equivalents of mirrors. Journal of Comparative Psychology, 99(2), 177-185.

Myowa-Yamakoshi, M., Tomonaga, M., Tanaka, M., and Matsuzawa, T. (2004). Imitation in neonatal chimpanzee (Pan troglodytes). Developmental Science, 7(4), 437-442.

Rizzolatti, G., Fogassi, L., Gallese, V. (2001). Neurobiological mechanisms underlying the understanding and imitation of action. Nature Review Neuroscience, 2(9), 661-700.

Savage-Rumbaugh, E. S. (1986). Ape Language: From Conditioned Response to Symbol. New York: Columbia University Press.

Savage-Rumbaugh, E. S. (1990). Language as a Cause-Effect Communication System. Philosophical Psychology, 90(1), 55-77.

Savage-Rumbaugh, E. S., Fields, W. M., & Segerdahl, W. M. (2005). Culture prefigures cognition in Pan/Homo bonobos. Theoria, 20(3).

Stern, D. (1971). A micro-analysis of mother-infant interaction: Behavior regulation and social contact between a mother and her 3 1/2 month-old twins. Journal of the American Academy of Child Psychiatry, 10(3): 501-17

Stern, D. (1977). The First Relationship. London: Fontana/Open Books.

Suddendorf, T. and M. C. Corballis (2010). Behavioral evidence for mental time travel in nonhuman animals. Behavioural Brain Research 215(2), 292-8.

Trevarthen, C. (1998). The concept and foundation of infant intersubjectivity. In S. Braten (Ed.) Intersubjectivity, Communication and Emotion in Early Ontogeny (pp. 15-46). Cambridge: Cambridge University Press.

Trevarthen, C. a. H., P. (1978). Secondary intersubjectivity: confidence, confiding and acts of meaning in the first year. In A. Lock (Ed.), Action, Gesture and Symbol: The Emergence of Language. London: Academic Press.

19 comments to Human Language—Human Consciousness


    There is much to agree with in Sue Savage-Rumbaugh’s reflections on human and nonhuman primates. Sue has probably spent more real time rearing and observing our closest hominoid cousins than any other human being has done. Bonobos are indeed astonishingly intelligent and capable, and become still more human-like when reared in daily contact with humans.

    But there is one radical inference Sue makes that it will be hard for most people to agree with: Bonobos have acquired a (“kind of”) language: “the kind of language they have acquired — even if they have not manifested all major components yet — is human language as you and I speak it and know it.”

    Let us reflect for a moment on languages and kinds: Humans have many kinds of languages, but there is one thing all those languages have in common: Anything you can say in one of them, you can say in any of the others. And you can say anything and everything that can be said in any one of them. Not necessarily in the same number of words (and you might have to define a few new ones); not necessarily equally elegantly; but anything and everything.

    (Some readers may find the foregoing assertion as hard to agree with as Sue’s that bonobos have language. I suggest they test their intuitions by finding a counter-example: either a human language in which you can say this, but not that; or something you cannot say in any human language. Until someone comes up with such a counter-example, I will provisionally take it to be a true property of language — not human language, but language itself — that if you have it, you can say anything and everything that can be said [or gestured or written, as the modality need not, of course, be vocal], and if not, not.)

    Neither Kanzi nor her kin or kind can say everything (or anything faintly near everything). I accordingly conclude that they cannot say anything. They can do a lot — far more than anyone ever imagined nonhuman primates could do. And what they can do includes an astonishing amount of intelligent, purposive communication with humans, using some of the same components to communicate that humans use for language: They can communicate purposively by sending and receiving computer images as well as by responding to human spoken sounds. But the undeniable fact is that — no matter how much linguistic understanding we attribute to them — they cannot enter into this “conversation” we are having in this Forum, not even into a rudimentary approximation to it, whereas any speaking human being, speaking any (spoken or gestural or written) language, can; even a child.

    And the most likely reason for that is that bonobos cannot understand propositions as propositions (statements with a truth-value: true or false); and that what they do understand and express when we think they are understanding or making propositions is not what we think it is. The “narrative” we project on it is more like the sound-track of a silent movie — one generated by our own language-prepared brains, irresistibly “narratizing” (as Julian Jaynes dubbed it) every scene we see, but especially every communicative interaction with another mind (and sometimes even, frustrated, with malfunctioning machines). We are projecting propositionality where it is absent.

    (And this is not merely about “aboutness” in the sense Sue intends it — not just about the intended object or “referent” of attention, shared attention, pointing, gesturing, or miming; it is about making and meaning subject/predicate assertions with truth values. For that is what gives language its unbounded expressive power, allowing us to make any and every proposition. Nor does that have anything to do with “consciousness,” i.e., feelings, which bonobos, and of course most — probably all — animals have; nor with the “self/other” distinction, which many species can make, to varying degrees, in the practical, sensorimotor sense, but none but ourselves can make in the linguistic sense.)

    It is hard to understand why creatures as stunningly intelligent and capable as bonobos cannot acquire language. I’d say that that was an even more remarkable and puzzling fact, yet to be understood and explained, than the remarkable intellectual and communicative feats that they have indeed proved capable of mastering — but of course it is precisely what they can do that makes what they can’t do all the more perplexing: Why can’t they say anything and everything, given what they can demonstrably do, if it’s really language?

    Sue’s reply is: “cultural differences”; and with Teco she’s hoping to close the cultural gap. But with any human child, the gap is closed almost immediately, in infancy, once it acquires (any) natural language. (Some unnatural languages can be designed that defy the child’s language-learning capacities, but that’s another matter; even those artificial languages still have the full expressive power of any natural language.) So until Teco can join this conversation, I will assume that what is going on is a good deal of hopeful, irresistible propositional over-interpretation (by humans) of some remarkable intellectual and communicative capabilities and performance (by bonobos) — but not a conversation, not propositions, and hence not language.

    Harnad, S, (2010) From Sensorimotor Categories and Pantomime to Grounded Symbols and Propositions. In: Handbook of Language Evolution, Oxford University Press.

    ______. (2010) Symbol Grounding and the Origin of Language: From Show to Tell. In: Origins of Language. Cognitive Sciences Institute. Université du Québec à Montréal, June 2010.

    • I think you are correct in stating that “bonobos cannot understand propositions as propositions”, but at the same time you are missing the point, as revealed by your following claim that “the undeniable fact is that — no matter how much linguistic understanding we attribute to them — they cannot enter into this ‘conversation’ we are having in this Forum, not even into a rudimentary approximation to it, whereas any speaking human being, speaking any (spoken or gestural or written) language, can; even a child”.

      Infants come to construct propositional frames to put their communicative symbols into quite late in the day (see, for example, Lock, 1980), in the sense of predicating/entailing some property to an object or event other than their presupposed self. For example, it is difficult to claim that the early use of a ‘word’ such as ‘hot’ uttered when an infant is near a fire is propositional, that the infant is entailing the property of heat to a spectacle: ‘the fire is hot’. More likely, they are commenting on the fact that they now can identify that they ‘feel hot’. The propositional form, and the evidence that the infant is entailing the property to an object independent of itself, takes a while to construct.

      Once constructed, there is little evidence that infants deploy propositions as propositions. They are not, in fact, interested in “making and meaning subject/predicate assertions with truth values” for quite sometime. Language is for coordinating actions in a shared world, and propositional forms are useful for that, in that the subject of the proposition identifies a shared topic that can be practically oriented to. Being concerned with ‘truth values’ is more likely a cultural construction consolidated a few thousand years ago when ‘theory’ became a valued project. On this view the “‘conversation’ we are having in this Forum” is not relevant to deciding the status of the abilities bonobos, infants, or prehistoric humans, exhibit in their symbolically mediated conversational abilities. That language can be ‘about the world’ for the sake of argument or the pursuit of truth is a potential of language that has taken a long time to exploit, and required finding a way of life in which doing so is considered of value.

      Lock, A. (1980). The guided reinvention of language. London: Academic Press.


    Sue Savage-Rumbaugh proposes (I simplify) that language made us distinctively human. I agree wholeheartedly. Several fundamental errors lead me to disagree with nearly every specific. The importance of individual knowledge and individual competency, for example, characterizes Western not human culture and stems from the cultural assumption that each person knows best for him- or her-self (Handwerker 2009). More important, the proposal that language use transforms bonobos into something more human misses the point that humans interact with something close to human daily without appreciating our kinship. We don’t need language to make it so.

    This stems from what we might call ‘The Primate Fallacy’ — the failure to extend comparisons beyond primates. Savage-Rumbaugh thus does not engage the observations that (1) consciousness appears to have evolved as the component of ‘intelligence’ that improves the transfer of information from short-term to long-term memory, not as a facility related to symbols, reasoning, and planning; (2) intelligence plausibly originated some 50,000,000 years ago, (3) we find both both consciousness and intelligence widespread in birds and mammals — a goat I once shared with a friend constructed, modified, and used a tool in my presence, for example — and behavioral evidence suggests that consciousness extends also to fish and cephalopods, and (4) because these forms of consciousness encompass things like symbols, reasoning, and planning as well as distinctions between me-thee, past-present, and him-her — what she calls ‘human patterns of self-awareness and self-reflection’ — her characterization of distinctively human qualities exhibits fundamental flaws (see, e.g., Jerison 1973, Danchin et al 2004, Handwerker 2009). An afterthought — the artifactual markers that Savage-Rumbaugh cites as evidence of such consciousness evolved gradually over some 400,000 years (McBrearty and Brooks 2000).

    Roy D’Andrade (2002) points out that the intelligence of contemporary humans and the cultures they create owe much to the emergence of language. An ancestor who could talk about what he or she did, when, how, and why, could more readily share his or her knowledge. Sharing knowledge quickens the spread of important innovations and means that your ideas can build on those of others, even if those ideas or ways of doing things originated hundreds or thousands or hundreds of thousands of years earlier. The minds of people who store more information generate both more and more radically different new ways of thinking about the world and acting in it. This heightens personal awareness that things don’t work right and makes finding ways to effectively correct mistakes easier and faster. Significant advantages thus accrued to our ancestors, who, generation after generation, possessed minds that both received and stored increasing amounts of sensory information, processed it in ways that generated increasing numbers of innovations, and effectively shared them increasingly well. In short, language contributed significantly to the (Darwinian) selective advantages of cultural participation. Savage-Rumbaugh addresses an important problem and set of observations, but the framework she now uses to frame both may need refinement.

    Danchin, Etienne, Luc-Alain Giraldeau, Thomas J. Valone, and Richard H. Wagner. (2004) Public Information: From Nosy Neighbors to Cultural Evolution. Science 305:487-491.
    D’Andrade, Roy. (2002) Cultural Darwinism and Language. American Anthropologist 104:223-232.
    Handwerker, W Penn. (2009) The Origin of Cultures. Walnut Creek, CA: Left Coast Press.
    Jerison, H.J. (1973) Evolution of the Brain and Intelligence. New York: Academic Press.
    McBrearty, S. & A. Brooks. (2000) The Revolution That Wasn’t: a New Interpretation of the Origin of Modern Human Behavior. Journal of Human Evolution 39:453-563.

  • Conjoint, situated, expressive movements: developing our own human relations to our human surroundings

    John Shotter
    January 4th, 2011 at 11.15am • reply

    In my comments below, I do not want to evaluate Sue Savage-Rumbaugh’s claims as to their appropriateness or adequacy in relation to our current understandings as to what a human language is. What I want to do instead is celebrate the very special nature of her unique research and to explore the whole new realm of inquiry — to do with the expressive-relational nature of animals’ living movements — that it opens up. For we can assume that, as animals face the task of moving around in relation to their needs in different (and in sometimes unusual) environments, occasionally at least, they will meet circumstances that ‘call out’ uniquely new responses from them, responses not previously performed, responses that offer the beginnings of new forms of consciousness to, for, and within them.

    When I began studying academic psychology in 1959, everyone’s focus was on the already existing ‘hidden things’ supposed to be occurring within the single individual somewhere, ‘things’ that we could only relate to indirectly, in terms of the observable, theoretically predicted, outcomes of causal manipulations. For, in the days of Behaviourism — Cartesian, mechanistic behaviourism of the day — we were meant to start our inquiries by observing very general “colourless movements and mere receptor impulses as such” (Hull, 1943, p.25), and from these to build up to our more complex behaviours — where it was assumed, of course, that such “colourless movements” presented themselves to us as the very “colourless movements” they were, without any effort being needed on our part to pick them out as such.

    I begin in this way for two inter-related reasons: One is just to note the extent to which, over the last 15 years or so, I have found myself becoming more and more interested in the minute details of the everyday social lives of animals — all kinds of animals, insects even — and what this can teach us about our own relations to the others and othernesses in our surroundings. Not only has Sue Savage-Rumbaugh been a major contributor to this movement, away from very general theories expressed in abstractions towards the noticing and documenting of concrete social events in detail, but she has gone much further: In wondering if “different styles of caregiving” foster different kinds of consciousness, she has not been content merely to observe experiments of nature, but has ‘intervened’ extensively in rearranging those patterns. Central to her interventions has been her own responsiveness to ‘teaching moments’ occurring in her ‘up close and personal relations’ with Kanzi and her family — touching, pointing, distracting, and many other things, to draw and animal’s attention to something ‘meant’ by her, actually there, in the situation of their joint involvement, plus all the time talking, moving, human, intonated, expressive talk that, in its musicality (Trevarthan, 2002), works to shape the relations her and Kanzi.

    Clearly, in my own movements I have slowly become very differently oriented, very differently related as an academic psychologist to my surroundings, picking out and attending to ‘things’ in them in very different ways than before — and this is my second reason for beginning with my own change of focus over the years.

    “Knowledge is in the end based on acknowledgement,” remarks Wittgenstein (1969, no.378). Before being able to gain what we call ‘knowledge’ about the ‘things’ around us, we need to be able to recognize them and move around in relation to them in our everyday practices as the ‘things’ they are — and such practical recognitions cannot be taught by teaching propositions or by offering propositions. The ability explicitly to say, “This is an X,” requires our already knowing implicitly what X-ness is. This what-ness of ‘things’, the capacity to orient to one’s surroundings in something of a human rather than a bonobo fashion, is what Sue Savage-Rumbaugh sees as being taught in the different care-giving patterns of her concern. So, while Kanzi and her family are bonobos, she notes that “they explore art, they explore music, they explore creative linguistic negotiation, they have an autobiographical past and they think about the future. They don’t do all these things with human-like proficiency at this point, but they attempt them if given opportunity. Apes not so reared do not attempt to do these things.”

    This, I think, is amazing. The attentive patterns of situated care-giving studied by Sue Savage-Rumbaugh seem, then, to be concerned, not with the acquisition of knowledge per se, but with the acquisition of those attitudes, ways of relating, or orientations to do with seeking to acquire it as a spontaneously felt, embodied need. If this is so, then this is an important achievement indeed. For those of us worried about how to relate to disaffected, disoriented, delinquent youngsters might find Sue’s ways of re-orienting bonobo’s towards more human ways of acting relevant to our concerns also.

    Hull, C. L. (1943) Principles of Behavior: An Introduction to Behavior Theory. New York: Appleton-Century-Crofts.
    Trevarthan, C. (2002) Origins of musical identity: evidence from infancy for musical social awareness. In, MacDonald, R., David J. Hargreaves, D. J. and Dorothy Miell, D. (Eds.) Musical Identities. Oxford: Oxford University Press (pp.21-38).
    Wittgenstein, L. (1969) On Certainty ed. G.E.M.Anscombe and G.H.von Wright, trans by Denis Paul and G.E.M.Anscombe. Oxford: Blackwell.

  • Response to Savage-Rumbaugh essay “Human language – Human consciousness”

    Stones and bones are what fossilize in the archaeological record and so it is the story they tell that absorbs the lion’s share our attention as we try to understand the evolution of the human mind. Undoubtedly, tool making and changes in body and brain size played an important role in making us human. But Sue Savage-Rumbaugh rightly draws attention to a neglected truth of evolution: selection on the adult phenotype is actually selection on a particular ontogenetic pathway leading to a certain adult phenotype. Our real challenge is in understanding how, over the course of hominin evolution, an ontogenetic pathway that historically led to a hairy, quadrapedal, fruit-eating primate with 30 or so vocal calls, altered to become one whose end point is a naked, bipedal, meat-eating primate with a syntactically complex linguistic communication system. Savage-Rumbaugh points to a critical ontogenetic difference between (nonhuman) apes and humans — ape infants cling to their mothers; human infants are carried by their mothers. This physical difference produces a critical relational difference: ape infants’ earliest development is characterized by being an extension of their mothers, while human infants’ earliest development is characterized by being an object tended to by their mothers.

    Thinking in terms of shifting ontogenetic pathways forces us to look at the hominin evolutionary record differently. The most crucial event in hominin evolution may have occurred 1.2 million years before present (mybp) and not 2.6 mybp. It is about 1.2 mybp that genetic evidence indicates that our ancestors became hairless. By at least this time then, mothers had to relate to their infants in a radically new way. By contrast, the first evidence of stone tools emerges around 2.6 mybp. It is at this time that adults began to create tool technologies that changed how they gathered resources. Was adult tool-making a prerequisite condition for the later change in ontogeny? Possibly, but apes make tools and as best as we can tell this has not had any profound effect on their development. Furthermore, the first hominin stone tools are not associated with any significant change in brains, bodies or even diet. Stone tool making was probably just one of many currents whose confluence ultimately produced the right set of conditions for pushing primate development toward a more “human” trajectory. Hairlessness, by contrast, probably marks a more significant inflection point in the evolution of hominin ontogeny.

    A baby can’t cling to a hairless primate forcing mothers to monitor helpless infants in new ways. Rhythmically exchanged vocal calls (found in foraging baboons) may have become adapted for use as an “at a distance” monitoring system. These exchanges may have set the evolutionary foundation for the proto-conversational turn taking so prevalent in human mother-infant interactions. These interactions, in turn, are fundamental to the development of both the deep emotional attachments and the sophisticated theory of mind that characterized human social life. The fact that encultured apes show social and cognitive capacities not seen in wild apes does not tells us that apes can become human if given the right upbringing. Instead, it shows us the potential of the primate raw material when the ontogenetic path is shifted. It was this raw material that natural selection began working on a million years ago to shape our unique social/cognitive capacities.

  • Tom Givón

    T. Givón

    Sue Savage-Rumbaugh’s work on the cognitive and communicative capacities of apes is widely known and greatly admired. Her current essay is far ranging and thought provoking. My comments are meant to amplify on some of the themes she raises.

    1. Parenting styles: The human parenting style described in the essay is typical of Western, educated, urban middle class. The parenting style of hunting-and-gathering humans would have been a more appropriate standard of comparison. Here Sarah Hrdy’s (2009) work on allo-parenting and Frank Marlowe’s (2005, 2010) and Robert Kelly’s (2007) work of the territorial ecology of human foragers would have been more relevant. The home range of human foragers is vastly expanded relative to chimps and bonobos, indeed comparable to that of social carnivores (lions, hyenas, dogs, wolves). Such territorial expansion gave rise to the Central Place Provisioning (CPP) residence pattern of human foragers, whereby gathered-and-hunted food is brought back daily to home base and shared, and where children that are too young to keep up with foraging adults but too heavy to carry — toddlers in the 2.5-5.5 yr. range — are left at home base in the care of allo-parents. Hrdy (2009) views this as the crucial ingredient, for both mothers and children, in expanding the capacity of Mind Reading (ToM), a central ingredient in the rise of human communication (Dunbar 1998; Tomasello et al. 2005; Givón 2005).

    2. Time traveling: While the capacity to view oneself consciously in multiple time-frames (Studdendorf & Corballis 2010) is an important part of the human mental tool-kit, the crucial adaptive impetus for declarative, displaced-reference human communication was probably much more mundane: the expanded foraging range and Central Place Provisioning system of human hunters-gatherers on the Savannah. This territorial pattern is documented back to, at least, Homo habilis and Homo erectus (Klein 1999). The long-distance ranging of both hunters and gatherers, and the monthly-to-bi-monthly moving of home base to a new foraging range, created an informational imbalance between sub-groups in the social unit, whereby crucial adaptive information (food, shelter, hostile encroachment, new territory) about there-and-then referents is available only to the few, but must be shared with the many to insure consensual decision making and coordinated action. This was, I suspect, the main adaptive impetus for the rise of declarative communication about displaced referents, a system so characteristic of human language and so unlike pre-human communication (Bickerton 2009; Givón 2009).

    3. What bonobos can and cannot do: Sue Savage-Rumbaugh makes an important evolutionary point — that the cognitive and communicative capacities of apes vastly outstrip what they actually display in their natural habitat. But claiming that the kind of language Kanzi et al. have acquired “…manifests all major components…” of human language is a considerable stretch. Language-savvy Bonobos communicate largely in one-word lexical pidgin, characteristic of one-year-old humans (Bloom 1973; Scollon 1976; Givón 2009). Grammar is not a significant component of their communication (Givón and Savage-Rumbaugh 2009*), nor is the abstract vocabulary that is a key pre-requisite to the genesis of grammar. And while they can answer questions with technically-declarative responses, their spontaneous communication with humans remains overwhelmingly manipulative. Likewise, while they can remember the past and imagine the future, their communication remains, overwhelmingly, about here-and-now, you-and-I, this-and-that visible.

    4. Nature vs. nurture: It is unnecessary and perhaps unwise to counter extreme Cartesian-Chomskian nativism with equally-extreme empiricism. A claim such as “…therefore the apparent differences between humans and apes are not biologically fixed, but they are biologically and culturally instantiated…” seems unnecessarily extreme. Admitting the role of evolved biological capacities detracts nothing from the great importance of contextual mediation and behavioral flexibility. Bio-evolution imposes clear limits on the range of intra-specific variation and contextual flexibility — without abolishing flexibility and variation (Mayr 1969, 1976, 1982). And 7 million years of separation is surely long enough for some DNA-coded bio-neurological evolution to have taken place. And lastly, the proverbial 98% shared DNA must be taken in context of 4 billion year of bio-evolution, whereby the overwhelming bulk of primate genome codes the evolutionary accretion of the 4-billion years prior to the emergence of the primate line.

    T. Givón
    Institute of Cognitive and Decision Sciences
    University of Oregon

    Bickerton, D. (2009) Adam’s Tongue, NY: Hill and Wang
    Bloom, L. (1973) One Word at a Time: The Use of Single-Word Utterances Before Syntax, The Hague: Mouton
    Dunbar, R. (1998) “Theory of mind and the evolution of language”, in J.R. Hurfod. M. Studdard Kennedy and C. Knight (eds) Approaches to the Evolution of Language, Cambridge: Cambridge University Press
    Givón, T. (2005) Context as Other Minds: The Pragmatics of Sociality, Cognition and Communication, Amsterdam: J. Benjamins
    Givón, T. (2009) The Genesis of Syntactic Complexity: Diachrony, Ontogeny, Neuro-Cognition, Evolution, Amsterdam: J. Benjamins
    Givón, T. and S. Savage-Rumbaugh* (2009) “Teaching grammar to apes”, in J. Guo, E. Lieven, N. Budwis, K. Nakamura, & S. Ozkaliskan (eds) Cross-linguistic Approaches to the Psychology of Language: Research in the Tradition of Dan Isaac Slobin, NY/London: Psychology Press
    Hrdy, S. (2009) Mothers and Others: The Evolutionary Origins of Human Understanding, Cambridge, MA: Harvard University Press
    Kelly, R.L.. (2007) The Foraging Spectrum: Diversity in Hunter-Gatherer Lifeways, NY: Percheron
    Klein, R.G. (1999) The Human Career: Human Biological and Cultural Origins, 2nd edition, Chicago: University of Chicago Press
    Marlowe, F. (2005) “Hunter-Gatherers and Human Evolution”, Evolutionary Anthropology, 14
    Marlowe, F. (2010) The Hadza: Hunter-Gatherers of Tanzania, Berkeley: UC Press
    Mayr, E. (1969) Populations, Species and Evolution, Cambridge, MA: Harvard University Press
    Mayr, E. (1976) Evolution and the Diversity of Life, Cambridge, MA: Harvard University Press
    Mayr, E. (1982) The Growth of Biological Thought, Cambridge, MA: Harvard University Press
    Scollon, R. (1976) Conversations with a One Year Old Child, Honolulu: University of Hawaii Press
    Studdendorf , T. and M.C. Corballis (2010) “Behavioral evidence for mental time traveling in non-human animals”, Behavioral Brain Research, 215 (2)
    Tomasello, M.M., J. Carpenter, J. Call, T. Behne and H. Moll (2005) “Understanding and sharing intentions: The origins of cultural cognition”, Brain and Behavior Sciences, 28

    [* Note: Co-authorship of the marked chapter cited has been contested -- Eds.]

  • “Etymologically, of course, the term “consciousness” is a knowledge word. This is evidenced by the Latin form, -sci-, in the middle of the word. But what are we to make of the prefix con- that precedes it? Look at the usage in Roman Law, and the answer will be easy enough. Two or more agents who act jointly — having formed a common intention, framed a shared plan, and concerted their actions — are as a result conscientes. They act as they do knowing one another’s plans: they are jointly knowing.” (Toulmin, 1982, p.64).


    Sue makes a number of metaphysically and empirically controversial points here — for example, that bonobos can partake in “human language as you and I speak it”; and that consciousness, as you and I have it, “burst upon the … scene between 40,000 and 90,000 years ago”. But in going on to debate these points (which I will do in commenting on other commentaries rather than here), we need to recognise that the most important controversial point she makes is both an empirical and ontological one.

    Work in human-infant interaction provides the view that human infants develop as human beings because they are treated by their caretakers as if they already were human beings. Infants are not presented with a booming, buzzing lived-in world that they have to bring their considerable intellectual powers to in order to make sense of the significance of events and objects in the world they have entered. Rather, their perceiving is structured as their attention is drawn to the local significance of events and objects as they are transacted to them in their real-time participation in shared worlds of socio-cultural interaction. In the developmental context, developing human consciousness is distributed (Toulmin, above) unequally to begin with. Adults, in their conduct, create a zoped [zone of proximal development] (Vygotsky, 1966) that imbues the sense infants can make of their jointly experienced world; infants develop so as to join in.

    Of course, infants bring an evolutionary history to this task of sense-making, and the general zeitgeist of contemporary psychology has made elucidating these individually-based abilities its focus, sidestepping an appreciation of the crucial role of the shared, relational consciousness that provides the grounds on which these abilities develop. You can’t, ethically, do an experiment that would shake the foundational importance of these paradigmatic blinkers. This, though, is what Sue has achieved in her work: an ontological shift. If you treat a bonobo that naturally develops a particular non-human way of life when raised by its conspecifics as if it were human — share with it a world imbued with human significance in a way that is responsive, moment-by-moment, to the joint reality human and animal are engaged in and attending to — then an otherwise unbelievable transformation occurs.

    This achievement puts us in a new ontological ball-park. As Sue points out, the notion of “epigenetic markers” may well provide a framework for understanding the biological bases by which human evolution was made possible, and mirror-neuron systems may well play an important role as well. But the real question becomes: how, in evolutionary time, were the zopeds that provide the epigenetic landscape enabling the tuning of possible into actual human consciousness constructed, sui generis, between individuals at the same level of cognitive ability? Zopeds can’t come out of nowhere to co-opt and transform otherwise mute biology. I have outlined some first thoughts on this elsewhere (Lock, 2000).

    Lock, A. (2000). Phylogenetic time and symbol creation: where do ZOPEDS come from? Culture and Psychology 6: 105-129
    Toulmin, S. (1982). The genealogy of “consciousness”. In P. F. Secord (ed.) Explaining human behaviour (pp. 53-70). Sage, Beverly Hills
    Vygotsky, L.S. (1966) Development of higher mental functions. In A.N. Leontyev, A.R. Luria and A. Smirnov (eds.) Psychological research in the USSR. Moscow: Progress Publishers.

  • I applaud Sue Savage-Rumbaugh’s attempt to tackle this fascinating and very important area. The idea of the power of cognition, culture and language in shaping our perceptions and actions is often overlooked when constructing evolutionary scenarios to explain why humans do what they do. The human niche is constantly modified via linguistic and other cognitive elements alongside the other social and ecological facets of being and becoming human. I am convinced that niche construction, especially social niche construction is an extremely important influence in human evolution (Fuentes 2009, Laland et al. 2000).

    Sue Savage-Rumbaugh’s work with bonobos is remarkable and gives her a particular insight into the minds and behavior of some members of the genus Pan that most of us will never have. However, I must agree with a few of the other commentaries and note that Dr. Savage-Rumbaugh has misconstrued some facets of human evolution and possibly gone further afield than warranted in her ideas about both the developmental plasticity in Pan and the relative role that neurological and other physiological changes have played in human evolution.

    One of the first elements of weakness in this scenario is the over reliance on the mother-infant dyad as the primary, and more or less exclusive dyad of interest from the infant’s perspective. This is the case for apes, but not for humans. Recent overviews (Gettler 2010, Hrdy 2009) demonstrate that human infants are extremely interactive with multiple caretakers and others from very early on, unlike most other primates (and mammals). Infants are part of the human social niche; the model of mother-infant pair as an isolate does not work well for trying to fully understand human development. One should also note that human infants do have agency beyond crying, and use a diverse array or visual and tactile cues to communicate.

    I agree, and find it fascinating, that bonobos reared by humans do co-opt a variety of human signals and postures, even some behaviors, but this does not indicate that the differences between the genus Homo and the genus Pan is largely developmental. It is not. The similarities in structure and physiology between our genera can result in a range of overlap when one species is raised in the cultural context of the other. This can also occur with dogs, but the lack of opposable thumbs and less developed neocortexes limits the amount of co-option they can undertake. Because of the phylogenetic similarities between bonobos and humans we should expected more convergences than between humans and dogs. But there is no justification for the assumption that raising bonobos like humans will result in bonobos who are more human. It results in bonobos who use a range of behavior that is 1) possible for the bonobo to produce and control, and 2) effective for communicating with humans. This says more about the high social intelligence and behavioral malleability of bonobos than it does about cross-species transfiguration via development.

    The argument that “Human Consciousness seems to burst upon the evolutionary scene in something of an explosion between 40,000 and 90,000 years ago” is not correct. As noted in the commentary by Givon, this is just too recent and too simple a view of human consciousness. There is evidence of “art” at earlier times than 90,000 years plus evidence of sea crossings, complex hunting, long-distance resource exploitation, fire use, shelter construction, etc… before this time period as well. I do not doubt that in the last 100,000 years or so humans have become progressively more complex cognitively, for a variety of reasons, including the language-development-cognition feedback circuit that Savage-Rumbaugh focuses on. But it is not so simple as she paints here. She wonders how we could have “evolved so quickly from an apish mentality”… but we did not… it took millions of years. Also, it is a certainty that modern day members of the genus Pan have also evolved (changed) from the shared ancestor 7 million years ago (or so) so an “apish” mentality based on living apes is also a modern state, not necessarily reflective of our last common ancestor.

    Finally, I am a big fan of epigenetic approaches and completely agree that is very important to consider them when we think about developmental and evolutionary change. However, there is no evidence that Kanzi learning aspects of human language or being interested in painting actually causes heritable changes to the neurophysiology of the brain. The argument that some currently unknown epigenetic markers are activated by cross-fostering a member of the genus Pan is on shaky ground. Maybe, this is a form of behavioral inheritance or possibly in Kanzi’s offpsring’s case, a symbolic inheritance (see Jablonka and Lamb 2005).

    Despite my criticism, I am a fan of this endeavor. We need to move beyond standard approaches when modeling the evolution of complex cognition, language and the human niche. Courageous and engaging ideas and research like that conducted by Dr Sue Savage-Rumbaugh are a terrific contribution to the discourse.

    Fuentes, A. (2009) Re-situating Anthropological approaches to the evolution of human behavior. Anthropology Today 25(3):12-17
    Gettler LT (2010) Direct male care and hominin evolution: Why male-child interaction is more than a nice social idea. Am Anthropol 112:7–21.
    Hrdy SB (2009) Mothers and Others: The Evolutionary Origins of Mutual Understanding (Belknap Press of Harvard University, Cambridge).
    Jablonka, E and Lamb, M. (2005) Evolution in four dimensions: Genetic, epigenetic, behavioral, and symbolic variation in the history of life. Cambridge, MA: MIT Press.
    Laland, K., Odling-Smee, J. and Feldman, M. W. (2000) Niche construction, biological evolution and cultural change. Behavioral and Brain Sciences 23: 131-175.

  • Aristotle famously said that humans were the only animals to have language, or, to be more precise, to have “logos” (Aristotle 1957). When he developed this idea in his Politics, Aristotle was not the first Greek to link theoretical ‘humanness’ and the faculty of language (and one could find equivalent claims in non-Western parts of the world), but the way he framed the debate was extremely powerful and is widely (and often tacitly) accepted today, even in the sciences. “Logos” is a very polysemic word in Greek, and I would need a lot of time to explain what is said on this matter by Aristotle (and not the dominant tradition). But let us be bluntly brief. “Logos,” in this text, refers to discursive rationality, or the intellectual ability to consider and distinguish basic, bipolar concepts (such as just and unjust, good and bad) through speech. Other animals, Aristotle says, have no “perception” of such things, even though they feel “pain and pleasure,” as humans do, and are apt to “signal” it to “each other” by using their “voice.” That is to say, in humans, mental “perception” is influenced by language (in general). “Logos,” then, refers not only to language as it is realized through discourse, it also implies that the human mind, in its logic and ordering of the real, depends on having language. Finally, since Aristotle is both a philosopher and a naturalist, one could go one step further and advance that, in this hypothesis, any non-human animal having at least some human language would begin to perceive concepts, in order to consequently display some moral behavior, to re-arrange its mental life through internal and external dialog (dia-logos), i.e. to become human.

    At this point, it should be obvious that several claims made by Sue Savage-Rumbaugh in her present essay conform to this philosophical perspective (effect of language on the mental life of animals, subjectivity, logic, morality,…). I would even say that, to a certain extent, and along many other goals, Savage-Rumbaugh’s research might be seen as a new, ambitious and experimental way of (finally!) testing an enduring hypothesis on animal humanness, a hypothesis that — like it or not — is still absolutely central in contemporary science and politics. While Sue’s bold manner of thinking goes far beyond disciplinary norms, her experiments, results, and theories, have to be more widely considered by those ‘humanists’ who seek to approach language and thought outside their human hypostasis. On the other hand, many scientists who simply (too simply) oppose Sue’s work might be well advised to re-examine their own implicit philosophy.

    From there, I could submit numerous considerations or comments, and express some possible disagreements (regarding the conceptual armature of the demonstration in particular). But given the format and the scope of this forum, I will just ask the author for one clarification. Once we admit that, in eloquent primates, ‘consciousness’ (if we have to to stick to this word) and language are closely intertwined ( I would agree with this), what is the exact nature of this link? It sometimes seems that Savage-Rumbaugh envisions language as something creating the self, morality, etc. Or could it be (as I sometimes understand it) that language rather expands, shapes, orders and/or channels already existing mental phenomena? In recent years, and even though this topic is also highly controversial, many observational and experimental studies tend to suggest that, for instance, corvids could spontaneously have a mental representation of their future states (Raby 2007); that some dolphins are not only self-aware of their own image but also bring objects in order to play with their reflection in the mirror (Marten 1995); that elephants categorize death and display a ritual behavior toward skulls and bones of conspecifics (McComb 2006), etc. Such observations and data might indicate that, in very different animals, elements of ‘human-like consciousness’ are indeed present, even if they do not have human language, and even though they are not primates.

    Aristotle (1957) Politica, Oxford, Clarendon Press.
    Marten, Psarakos (1995), “Evidence of self-awareness in the bottlenose dolphin (Tursiops truncatus)” in Self-awareness in Animals and Humans: Developmental Perspectives, New York: Cambridge University Press, 1995.
    McComb, Baker, Moss (2006), “African elephants show high levels of interest in the skulls and ivory of their own species”, Biology Letters, 2-1.
    Raby, Alexis, Dickinson, Clayton (2007), “Planning for the future by Western Scrub-Jays,” Nature 445.

  • I am in Sue Savage-Rumbaugh’s debt for her explanation of a bonobo-human difference that I previously hadn’t appreciated. Different early-developmental pathways obviously lead to significant cognitive differences at subsequent stages. But could this particular difference bear the weight Sue places on it? I doubt it. And not just for the reasons that previous commentators have already well expressed.

    I’d rather deal with a basic assumption that underlies, not just Sue’s article, but a great deal of current thinking on how we became human. It’s the assumption that humans are just like modern apes but with one or two crucial add-ons. These add-ons can be anything from Sue’s mother-infant separation to Chomsky’s (2010) recursive merge — the assumption’s
    shared across a wide spectrum of theories — but the underlying notion remains the same.

    It was undoubtedly good to get rid of the idea that a vast, God-given gulf separates humans from the rest of nature. But as every Frenchman knows, the good is the enemy of the best. The best, in this case, being an understanding of how biological evolution actually works. And it does not work by taking some existing species and adding a few gimmicks and/or tuning up some gimmicks the species already has.

    Bottom line, humans are still being treated as something special. People keep looking at apes and other primate species to find “precursors” of human behaviors. Nobody looks for precursors of web-spinning in spiders, or bark-penetration in woodpeckers, or dam-building in beavers, or echolocation in bats. Everyone tacitly accepts that these behaviors — each as unique as language, please note — developed in isolation, as a result of a particular species exploiting a particular ecological niche. By now we should know enough evo-devo to realize the pluripotentiality of genes, the multiplicity of ways in which the “same” gene can express itself under different environmental conditions.

    Other species, no matter how genetically close, are, when all’s said and done, other species. When I met Kanzi, my immediate impression was, “Wow! Watch out for this guy!” I wasn’t faced with an intelligence that was kind of like mine but not quite up to snuff. I was faced with a formidable intelligence of a quite different order, with different skills, different functions, different goals. Luckily for me we were forced to meet on my turf. If we’d met on his turf, I know who’d have come off best, and it wouldn’t have been me.

    What would you do if you were placed in captivity by members of another species, aliens, say, who wanted to study you? You’d try to get along with them as best you could. You’d use your intelligence to figure out what they wanted. They want X? Okay, I’ll give them X! But I’m still me, and as far as I’m concerned, X is not really a part of that at all.

    Remember the old joke about how many psychiatrists it takes to change a light-bulb? One, but the light-bulb has to want to change. In order to develop any novel, unique trait, a species has to want to change, or I should say has to desperately need to change. It has to be faced with some ecological problem that can be solved only by developing some novel, unique trait — some X. The alternative is to go extinct.

    I’ve already tried to show how this could have happened with humans (Bickerton 2009). I could be quite wrong in the
    details, but not, I think, in the basic assumption. Something happened to us on the way from the Last Common Ancestor that didn’t happen to the other apes, that made us behave quite differently from them. We built our own X, which of course, as Sue agrees, is language. It was sheer luck — good or bad, you can’t tell yet — that this particular X ended up by putting us in Paris and New York while they remained in their dwindling tropical forests.

    Bickerton, D. (2009) Adam’s Tongue: How Humans Made Language, How Language Made Humans. New York, Hill & Wang.
    Chomsky, N. (2010) “Some simple evo-devo theses: how true might they be for language?” in The Evolution of Human Language: Biolinguistic Perspectives, ed. Larson et al. Cambridge, CUP.

  • Sue Savage-Rumbaugh has made a bold two part conjecture — that 1) the earlier and more intimately ape infants are exposed to human care-giving patterns and linguistic environments the more likely it is that they will acquire and demonstrate components of human language, and 2) that this differential rearing will result in epigenetic changes observable in the behavior of their offspring. Lacking the biological expertise to comment on the likelihood or mechanisms of epigenetic inheritance I will restrict myself to a discussion of the first conjecture.

    SSR’s work, which some might characterize as experimental anthropology, constitutes a sustained arc of research from Lana, the first ape to use a lexigram keyboard, through Sherman and Austin, Kanzi and Panbanisha, and, finally, to Teco, Kanzi’s six month old son. Each of these experimental groups in succession received earlier and more intense exposure to human language and culture and as the exposure increased so did their demonstrated facility with language.

    Lana was two and a half years old when her training began and she was purposely not exposed to spoken language or extensive personal interaction with her trainers. In fact Duane Rumbaugh had hoped that simple interaction with the computer itself might be sufficient to produce an ‘analog’ of human language. When SSR began working with Lana she realized that although Lana was able to string together lexigrams to make requests or answer questions, her understanding of the symbols was not coextensive with a human’s understanding. Her subsequent work with Sherman and Austin was an attempt to better ‘ground’ these lexigrams and to go beyond conditioned stimuli to symbols. After arduous training the chimps did learn to treat lexigrams symbolically and to use them to make requests and interact with one another. Yet, like Lana, neither chimp ever came to comprehend spoken English. The story of Kanzi is well known, how he acquired his language facility spontaneously through observing his mother’s training sessions. Exposing his sister Panbanisha to a similarly rich language environment was met with equal success.

    The birth of Teco has raised the experimental stakes even higher. Whether Teco’s inability to cling was the result of epigenetic changes or natural accident, it has provided a serendipitous occasion for the most dramatic test of rearing-effects yet. The logical prediction of SSR’s first conjecture is that Teco, who is receiving maximum linguistic interaction with a human mother from the earliest possible time, should show the greatest gains of all in linguistic capabilities. I applaud Sue for providing us with a testable/refutable hypothesis and for her daring experimental creativity. I have had the opportunity to visit with Teco and I was amazed at the precocious hand and eye coordination he is already demonstrating and his keen interest in the lexigram board, the words spoken to him, and even the games on his iPad. Time will tell whether SSR’s conjectures prove to be right or wrong, but as James Joyce commented about persons of genius, ” they make no mistakes … their errors are the portals of discovery”.

  • The evolution of language: Hardwired?
    Edward A. Wasserman
    Department of Psychology & Delta Center
    University of Iowa

    “From the day we’re born, we really do know where we’re headed. The whole of humankind is gifted with an innate sense of direction. It’s spinal. Inborn.” So proclaims one of the video advertisements from MSNBC’s recent “Lean Forward” campaign, entitled “Hardwired.” The “Mad Men” who crafted this narrative may have hoped to promote a progressive sensibility, but they seem not to understand the true nature of human development.

    We may each have our own penchants and peculiarities, but how we develop as individuals in an incomprehensibly complex world is one of life’s great mysteries. We both change and are changed by our environment. Such an interactive developmental process is anathema to the notion of destiny that is so strongly implied by the opening lines of this promotional video.

    Animals too face similar intricacies of developing in the natural world. As does each of us, each animal comes to exhibit species-typical responses as well as idiosyncratic actions. A full understanding of behavioral development demands that we consider both humans and animals from a common evolutionary perspective.

    Sue Savage-Rumbaugh has been at the forefront of efforts to discover the potential of nonhuman apes for acquiring language behavior. Forget destiny and hardwiring. Savage-Rumbaugh has fostered bonobos with languaged humans to see if particular developmental experiences can produce bonobos with demonstrable language skills.

    Is this notion pure fantasy? Kanzi’s reported language behaviors say “no” (Savage-Rumbaugh, 2009). But, Kanzi’s communication skills should not come as a complete surprise. The 18th century French physician and philosopher, Julien Offray de La Mettrie, was mightily impressed by the success of the Swiss physician Johann Konrad Ammann in teaching nonverbal deaf persons to speak by deploying what today we call behavior modification techniques. Might this or other tutelage also succeed with animals? La Mettrie said “yes,” but only if the proper animal were to be chosen. He believed that the ape would be the best selection.

    In L’Homme machine (1747/1996), La Mettrie explained this choice: “such is the likeness of the structure and functions of the ape to ours that I have very little doubt that if this animal were properly trained he might at last be taught to pronounce, and consequently to know, a language. Then he would no longer be a wild man, nor a defective man, but he would be a perfect man, a little gentleman, with as much matter or muscle as we have, for thinking and profiting by his education (p. 103).”

    If Kanzi’s competences have reached their limits (and they may well not have done so), then La Mettrie may have been wrong about the potential of apes for producing speech sounds, but he may have been right about the potential of apes for understanding human speech. And, as engaging as Kanzi can be, I can personally attest to his being far from a “little gentleman.”

    One line from the MSNBC video does ring true: “We were built to evolve.” Humans did evolve. And, we did develop language. Who says that other animals are incapable of doing so if they were given sufficient time and proper experience? The notions of destiny and hardwiring are passé in light of our growing understanding of development and evolution. We need to view language from a fresh perspective (see McMurray & Wasserman, 2009) and to remain open to its emergence in other animals. After all, when he was born, could Kanzi ever have imagined where he was headed or that he would acquire a symbolic language? For that matter, could we?

    La Mettrie, J. O. D. (1747/1996). Machine man and other writings (translated and edited by A. Thompson). Cambridge: Cambridge University Press.
    McMurray, B., & Wasserman, E. (2009). Variability in languages, variability in learning? Behavioral and Brain Sciences, 32, 459-460.
    Savage-Rumbaugh, S. (2009). Empirical Kanzi. Skeptic, 15, 25-33.

  • Just how important is experience?

    Stan A. Kuczaj II

    Department of Psychology & Marine Mammal Behavior and Cognition Laboratory

    University of Southern Mississippi

    Sue Savage-Rumbaugh and her colleagues have demonstrated that chimpanzees can be quite clever and that some of them can engage interested humans in communicative exchanges. I believe these findings are unequivocal and commend Sue for her pugnacious efforts to unravel the mysteries of the chimpanzee mind. However, the interpretation of these findings has always been controversial. Some scholars argue that the cognitive sophistication of chimpanzees blurs the distinction between chimpanzees and humans, while others suggest that the newfound abilities of chimpanzees clearly fail to move chimpanzees across the boundary that separates human and chimpanzees. Most, if not all, of the commentaries on Sue’s article in this forum will likely fall into one of these two camps.

    Much of this debate focuses on whether or not chimpanzees are capable of learning something akin to human language. Given sufficient encouragement and human enculturation, at least some chimpanzees have shown much more success than many scholars and laymen would have believed possible fifty years ago. But is what they have learned a human language? The answer to this important question depends on how one defines human language. I know that Sue is often frustrated by what she views as the moving target in terms of the accepted criteria for human-language-like performance by her chimpanzees, but it is nonetheless the case that many scholars doubt that Kanzi (or any other chimpanzee) has in fact acquired a human-language-like system. He has learned a complex communication system, but is his performance, remarkable as it is, reflective of an abstract productive rule-governed system that allows it users to communicate about past and future events, hypothetical events, or even events that will never occur? Sue and many others believe the answer to this question is yes, but there are numerous researchers and theorists who would answer no.

    For the moment, let us assume that Kanzi and his brethren have not acquired something akin to human language. Is it the case that changes in maternal styles or other aspects of early experience could alter this outcome? Put another way, is it possible that certain forms of social interactions during an animal’s early life could result in radical changes in behavior and even alter a species’ predispositions? Sue suggests that “apparent differences between humans and apes are not biologically fixed, but (instead) biologically and culturally instantiated”. Of course, early experience is important. For example, children do not learn English if they never hear the language. However, the extent to which early experience can alter predispositions depends on many factors, including individual temperament and the extent to which predispositions are hardwired. Sue and her colleagues have already shown that certain forms of early experience yield higher levels of communication in apes than would otherwise be the case. But are the apes more like humans than like bonobos as a result? Or are they apes that have developed abilities to communicate with humans as a result of their enculturation? Attempts to raise wolf pups as if they were domesticated dog pups fail to produce domesticated wolves despite the fact that dogs and wolves are quite similar genetically. So although early experience can certainly affect developmental outcomes, there are limits. It will be interesting to see what these limits are in the case of the enculturated apes that Sue studies.

  • Paul J. Thibault

    What We Think We Know about Language, and What We can Learn about it from Kanzi and Family

    - Paul J. Thibault,
    - Faculty of Education, University of Hong Kong, Hong Kong SAR, P. R. China; University of Agder, Kristiansand, Norway
    Many commentators on Sue’s work over the years have started from the sceptical view that whatever Kanzi is doing and however clever and remarkable he obviously is, he is not using ‘language’ (see Savage-Rumbaugh et al 1998; Segardahl et al 2005 for discussion). In this vein, Steven Harnard, in his contribution to the current discussion, tells us that Kanzi cannot understand and express propositions (statements with truth-value). This claim seems to suggest a priori that the possibility space of natural languages just is the class of statements with truth-value (propositions). There is no consideration or mention of other possible distributions of non-propositional utterances with which propositions paradigmatically contrast in the overall possibility space of a given language. This conclusion is at variance with the fact that many utterances in natural languages such as English, Potunghua, etc. are clearly not propositions in this sense yet most people have no difficulty accepting them as utterances of this or that language. Where in this view do utterances featuring imperative or oblative mood figure? These mood categories do not realize statements to which truth-values can be assigned, yet most people would have no difficulty construing them as meaningful, linguistic utterances. To deny imperative utterances such as Give me the ball! linguistic status because they are not statements with truth-value is to drastically shift the goal posts even for the vast majority of human speakers of English.

    What Sue’s work invites us to do, in my view, is to radically rethink the nature of language itself and what (we think) we know about it. Kanzi and family, I contend, have a thing or two to teach us about this. Rather than assuming that the sine qua non of language is the communication of propositional thoughts, we can question whether ‘communication’ in this sense is indeed the primary function of language. Language events (e.g. speech, writing, sign, lexigrams) are, following Clark (1997) and Hutchins (1995), external resources or affordances that constrain and affect ways of acting and thinking. In their interactions with humans, Kanzi, Panbanisha and other bonobos at the IPLS are participants in distributed forms of thinking and acting that have the capacity to transform the local ecology. This occurs in a culturally organized world in which dialogically coordinated forms of action and experience are possible, as is also the case for human infants. Such a world depends on much more than ‘language’ per se. It depends on functionally organized modes of action with which language behaviour can be integrated, the biophysical dynamics of interactive behaviour, including vocalizing (bonobo and human), the physical properties of the lexigram board, environmental affordances, and so on. Instead of focusing on the telling of propositions in linguistic form, we can look at how bonobos like Kanzi and Panbanisha jointly manage their dialogically coordinated activity with humans. This requires a very different kind of empirical research programme than one based on the search for abstract propositions. Sue has been a pioneer in this endeavour over many years.

    The propositional view excludes the many aspects of languaging behaviour that are not reducible to the formal abstracta (‘grammar’) required for the communication of propositional knowledge. Language behaviour is in the first instance based on the real-time dynamics of interacting bodies in and through which individuals relate to each other (Cowley 2008; Cowley et al 2004; Thibault 2008, In press). Individuals make use of the dynamics of vocal, gestural and other forms of bodily expression or signaling behaviour to coordinate perspectives and to move people – affectively, cognitively, interactively. Individuals respond to the micro-temporal details of the dynamics in ways that have the potential to bias action and perception and how these are or may be integrated with wordings.

    A few observations are in order here. First, Kanzi, Panbanisha and other bonobos at the IPLS in DesMoines have the capacity to coordinate their actions with those of humans in ways that are clearly social. Secondly, they use semiotic meditational resources to do so. Thirdly, they have the capacity to intentionally and affectively modulate their interactive behaviours in concert with their human interlocutors. Fourthly, they orient to and exploit for their own purposes local resources (e.g. lexigrams, vocalizations of human partners, other interactively salient bodily and environmental events). Fifthly, they engage with the ways in which their human interlocutors manifest their intentions and understandings of the situation. Sixthly, through statistical learning they connect phonetic (in human speech) and lexigram patterns to both aspects of situations and to their own needs and wants. For example, Kanzi’s actions are motivated by his own projects. He has his own perspectives on and understandings of the situations he participates in with his human interaction partners. These perspectives are informed by his status as a bonobo who participates in a joint pan-homo ecology (Thibault 2004, 2005a, b).

    The perspectives of Kanzi, Panbanisha and others cannot be totally assimilated to those of their human interlocutors for reasons of both biology and culture. And yet it is the differences (as well as the commonalities) of biology and perspective that are the motivators of dialogically coordinated pan-homo and human-human interaction. Moreover, the local pan-homo ecology of IPLS is perfused with norms that help to scaffold the negotiation of the differences between bonobo and human. The issue of our ‘narrativizing’ the observable behaviours of bonobos such as Kanzi, rather than refuting Kanzi’s languaging abilities, serves in actual fact to bolster my point that his understandings and those of his human interlocutors and observers are unlikely to coincide in all respects. Instead, the ability of humans to project their narrativized norms and expectancies onto Kanzi serves to illustrate that in dialogically coordinated social interaction, it is often problematic to say that norms, values, understandings, and perspectives are ‘shared’ by interlocutors. In actual fact, much of the observed interaction between bonobo and human only partially fits available interaction schemas. This leads to an interesting empirical consideration.

    It is an empirical fact that many encounters between individuals cannot be modeled on the basis of readily available discourse formats and genre schemas. Interactants are frequently required to construct models of the interaction on-the-fly without being able to refer to explicit genre or other conventions. Furthermore, the individuals so engaged must jointly coordinate their actions on the basis of these on-the-fly models of their interaction. Many spontaneous , even one-off everyday encounters between persons have this character. Moreover, the inherent recursivity of this coordination means that every linguistic move – e.g. in a dialogue (seemingly) between two persons – is in actual fact a move in multiple language games that involve everyone – past, present, future – who is potentially affected by the given move by a particular person in a particular conversation in a particular time and place and/or who may come to learn about it. Norms and conventions emerge over time to manage these uncertainties and the multiple reflexivities they entail.

    Indeed, it is the uniqueness and specificity of the perspectives – both bonobo and human – that are at play in their joint social interaction encounters that enable joint contextualization to take place. Kanzi and other like-minded bonobos have learned to use lexigrams and possibly also vocalizations as semiotic meditational means whereby others can interpret their sign-making as indexing their needs and wants. In this way, the contributions of both bonobo and human partners gel in ways that manifest a local interpersonal coherency (Thibault 2004). There is no need for Kanzi to communicate propositional thoughts to be a languaging agent. Instead, Kanzi and his human partners-in-interaction are able to coordinate their perspectives, bodily dynamics, vocalizations, and lexigram activity with those of their human interlocutors in ways that elicit shareable interpretations that can be integrated with the real-time event and the emerging sense of interpersonal coherency.

    There is no need, then, to posit the presence or absence of propositional thinking as evidence or not of languaging behaviour. Much everyday talk between humans is far less concerned with propositional thinking than the folk-theory would have us believe. As an episode analysed by the present writer shows (Thibault 2005c), Panbanisha, in that case, demonstrates the ability to some extent to reflect upon and perhaps to comment on her own behaviour when scolded by Sue for jumping on a dog during a walk together at their former home in Georgia. The evident moral self-reflexivity of Panbanisha’s behaviour and lexigram response on that occasion shows something akin to if perhaps less specified with respect to human contrition. Panbanisha even gives a lexigram undertaking not to repeat the misdemeanour. If bonobos can exhibit even primitive (by our standards) forms of self-reflexivity in their semiotic acts that are open to the moral values of her human interlocutor, then this would suggest that bonobos, like humans, are liable to cultural colonization of their brains in ways that can pave the way to some form of languaging behaviour as individuals seek to anticipate what their interlocutors know and expect about their shared worlds and how to behave in them.

    Knowledge of this kind is grounded in first-person experience rather than abstract propositional knowledge. Kanzi and Panbanisha probably don’t know that George Washington was the first President of the U.S.A., or that we are taking part in this Internet discussion on Sue’s paper. Neither of these facts are grounded in their everyday worlds. But they do know about fires, barbecues, pretend play with masks, lexigram symbols, etc. They know about these things (and much more) on the basis of their daily engagements with humans, the world they share with humans, their assumptions relative to their perspectives as to how such knowledge fits in with their behaviour and that of their human interlocutors in this world. In this way, they are also able to know a great deal about their interlocutor’s intentions. Zlatev (2008: 231) argues that “a degree of third-order mentality” appears to be available to the cognitive and communicative potential of apes. He argues that the realization of this potential depends upon “an environment that is rich in opportunities for developing triadic mimesis, i.e. a particular form of enculturation” (Zlatev 2008: 231). On this basis, Itkonen (2008: 288) discusses the “relevant three-level knowledge” and concomitant reflexivity as the basis on which interpersonal coherency in interactional encounters is created between persons in, for example, a transaction between a client and a bank teller. The interpersonal coherency that is typically achieved in such occasions is grounded in both the forms of reflexivity mentioned above and in relevant aspects of the social and physical environments (Itkonen 2008: 288). We can postulate on the evidence of the interpersonal coherency achieved between Kanzi, Panbansiha and some other bonobos at the IPLS that: Kanzi knows-1 what to do; he knows-2 that his human interlocutors know-1 what to do; he also knows-3 that his human interlocutors know-2 that he knows-1 what to do. His knowing may be less specified (e.g. semantically) than that of his human interlocutors, as seen from our human scientific perspectives, and its is almost certainly implicit. It is unlikely that he has formalized an explicit hierarchy of levels of knowing and hence of reflexivity! However, its very implicitness is also characteristic of how humans act for the most part in many occasions of every talk.

    To conclude, Sue’s work demonstrates that humans and bonobos (Kanzi and family) align with and coordinate with each other in talk. In doing so, they orient to and realize values in their shared worlds, achieving occasions of local interpersonal coherency as they do so. Regularities in the behaviour between persons and bonobos (not abstract symbols or propositional forms per se) serve as standards that persons can use to evaluate others’ behaviours in ways informed by their own perspectives and experience. Both humans and bonobos accordingly give that behaviour value and meaning relative to their own perspectives. We do not ‘use’ a symbol system except in the second-order cultural sense imposed on us by literacy, writing, and education. But that is not how much talk works even between humans. Rather, the second-order entities we humans learn to call ‘words’ and ‘sentences’ put constraints on what we do and, developmentally speaking, what we become as persons (Cowley 2008).

    Some of these constraints are symbolic not because of formal syntactic operations but because of empirical constraints deriving from the ways we learn to manipulate the symbols in interaction in order to achieve high-order control over aspects of culturally shared worlds. I would suggest that many interactional phenomena operate according to (interpersonal) constraints that arise from the self-organizing processes whereby persons and bonobos align to cultural values in their local ecology. The issue then is not, as in most conceptions of inter-cultural and inter-species communication, how one uses the symbol system of a given language when talking with ‘non-native users’ and vice versa, but how agents-in-interaction align to and are constrained by values that shape the interaction itself and its regularities. As Goffman (1983) showed, many of these constraints give interacting bodies value and meaning for the selves in interaction. Language is material and dynamical, in this sense, not abstractly symbolic. If it were the latter, it would have no causal powers in the world. Material interactivity affects us. This would lead to interesting linkages between body dynamics, values, and affect, but that will take me too far for now.


    Clark, Andy 1997. Being There: Putting brain, body, and world together again. Cambridge, MA: The MIT Press.

    Cowley, Stephen J. 2008. ‘The codes of language: turtles all the way up?’. In Marcello Barbieri (ed.), The Codes of Life: The rules of macroevolution, pp. 319-345. Berlin: Springer.

    Cowley, Stephen, Moodley, S., and Fiori-Cowley, Agnese 2004. ‘Grounding signs of culture: primary intersubjectivity in social semiosis’. Mind, Culture, Activity, 11(2): 109-132.

    Goffman, Erving 1983. ‘The interaction order’. American Sociological Review, 48: 1-17.

    Hutchins, Edwin 1995. Cognition in the Wild. Cambridge, MA and London, England: The MIT Press.

    Itkonen, Esa 2008. ‘The central role of normativity in language and linguistics’, In Jordan Zlatev, Timothy P. Racine, Chris Sinha, and Esa Itkonen (eds.), The Shared Mind: Perspectives on intersubjectivity, Amsterdam/Philadelphia: John Benjamins, pp. 279-305.

    Savage-Rumbaugh, Sue, Shanker, Stuart and Taylor, Talbot J. 1998. Apes, Language and the Human Mind, Oxford and New York: Oxford University Press.

    Segerdahl, Pär, Fields, William and Savage-Rumbaugh, Sue 2005. Kanzi’s Primal Languge: The cultural initiation of primates into language. Basingstoke and New York: Palgrave.

    Thibault, Paul J. 2004. ‘Agency, individuation, and meaning-making: reflections on an episode of bonobo-human interaction’, in Language Development: Functional Perspectives on Evolution and Ontogenesis, Geoffrey Williams and Annabelle Lukin (eds.), London and New York: Continuum, pp. 112-136.

    Thibault, Paul J. 2005a. ‘Brains, bodies, contextualizing activity and language: do humans (and bonobos) have a language faculty, and can they do without one?’. Linguistics and the Human Sciences 1(1): 99-125.

    Thibault, Paul J. 2005b. ‘What kind of minded being has language: Anticipatory dynamics, arguability, and agency in a normatively and recursively self-transforming learning system, Part 1’. Linguistics and the Human Sciences 1(2): 261-335.

    Thibault, Paul J. et al 2005c. (with Benson, James D. and Greaves, William S.). Scientific report as contribution towards Project #2 — Savage-Rumbaugh, Fields, Segerdahl, Thibault, Benson, and Greaves, Experimental investigations: The effect of the intentional introduction of forgiveness into a Pan/Homo culture. Funded by the Campaign for Forgiveness, Templeton Foundation (USA).

    Thibault, Paul J. 2008. ‘Face-to-face communication and body language’. In Handbooks of Applied Linguistics (HAL) Linguistics for Problem-Solving: Perspectives on Communication Competence, Language and Communication Problems, and Practical Solutions, Karlfried Knapp and Gerd Antos (eds.), Volume 2: Interpersonal Communication. Gerd Antos & Eija Ventola (eds.). Berlin. Mouton., pp. 285-330.

    Zlatev, Jordan 2008. ‘The co-evolution of intersubjectivity and bodily mimesis’, In Jordan Zlatev, Timothy P. Racine, Chris Sinha, and Esa Itkonen (eds.), The Shared Mind: Perspectives on intersubjectivity, Amsterdam/Philadelphia: John Benjamins, pp. 215-244.

  • Heidi Lyn

    Much of the argument surrounding the animal language field centers on one question — have these animals “acquired language”? Many researchers within the field find the question frustrating. No, neither Kanzi nor Panbanisha (nor any other language-using ape) can contribute to this forum. No, these apes do not use language in precisely the same way even a 2 1/2 child does, although this stage of language development is frequently used as the comparison base for their language abilities. However, neither does a dog, a dolphin, or a goldfish, for that matter. Does that make bonobos, dogs, and goldfish equivalent in their language abilities? After all, none of them “has language”.

    What Sue Savage-Rumbaugh and other animal language researchers attempt to do is look at what, of language, animals can learn. For instance, Rico, a Border Collie dog, could correctly respond to over 100 German words (Kaminski, Call, & Fischer, 2004), to date, a feat that no other non-primate species has mastered. However, no dog has been shown to use symbols productively. And as far as we know, goldfish are incapable of any symbolic processing. Are these differences uninteresting?

    The bonobos and chimpanzees in the Savage-Rumbaugh studies have been shown to be capable of many feats related to the acquisition and use of language : the acquisition of symbolic capacities without explicit training (Brakke & Savage-Rumbaugh, 1995, 1996; Greenfield & Savage-Rumbaugh, 1991; Savage-Rumbaugh, 1986; Savage-Rumbaugh, McDonald, Sevcik, Hopkins, & Rupert, 1986; Savage-Rumbaugh, et al., 1993); using symbols to name objects in double blind studies (Brakke & Savage-Rumbaugh, 1995, 1996; Savage-Rumbaugh, et al., 1986); associating novel English names with novel objects with very few exposures to both object and word (Lyn & Savage-Rumbaugh, 2000); utilizing imitation in an intentionally communicative context (Greenfield & Savage-Rumbaugh, 1993); making semantically-based combinations across both lexigram and gestural combinations (Greenfield, Lyn, & Savage-Rumbaugh, 2008; Greenfield & Savage-Rumbaugh, 1991; Lyn, Greenfield, & Savage-Rumbaugh, in press); mentally representing symbols on several levels at the same time (Lyn, 2007) and comprehending English sentences at least at a similar level to a two-and-a-half year old child tested in the same manner as the ape (Savage-Rumbaugh, et al., 1993). These abilities, I believe, are what Sue Savage-Rumbaugh refers to when she says that Kanzi and the others “use human language as you or I speak it” even if they do not show all components of said language.

    Many of the above findings came as a direct result of critiques of the animal language results. Critiques that suggested that, while the findings are interesting, the apes have not been shown to do X (with X depending on the interests of the critic). The most recent example is the current discussion in the evolution of language field about whether the ability to “declare” or use a symbol simply to share information, was the trigger for the development of human language (Moll & Tomasello, 2007; Tomasello, 2007). Regardless of the multiple reports of language-using apes making declarations, these critics claimed that “apes do not point declaratively, ever” (Tomasello, 2007). In response, myself, Sue Savage-Rumbaugh, and several other colleagues re-analyzed the data from Panbanisha and Panpanzee, resulting in considerable evidence that apes do indeed point declaratively (Lyn, Greenfield, Savage-Rumbaugh, Gillespie-Lynch, & Hopkins, in press).

    This is not to say there are not important distinctions (even in the case of the declaratives, the apes declare much less frequently than do human children). However, if we ignore the apes’ declarations since they are not just like those of human children, don’t we lose an important piece of information? Apes do declare. If we are looking for a biological difference between ourselves and these apes that explains why we “have” language and they don’t, the ability to use declarations isn’t it.

    Dismissing the abilities of these apes as “not language” misses the point — language is more than a dichotomous variable. The exploration into the abilities of the bonobos and other species give us the chance to understand more fully the pieces that make up human language and why, and when, they may have evolved.

    Brakke, K. E., & Savage-Rumbaugh, E. S. (1995). The development of language skills in bonobo and chimpanzee – I. Comprehension. Language and Communication, 15, 121-148.
    Brakke, K. E., & Savage-Rumbaugh, E. S. (1996). The development of language skills in Pan – II. Production. Language and Communication, 16, 361-380.
    Greenfield, P. M., Lyn, H., & Savage-Rumbaugh, E. S. (2008). Protolanguage in ontogeny and phylogeny: combining deixis and representation. Interaction Studies, 9, 34-50.
    Greenfield, P. M., & Savage-Rumbaugh, E. S. (1991). Imitation, grammatical development, and the invention of protogrammar by an ape. In N. A. Krasnegor, D. M. Rumbaugh, R. L. Schiefelbusch & M. Studdert-Kennedy (Eds.), Biological and behavioral determinants of language development. (pp. 235-258): Lawrence Erlbaum Associates, Inc, Hillsdale, NJ, US.
    Greenfield, P. M., & Savage-Rumbaugh, E. S. (1993). Comparing communicative competence in child and chimp: the pragmatics of repetition. Journal of Child Language, 20, 1-26.
    Kaminski, J., Call, J., & Fischer, J. (2004). Word learning in a domestic dog: evidence for “fast mapping”. Science, 304, 1682-1683.
    Lyn, H. (2007). Mental representation of symbols as revealed by vocabulary errors in two bonobos (Pan paniscus). Animal Cognition, 10, 461-475.
    Lyn, H., Greenfield, P. M., & Savage-Rumbaugh, E. S. (in press). Semiotic combinations in Pan: A cross-species comparison of communication in a chimpanzee and a bonobo. First Language.
    Lyn, H., Greenfield, P. M., Savage-Rumbaugh, S., Gillespie-Lynch, K., & Hopkins, W. D. (in press). Nonhuman primates do declare! A comparison of declarative symbol and gesture use in two children, two bonobos, and a chimpanzee. Language and Communication.
    Lyn, H., & Savage-Rumbaugh, E. S. (2000). Observational word learning by two bonobos: ostensive and non-ostensive contexts. Language and Communication, 20, 255-273.
    Moll, H., & Tomasello, M. (2007). Cooperation and human cognition: the Vygotskian intelligence hypothesis. Philosophical Transactions of the Royal Society B, 362, 639-648.
    Savage-Rumbaugh, E. S. (1986). Ape language: from conditioned response to symbol. New York, NY, US: Columbia University Press, (Chapter Chapter).
    Savage-Rumbaugh, E. S., McDonald, K., Sevcik, R. A., Hopkins, W. D., & Rupert, E. (1986). Spontaneous symbol acquisition and communicative use by pygmy chimpanzees (Pan paniscus). Journal of Experimental Psychology: General, 115, 211-235.
    Savage-Rumbaugh, E. S., Murphy, J., Sevcik, R. A., Brakke, K. E., Williams, S. L., & Rumbaugh, D. M. (1993). Language comprehension in ape and child. Monographs of the Society for Research in Child Development, 58, v-221.
    Tomasello, M. (2007). If They’re So Good at Grammar, Then Why Don’t They Talk? Hints From Apes’ and Humans’ Use of Gestures. Language Learning and Development, 3, 133-156.

  • William Fields

    This post wanders across a wide range of epistemological domains without explanation. Many of the contributions seem to ignore the critical, central, and original thread of the blog, “the science is in the rearing,” and therefore it is difficult for me to build upon the energy of the posts with any level of correspondence to the vast literature of Rumbaugh or Savage-Rumbaugh. Like freshmen in college dealing with pro/con topics such as gun control, the bloggers’ energies focus upon the nature of Kanzi’s ability in terms of whether his productive and receptive competence for English is human language. In my view, one can talk about this either scientifically or ethnographically. The blog, unfortunately, essentializes points out of either context. The contributors appear to strive to make intellectual common sense out of Kanzi’s behavior from a folk perspective and a subjective assessment of what they know about the research. Clearly, the contributors are not speaking empirically from the literature, for if they were they would conclude that the Kanzi facts do positively infer a subset of behavioral data, which may or may not meet a systematic and rigorous definition of language. That depends on the definition of language. (One day Chomsky is in and the next day he is not — independent of data.) Without a familiarity of the published research contrasted upon methodized definitions of language that do not violate the ethnographic facts of human language (see Dan Everett and the Pirahã people), how can one seriously continue this conversation as though it addresses the empirical facts of language and Kanzi? It is strikingly reminiscent of the one-sided communication of epistemology between the Church and Galileo that is cast as a dialog of logic. Or Kanzi’s illogical assumption that just because the beaver in our pond can “bite” a tree down with its teeth, the beaver therefore has the ability to come in the building and bite Kanzi while he sleeps.

    Let’s be clear, the subjective assessments of whether Kanzi has language (or not) has nothing to do with the data analysis that tells us he does. If the readers invest in an authentic engagement of literature from Lana to Kanzi (this includes the Sherman and Austin research and excludes the Panzee and Panbanisha and the Nyota research), which I realize is a significant expenditure in time (certainly I don’t know anybody who has been willing to do it, including reviewers like Clive Wynn, see our response in The Skeptic 2009), one will have a different empirical point of view about ape language research. If you believe empiricism can convey a truth which is not immediately accessible to one’s subjective common sense and you accept that Kanzi does have language, no matter what your ethnocentric human values tell you, then it is possible for the authentic investigator to move to a qualitative assessment (ethnographic) which is predicated upon an empirical baseline: Kanzi has language. With this is mind, it becomes the ethnographer’s job to cross the bridge from an etic to an emic perspective to understand the cultural quality of Kanzi’s abilities, as opposed to an ex cathedra attempt to map Kanzi’s cultural uniqueness in a match for match of first-world human uses and forms of written language (keeping in mind the majority of human languages are not written ones).

    And to frame this approach as scientific realism is astonishing. As Merlin Donald points out, while Kanzi does have some linguistic abilities, he will never be a Cicero. Well, in my opinion, neither will Merlin Donald. This does not deny the fact that Donald may be a linguistic being. Of course, that conclusion is dependent upon the outcome of empirical data. As far as I know, my good friend Merlin has not been a test subject for receptive and productive competence for English. So, empirically, I know far more about Kanzi’s abilities that I know about Donald’s. As a functional social matter, I have had conversations with both and my subjective reality tells me that Kanzi and Donald are linguistic and symbolic beings. Empirically, I can only speak about Kanzi. Now, if I could get Kanzi to authentically challenge his categories of “bite” and understand the cultural use of teeth in the beaver’s world, everybody would get a better night’s rest. Unfortunately, Kanzi’s cultural use of symbolic expression is far too human in its mythological practice. So as I spend the next decade with humans about “is it language or isn’t it,” I will, on the ape side, continue to have to reassure Kanzi that the beaver does not have the ability to bite him. Based upon the data trends, I can safely say nobody, ape or human, is going to listen to me. By the way, did I mention “the science is in the rearing!”

  • The original essay was designed as an initiation point in a discussion of why human beings and apes live in, and construct, very different life styles and worlds, though we are sibling species. The traditional answers: “brain structure” and or “brain size” — do not explain how humans with only one functioning hemisphere are able to lead relatively normal human lives. The traditional answer: “genetic differences” does not explain why most of the genetic differences between ourselves and apes lie in genes that control immunological differences, not in those that determine bain function and bodily structure (Perry, Yang, Marques-Bonet, Murphy, Fitzgerald, Lee, Hyland, Hurles, Tyler-Smith, Eichler, Carter, Lee and Redon 2008). Thus while our genetics and our brain structures do not set us apart, to the degree that seemingly they should; there is something that does differ radically between humans and apes. This is the immediate post-natal neuro-developmental ontological trajectory of our infants.

    The sensory motor-world experienced by ape infants is dominated by their ever-present need to cling constantly to the mother. Indeed, the most salient feature of the Order Primates is that primate infants cling from birth onward. Consequently primate infants experience the same world as the mother, whilst their brains are developing. They not only see, smell, hear and taste everything the mother encounters, they also experience constant emotional and kinesthetic feedback from her body during every experience she faces. They learn to anticipate and share her emotions toward everything. To survive they must achieve, and constantly maintain, the necessary bodily coordination with the mother that allows them to travel as a seamless unit through three-dimensional space (Savage-Rumbaugh and Fields, in press-b).

    Human infants, by contrast, are supported from birth by their caregivers. Because they are not required to focus their attentional and physical energies on clinging, they are afforded a radically different world view from other members of Order Primates. Their world is one that is constructed by the linguistic culture into which they are born. This world passed on anew, each generation, through acts of meaning-makings, which take place between caretaker and child from birth forward. Human children are thereby availed the opportunity of forming object/other distinctions in a way no other member of the Order Primates could possibly accomplish, because their hands, feet and attention are free to develop in myriad ways.

    Until one has reared an ape infant and experienced, firsthand, their overpowering need to cling, it is difficult to cognize the extent to which the clinging desire inhibits, in ape infants, nearly ALL the things that human infants begin to do shortly after birth. Human babies focus upon the development of object manipulatory skills, face to face gazing and exchanges of smiles and laughter. Ape infants are far too busy clinging to do these things. They cannot afford to lie on their backs and wave their hands and feet while laughing and smiling at the caretaker. They experience emotional disturbance whenever the mother moves and this emotional reaction initiates the clinging response. At any movement on the mother’s part, they must cling at once, even more tightly than they were already clinging to her while she was at rest. This overpowering need produces a high level of emotional arousal, centered around movement. It shapes the ape infant’s orientation toward, and categorization of, its social and its physical environment. It shapes the infant’s reactivity patterns and its level of arousal. It shapes the infant’s muscles in the shoulders, arms, necks and feet. As it does so it moves the ape infant toward a quadrupedal orientation. Just as the baby’s hands are used mainly for support, so do the hands become a significant means of support for the adult ape.

    It is unlikely that a human infant, were it to spend its first few years completely focused upon clinging to the mother, would still manage to acquire a human form of language; so pervasive is the difference between clinging and not clinging (Savage-Rumbaugh and Fields, in press-b).

    Regardless of the potential implications of how clinging affects cognitive development, many of the respondents to the article elected not to address this question. They preferred to offer their personal opinions on the topic of whether the bonobos of the experimental group (EG — Kanzi, Panbanisha, and Nyota), do or do not have language capabilities. In so doing they chose to ignore the fact that the bonobos in the EG contrast markedly with those in the control group (CG — Matata, Maisha and Elykia). Some of respondents assumed that virtually any bonobos, regardless of rearing, could accomplish feats of language, as long as a human experimenter was present to explicate, expedite, and narrate their communications. This is not true.

    It is the unique rearing environment of the EG that has produced linguistic abilities not even approximated by the CG, and/or by other apes who have been subjects of language studies. This can be attributed, in large degree, to the role that human beings have played in infant rearing in a bi-species world. The nature and significance of infant carriage and support, within a cross-species rearing environment, has only recently been discussed in detail; but this consitutes the most salient difference between these apes and all others who have been exposed to language (i.e. Washoe, Nim, Koko, Lana, etc.).

    All of the bonobos in the EG were exposed to significant amounts of human carriage, support and love from two months to five years of age. The CG offspring have spent their time clinging to Matata who also gave them love and support, but of a very different form. As a result of this rearing variable, not only did the EG begin to spontaneously understand, and use, human language during infancy; they also acquired many aspects of human social interaction and human culture. It is important to note that the bonobos in the EG were never reared as human infants, nor were attempts made to turn them into human beings. They were raised, instead, as part of the bi-species world which included both bonobos and human beings. Those in the EG formed strong social bonds to members of both species. Those in the CG bonded only to bonobos. The vast differences between these two groups with regard language comprehension and use has not been disputed. What is at issue is whether or not bonobos in the EG actually manifest the sum and substance of language as we human beings know and experience it — or whether what they are doing is somehow less than “real” language, or something more akin to wishful thinking on the part of those that work with them. The simple answer to this question is that if ‘wishful thinking’ and ‘narratizing’ can produce language in nonlinguistic beings, both work quite as well for the CG as they do for the EG. Wishful thinking and narratizing should also work for language delayed and autistic human children. The plain truth is, they don’t.

    Human ape infants are not going to acquire language simply because they do not cling. To acquire language, any infant must be enmeshed in a culture of language users. They must begin to develop a sense of self-awareness and a sense of free-will. Members of their culture must extend to them the embodied belief that linguistic competency is possible. As these rearing vectors of ape and child become increasingly similar, the linguistic skills they acquire become increasingly similar also. This is not a particularly challenging idea. Indeed nearly anyone knowledgeable in the field of developmental psychology would be likely to make such prediction. Yet some reviewers seem to view it as highly suspect.

    How far can one press the envelope in the case of bonobo infants, and how do differing vectors of manipulation result in differing forms of linguistic ability? We do not know how far the envelope can be pressed; we do know that we have not come close to those boundaries, nor have we attempted to press the envelope, only to expand it. We know that the intentional inclusion of Matata, a wild reared bonobo, into the Pan/Homo rearing environment has significantly affected a multitude of rearing parameters. The sounds, tactile and kinesthetic stimulation she provides have activated infant behavioral profiles in ways no human stimulation could ever accomplish.

    Some of the commentators, however, feel that there is no reasonable scientific question to be addressed, as all the answers are already in, and they know that apes cannot acquire language. Harnad succinctly states this view, on behalf of himself and some of the other respondents noting:

    1. Kanzi and family cannot say anything and everything.

    2. “…no matter how much linguistic understanding we attribute to them — they cannot enter into this “conversation” we are having in this Forum, not even into a rudimentary approximation to it — whereas any speaking human being, speaking any (spoken or gestural or written) language, can; even a child. And the most likely reason for that is that bonobos cannot understand propositions (statements with a truth-value) and that when they do make or understand propositions…what they are actually doing it not what we think it is.”

    It is tempting to respond Harnad’s assertions by representing summaries of published studies in which the data illustrate that apes can make and understand propositions (Savage-Rumbaugh, 1997; Savage-Rumbaugh, 1999; Savage-Rumbaugh and Fields, 2000; Savage-Rumbaugh and Fields, 2007; Savage-Rumbaugh, Fields and Spircu, 2004; Savage-Rumbaugh, McDonald, Sevcik, Hopkins, and Rubert, 1986; Savage-Rumbaugh, Murphy, Sevcik, Brakke, Williams, and Rumbaugh, 1993). But Harnad forwarns us that these facts are of no consequence because the issue has already been decided, presumably on grounds other than data. Harnad professes a kind of personal knowledge, apart from the scientific data, regarding that which Kanzi and family are unable to do. Since Harnad has not attempted research of this sort himself, nor collaborated with us, he must be drawing upon personal opinion. Perhaps he has spoken with those who have made brief visits to the bonobos, as did Derek Bickerton. Whatever his source, it is inadequate.

    So let us abandon data and simply answer Harnad from the perspective of logic. While Harnad is correct in his assertation that the EG cannot say “anything and everything” with the keyboard, what does this mean with regard to functional language? The EG communicates through what is called, in the field of mental retardation, an “augmentative communication device.” This means that they, in addition to vocalizing and gesturing, touch symbols on an electronic panel, or a paper symbol board. Because there are a limited number of symbols on this board, no one can say anything and everything with it. However, if one allows for metaphoric use of symbols, one can potentially say anything and everything. The EG allows for metaphoric use. For example, Kanzi recently called kale (not on the board) ‘slow lettuce’ — because it takes a long time to chew. Not all of Kanzi’s metaphors are understood by everyone, and he tries to say things at this keyboard that are not readily interpretable from the human perspective. Thus being able to say everything and anything is not equivalent to being understood by all other parties. Given that the keyboard was not designed to say anything and everything, the limitation which is of concern to Harnad is a limitation of the current system, not of the bonobos.

    The EG does need more words. The limits of his keyboard are NOT the limits of their world. They know many more English words than are presently on the keyboard, as demonstrated in Fields, (2007) and Fields, Segerdahl and Rumbaugh, (2007). Lacking a way to say things can be very frustrating for them. A few days ago Kanzi was trying to indicate his desire for a person who had entered the lab to come see him, but her name was not on the keyboard. He could hear her talking, but I could not. Kanzi gestured and glanced in the appropriate direction but I did not know she was in the building several rooms away. Finally Kanzi displayed at me, upset with my inability to properly interpret and narratize his communication. When I learned who was in the building, I began to put the facts together. I then inquired of Kanzi whether he needed names for everyone, and if that was why he had displayed at me. He adamantly answered “YES,” and as soon as a board was prepared with such names, he began to using the name of the person he had been previously trying to convey to me. The EG’s world of comprehension outstrips their lexical world of production by a factor of ten or more.

    Still, the question remains, are there things that EG could not say, even if the words were on his keyboard. In this vein it is perhaps worth observing that recently a dog has been demonstrated to comprehend over a thousand words. The dog does not have a keyboard and lacks any means of producing words. If the dog had a keyboard, would he employ it all, even in the limited, nonpropositional manner which Harnad and other critics attribute to Kanzi? We do not know. What we do know is that the EG can employ the keyboard to ask for things, to name things, to make statements of intent, to make statements of his state of knowledge, to recall past events and to refer to future events. Examples abound in Segerdahl, Fields and Savage-Rumbaugh, (2005); Lyn, Franks, and Savage-Rumbaugh, (2008); Greenfield, Lyn and Savage-Rumbaugh, (2008), but a more in depth analysis and filming of these kinds utterances of the EG is warranted and will be presented in the future. In addition, the level of utterance complexity has increased across time as Kanzi and Panbanisha have grown older and new data are now needed.

    What Harnad and citics fail to offer is helpful information regarding what they would accept as evidence for propositional communication, from either a dog or an ape. A proposition is a kind of truth statement, and as such, Harnad is right, propositions lie at the heart of human language. If apes (or dogs) did not know that words ‘stand for things’ and can be used to make statements about things, they would not be using language. Clear evidence for propositional use of language came early on in the field of ape language, long before Kanzi was born. Lana, the first chimpanzee to acquire lexical competency, made many spontaneous propositional statements, including “Please give banana which is black” for an overly ripe banana in 1977 (Rumbaugh, 1977). She understood her statements. In the case of asking for the banana, she pointed to the black banana (to distinguish it from the yellow one) and gestured that it be given. She was not taught to use gestures, yet she employed many of them proficiently and appropriately. Lana also engaged in conversations, not unlike the one in this forum in that there was give and take and clear evaluations of the truth value of statements that Tim Gill, the researcher who raised her, made (Savage-Rumbagh, in prep).

    However, Lana’s abilities were dismissed as ‘random variations on stock sentences’ — by critics who took the sum total of all of Lana’s utterances and concluded that most of them were either sentences on which she was trained, or “random” variation of sentences on which she was trained (Thompson and Church, 1980). Of course, unlike Kanzi, Lana was trained. Random utterances came only as she was being taught new things. Lacking comprehension of spoken language (she was not raised in a bispecies English speaking world as was Kanzi), her only method of acquisition was trial and error, by design. Thus she should not be faulted for employing this method. She should be lauded for her truly expert ability to figure out how the humans around her were trying to employ symbols in a mean-making exercise. Once Lana grasped a concept and had devised an accepted means of expressing it, she no longer produced random strings. She produced only novel correct propositions, in novel contexts, and she engaged in conversational give and take. In brief, Lana answered all the “important” questions about what apes could and could not do with language, in the first few years of her life (Rumbaugh, 1977; Savage-Rumbaugh, in prep). She did so in a clean pristine experimental manner with every utterance recorded. When language is acquired by listening to adults in the speaking community, as with Kanzi, myriad different variables exist. The fact that Lana could not produce any and all propositions was to be expected as she received a limited set of training sentences and words. Yet she unequivocally produced propositions and produced them correctly on a trial one basis in novel contexts.

    Unlike Lana, the bonobo EG did not receive an experimental “training” background. They can say anything they can understand once they map the lexical symbols, and their modes of usage, onto their formidable capacities for English comprehension. Kanzi began to make simple propositional statements, in the form of “X chase Y” by three years of age. Like Lana, Kanzi demonstrated that he knew what he was saying, for he would take X’s hand and lead them to Y. Having seen X chase Y, Kanzi would then request the inversion of these circumstances by asking Y to chase X. If there were more than two visitors, he would expand his requests to include others (P, Q and Z) and again make novel combinations. Kanzi wanted to visibly see the kinds of social relationships people manifest as they chased each other. Kanzi then expanded his requests to include tickling, hugging, grabbing, and many other forms of personal contact typically ‘prohibited’ by adult human social norms.

    In so doing, Kanzi is employing a human language to say something that the human recipients often did not understand, unless it was explained to them or “narratized.”

    The intent of Kanzi’s utterances were so different from their own value system, that the “anything and everything” Kanzi would utter to them about his social world, did not “make meaning” to them, for they were trapped within their own cultural framework.

    Still, it could be objected that although apes can make statements about things they want and that they know the truth of that which they want — they nonetheless are unable to realize the statements they are making have a truth value apart from their own desire. Can they judge statements as true or false? Do they know that their own propositions are abstract? Do they know the difference between statements that are true and those that are not? Lana was able to tell Tim when the wrong food was placed in her machine and able to ask Tim to remove it, on a trial one basis, employing novel utterances in both cases (Rumbaugh, 1975).

    If one takes the bonobos’ conversations as offered, the answer is again clear, yes they do understand this and they understand it quite well. However, there is a Catch-22 in reporting “data” to prove this. The statements which Kanzi and Panbanisha make must be accepted as having been produced with sentience and intentionality for the purpose of making meaning to the listener. In other words, we must assume that Kanzi and Panbanisha know what they are saying in order to determine if they can assign a truth value to a proposition. I offer a recent example. Kanzi’s favorite objects are balls. He likes some balls much better than others and he keeps favorite balls for months; others are less preferred and they may last less than a day before he bites them open and deflates them. Sometimes the other bonobos also destroy Kanzi’s balls, but it is generally Kanzi. A few days ago Kanzi was given a new kind of ball termed a ‘sky ball’ because it bounces 75 feet into the air. After a few days, I noticed that Kanzi’s sky ball was missing. I inquired as to whether or not he had bitten it. He denied doing so, shaking his head no and uttering a sound that indicates negation. I then asked if Panbanisha tore up the ball and he indicated that she did. I did not believe Kanzi and I so stated. Kanzi held firmly to the truth of his position — that he had done nothing to the ball. Later I learned that another person had seen Panbanisha tear up Kanzi’s ball. I apologized to Kanzi for not believing him and for incorrectly narratizing what had, in fact, occurred.

    The point to draw with this example is that I made an inquiry of Kanzi, about his knowledge of information regarding a past state of affairs. He provided the answer I requested, yet I did not believe him. One might say that my ‘narratizing’ in this case was that Kanzi either did not understand my question or that he had forgotten that he destroyed the ball. However, my ‘narratizing’ was incorrect. Kanzi knew the truth of this his statement and stayed with that truth in spite of my refusal to accept it. Many other similar examples could be cited, all different, but all consistent in their indication that Kanzi and Panbanisha understand the truth value of utterances, and not only utterances about present events but about past events as well.

    The supposition that their numerous propositions can, and should be, explained away by the narrative that researchers project into the situation is a fundamentally flawed notion, as the example above begins to elucidate. Language, by its very nature, is the establishment of a joint narrative: a joint way of linking events and viewing the world, initiated and maintained by dialogue. From this conjoined way of viewing the world to work, the narrative has to be both culturally and linguistically grasped by all parties of the dialogue. To put it bluntly a narrative that is not accurate will not work, either in the human world or in the bonobo world. For this reason no narrative will work with the CG. It is not possible to engage them in a dialogue and provide some kind of running narrative that corresponds to their vocalizations and their view of the world — because whatever language they may have, it is not one shared by the authors.

    To put it another way; it is precisely the narrative, both cultural and linguistic, that is constructed between the EG, and those who know speak with them, which allows for maintenance of a bi-species socio-linguistic world. It is this that permits symbols, and their combinations, to become an experience of meaning making. When meaning is made, promises can be kept, expectancies fullfilled, social life thereby negotiated and the social contract constructed. This process takes place daily within and between a community of human beings and bonobos. However, there is no precedent for this kind of research — apes have always been studied as beings who are not capable of such a thing. Thus the kind of data that support the very facts of their meaning-making, are also the kinds of data that are ruled “unacceptable” for apes, but “acceptable” for human beings.

    Within the limits of Kanzi’s and Panbanisha’s world they understand essentially anything and everything. As noted above, one can easily ask them complex novel yes/no questions about their world and get reliable and replicable answers. One of the most intriguing questions that can be asked is whether the CG has language. They always answer yes, but when asked to tell Matata many things, they will refuse to translate, yet others they will always translate. There appear to be categories of things that Matata does not comprehend and therefore Kanzi and Panbanisha cannot say to her anything and everything that they understand about their world. For example, Kanzi and Panbanisha recognize that some snakes are good and will not harm you while others are bad and must be avoided. They also recognize that some snakes are not real, but toy imitations. Matata treats all snakes as bad and all toy snakes as real. Therefore there are things that cannot be said to Matata about snakes because of the categories of her world.

    The gulf of differences between the EG and CG make those who interact with them daily acutely aware that language has changed the brains, and the bodies, of the EG in fundamental ways. There is no doubt but that language has provided the bonobos in the EG with a self identity, a past and a future, a self narrative, a self concept and even a rather large degree of pride and recognition of their own notoriety, which is evidenced in how they respond to televised productions of themselves. The bonobos in the CG forever consider themselves a part of all the other CG bonobos in their feeling world and in their thoughts. There are no parts of their world that belong to them and them alone, all is shared. In such a shared world, propositions do not require a truth/false value. All is truth — at least there can be no individual opinion about what is true and what is not true. The EG, on the other hand, realizes that most of what takes place in the world of language requires negotiation and opinion. They are often polite and even hesitant to express an opinion or preference that may offend someone in the bi-species group. However they are, nonetheless, definite about the assigned roles each person must play in the group. They do not hesitate to enforce sanction for any infraction of assigned roles, and they can elect to employ physical means. What has emerged is an unusual kind of bicultural approach to sociality. It is one that the bonobos have formed which merges bonobo and human normative values in novel and creative ways.

    It is this which causes Derek to note:

    “When I met Kanzi, my immediate impression was, “Wow! Watch out for this guy!” I wasn’t faced with an intelligence that was kind of like mine but not quite up to snuff. I was faced with a formidable intelligence of a quite different order, with different skills, different functions, different goals. Luckily for me we were forced to meet on my turf. If we’d met on his turf, I know who’d have come off best, and it wouldn’t have been me.”


    “What would you do if you were placed in captivity by members of another species, aliens, say, who wanted to study you? You’d try to get along with them as best you could. You’d use your intelligence to figure out what they wanted. They want X? Okay, I’ll give them X! But I’m still me, and as far as I’m concerned, X is not really a part of that at all.”

    Derek’s comments may be accurately applied to Matata, but not to Kanzi. Matata was removed from the wild and placed in captivity just before puberty. She has little time for humans. She behaves as though humans are clearly inferior to bonobos and looks down upon them. Kanzi was born into a 50 acre forest in Atlanta, Georgia. He roamed it at will, with no lead, till he was eight years old, in the company of human beings. He now has several acres that he roams at will on his own or with other bonobos and hopefully will have more. It has gone unrecognized by those who view the research from afar, that as the members of the EG grew up, many serious choices had to be made about the qualities of their future lives. No matter how far they went into the world of language, they would not be accepted into the human community. They would need to acquire bonobo sexual patterns and bonobo display patterns. They would need to understand how bonobo adults negoiate power. When human children acquire language they also acquire the human cultural mores, styles, strategies, rules, myths and rituals of their group. Members of the EG could have been taught these aspects of becoming human, had a decision be taken to develop their normative social sexual behavior according to human standards. Quite the opposite was done: it was determined that they should receive their instructions in these domains from Matata, who was most happy to assist. She seems to view the passing of bonobos’ ways of behaving to the next generation, as a moral imperative. She has definite ideas regarding how a bonobo should act, and engages in active teaching of these things to young bonobos.

    If one is a bonobo in the wild one must:

    1. master three dimensional space,

    2. remember where hundreds of foods are to be found and compute exactly when they will be ripe,

    3. negotiate the daily travel decisions of a hundred other bonobos like oneself, all of whom have ideas about where it would be best to go next,

    4. be ready to defend oneself and one’s group against any stealthy and fast predators in the mammalian kingdom, the insectivore kingdom, and the reptilian kingdom. Bonobos must be able to communicate with other members of their group when they are out of sight. This is most of the time because the vegetation is so thick as to make visual communication in adjacent trees impossible and, unless they are sitting on the same branch, visual communication within the same tree is not possible.

    5. be willing to share food, sex and child care, as these are three pillars of bonobo society.

    Matata did not leave learning up to “observation” — she took her charges by the hand, literally, and instructed them by physically insisting they learn about sex, and by physically insisting they acquire the detailed rules of food sharing, and punishing them with a bite if they failed a subtle distinction. According to their self report, she tells them many other things. While we cannot understand what she says — and cannot verify their reports — we can see that she studies animal photos intently and she intently models how one should treat dogs, cats and other animals. In addition she tries to instruct them in the hunting of animals outdoors, but to date, they have refused to acquire this aspect of her enculturation.

    Thus Derek is quite right to assert that they did not meet an intelligence that was like his own, but a bit dim witted (as human ancestors are always depicted in public documentaries that reconstruct our evolutionary history). He met something quite different from himself because it was Matata who trained Kanzi how to respond to strange males of the human species. Kanzi has the quickness and the musculature of an ape and he knows how to keep human males at a distance, where Matata believes they belong. What Derek did not see was Kanzi when he was calm and in his ‘human mode’ or style of interacting. When the issue of which male is in charge can be removed, suddenly Kanzi becomes much more reflective, he moves more slowly, he maintains steadfast eye contact, he seeks human approval, and he orients his comments and his vocal and lexical utterances toward the human framework of reality.

    Kanzi is a denizen of two worlds. Derek but one. Derek tried to narrate and anthropomorphize from his human perspective, but it was not working for him. Derek was suspicious of Kanzi and Kanzi was suspicious of Derek. There are many emotions to get past when male scientist meets male ape, in a kind of mental duel as to whether or not Kanzi’s use of language is ‘real.’ Kanzi’s comprehension of human intonation, facial expression and manner is detailed and rapid, and his response to anything other than honesty and positive acceptance can be quite pushy and overbearing. There are a whole host of emotions that both Kanzi and any male visitor must move beyond in order to achieve rapport. This does not happen rapidly on either side. What would Kanzi have been like had the moral expectancies of human culture been as intentionally instilled within his being as is the spoken English he comprehends and emulates with his bonobo vocal tract? We did not know. That experiment was intentionally not attempted. What is known is that the attempt to do so went farther in the human direction with Panbanisha and that she has developed a far greater capacity to self-regulate and control her emotional outburts than Kanzi.

    A significant impediment most visitors encounter is their lack of knowledge and understanding of apes per se. Having never personally met an ape prior to encountering Kanzi (except on display in a zoo), visitors are often overwhelmed by his “apeness”. While they may have seen apes before, they have not seen apes who are trying to communicate with them. There is no doubt that Kanzi is trying to communicate when you meet him. He exudes communication of all types and visitors are overpowered by a wealth of intentional communications, designed to lay bare their basic emotions. When Kanzi challenges them to chase and grab, in place of sitting quietly and speaking in long sentences, they become disoriented. Because Kanzi behaves like an ape, rather than a human being, they are tempted to discount his lexical utterances combined with gestures, glances and vocalization.

    They are prone to assume that if Kanzi’s language is “real” it should result in more normative comprehensible human-like behavior, which translates to a desire to see Kanzi behave like a little gentleman.

    But Kanzi was raised to BE an ape, not a person. This confusion of human culture with human language runs deep because human language and human culture have always gone hand in hand before. All humans, for example, wear clothes and disguise their sexuality. Kanzi does neither. Could he have been taught to do so? Most likely he could have, if he had never met other bonobos. However, the longer one is around the EG the more human-like they reveal themselves to be, as the deep changes wrought upon them by language per sebecome increasingly apparent. Taking in the truth of EG is not a one night stand, but with increasing familiarity the certainty of their linguistic ability grows rather than diminshes. The problem is not that the communications of the EG are truly difficult to understand. The problem lies in getting the newcomer to take what they are seeing, hearing and thinking — as serious intentional communication.

    Therefore nothing replacing the real-life encounters one experiences as they become accepted into the bi-species world. The realizations that pour over those who begin to understand the EG, and to interact without a wire between them, are life altering. Not many try, because of the sustained instensity of effort required to set aside one’s normal idea of ‘humanness’ in order to enter into the bi-species world. The only similar experiences are those encountered by anthropologists who left the Western World and sought to integrate themselves fully into groups of remote indigenous people, with no knowledge of their language or culture. Prior to the rise of ethnography, it was thought that the languages of indigenous peoples were primitive as were their minds. Anthropologists such as Malinowski and Boas discredited these views by becoming part of the groups which they set out to study.

    William Fields, the ethnographer who has succeeded in integrating himself into the bispecies world, acquiring its language and its culture, set out to either reveal the flaws in the work or to produce the first ethnographic account of nonhuman linguistic beings. Finding the bonobos far more linguistically competent than any reports to date had revealed, he set about to construct an ethnographic account. Only small portions of this ethnography have yet been published, but his encounter validates the fact that the EG has language — and that they have sufficient language to engage someone who did not rear them. They did so through dialogues which allowed him to learn enough about them to begin to live with them. As he entered the group, he was assigned a role, as are all anthroplogists who enter into the lives of small indigenous tribes. It is not possible to permit an outsider in without assigning them a role and expecting them to fulfill it, since everyone in the community — to be accepted — must have some value to every group and thus must fulfill their expected role. In the Pan/Homo group, ignorance of one’s assigned role will be tolerated only for a short time, and punishment or banishment from the group are the only options when the given role is not accepted. The story of the integration of William Fields in the group is yet to be told. It is not only the story of his increasing comprehension of their language, it is a story of the many prominent scientists and journalists attempting to co-opt him into denying the linguistic capacities of the EG, to discredit the work.


    As Andy Lock points out, language can be pressed into a level of philosophical abstractness that it typically does not enjoy in the real world usage we extend to individuals engaged in the business of living.

    “Work in human-infant interaction provides the view that human infants develop as human beings because they are treated by their caretakers as if they already were human beings. Infants are not presented with a booming, buzzing lived-in world that they have to bring their considerable intellectual powers to in order to make sense of the significance of events and objects in the world they have entered. Rather, their perceiving is structured as their attention is drawn to the local significance of events and objects as they are transacted to them in their real-time participation in shared worlds of socio-cultural interaction. In the developmental context, developing human consciousness is distributed (Toulmin, above) unequally to begin with. Adults, in their conduct, create a zoped [zone of proximal development] (Vygotsky, 1966) that imbues the sense infants can make of their jointly experienced world; infants develop so as to join in.”

    These kinds of insights Andy offers above are eloquently laid out in his book The Guided Reinvention of Language. This book clearly illustrates why language is not something that arrives de nouveau in the child solely as a function of maturation. Instead, language is something that the culture builds into each child anew in each generation. This book, plus Action, Gesture and Symbol (also edited by Andy Lock) were seminal works which determined the kind of environments that were designed for Kanzi and his family, with zones of proximal development (zopeds) intentionally laid out to foster language use in a manner that would make sense to an ape.

    As the correct kinds of zoped were created, the question arose how far apes could go along the language trajectory. We are continuing to answer this question. However Andy notes that “zopeds” cannot appear on their own. For Andy, the question is NOT about what an ape can acquire (or not acquire) but about what kinds of zopeds exist, and how the zopeds are scaffolded initially; that is, how did human consciousness bootstrap itself? It is a sort of chicken and egg question.

    It was precisely this issue I was seeking to address by considering what occurs when an infant is born that does not cling. If that infant is going to survive, it will have to call forth in the mother (and other caretakers) a complexity of behaviors and “other oriented consciousness” that are not normally manifest in apes. Any such infant, as a result, will itself become distinctly different from other apes. To initiate this process, an infant that does not cling at all would not be required, but rather one that simply clings with less intensity, thereby placing more of the burden for its survival upon the mother. Such would be sufficient to initiate a spiraling up of the propensity for maternal carriage and care. Indeed, if we compare apes to monkeys we already find that things headed in this direction.

    The objection that such is a “western” notion overlooking the fact that in other cultures many people participate in child care is misplaced. The primary caregiver can be mother, father, aunt, uncle, sibling, etc. The basic point remains the same. Infants who make their way in the world by clinging, will inevitably develop and refine neurological and anatomical programs that interfere with the “aboutness” of propositional communication. Propositional communication lends itself to things and thingness.

    It is things and thingness which belong to humans in a unique way. We are the only species that does not use our forelimbs for locomotion; we employ our forelimbs to manipulate things, construct things, grow things, measure things and carry things. As apes begin to do this, their zopeds are altered, and their neurodevelopmental trajectory can become bootstrapped in a new manner. Many of the behaviors that ape mothers naturally scaffold with much older infants, can occur at an earlier point in development when the infant does not cling, thereby changing the course of development. Thus the question is not a matter of from whence do the zopeds arise in the beginning, but rather one of the timing of their appearance in the developmental plan.


    Givón is concerned that longer, more grammatically complex, sentences are not produced. They are, but a definitive and complete account of this remains to be made. A definitive account will require the use of iPad devices that the bonobos can carry everywhere, and/or installation of electronic keyboards in their enclosures, along with cameras in each area that can track utterance context. In addition these data will require detailed analysis such as that currently underway by Benson, Greaves and Thibault — of the entire setting as it changes and reconfigures from topic to topic across time.

    It is worth noting that the kinds of sentences Givón wishes the bonobos to construct on a regular basis reside in a vocabulary which was not added to keyboard until Kanzi and Panbanisha were adults. The failure to locate these words whose meaning derives from function, not referential content (i.e. here, there, it, yours, mine, before, until, time, to, is, be, not, now, later, yesterday, tomorrow, up, down, right, wrong, etc.) resulted from the original assessment that the bonobos, at the time Kanzi and Panbanisha were small, were not likely to comprehend or wish to employ these abstract vocabulary items. They do comprehend these words, which suggests that this assessment was incorrect. Adding these words “after the fact” has not been the best way to build lexigram sentence structure into their utterance patterns, but science progresses in baby steps much of the time.

    Earilier work with Lana illustrated the ape’s competency for longer and more complex utterances in training-defined contexts. Had we sought to press Kanzi toward the kind of sentence structure employed by Lana, he would have succeeded as well. The choice that was made for Kanzi was to develop his spontaneous comprehension and to integrate language into life. Givón is judging too harshly, assuming from what has been reported to date, that the limits of Kanzi’s capacity, and/or that of other apes, is at hand. Such is not the case. Most of the data reported to date were collected prior to adulthood and prior to here at Great Ape Trust, where the control and experimental groups have become integrated — and where the comprehension abilities of Kanzi, Panbanisha and Nyota have increased.

    They are now more at ease with the kind of grammatical devices Givón seeks, in their comprehension of spoken human language. They have no difficulty following a series of utterances, such as “What did you do with it? Did you put it over there? I mean the one that I gave you earlier, where is it? Did someone take it? Do you know where they put it?” — all with just an initial identification of the referent of “it.” If one changes the order of the words in such sentences (i.e. did it what do with you) the meaning is lost, the grammar is no longer operative, and they do not follow what is said. Their inability to construct sentences of equal complexity may simply reflect the fact that these sentences are long and difficult on the lexigram board and often they are not modeled even by staff. Using the board to construct long sentences and/or using it fluently with the functional words is a difficult task even for human adults who are already competent language users. In fact, even with a sophisticated computer based training program, students still find if hard to make such sentences at the keyboard. The degree to which the use of such an augmentative system itself is difficult, above and beyond the task of language per se, is a significant factor in encouraging the production of short utterances, both by the bonobos and by human beings.

    Much more distressing with regard to Givón’s response however, is his citation “Givón and Savage-Rumbaugh, 2009″. Prior to reading this response I was completely unaware of any input I made to any paper which he wrote. Certainly I do not deserve to have been listed as co-author to the publication since I:

    a) did not see a copy of the paper prior to its publication
    b) do not agree with the conclusions of the paper, though the data, as reported, is accurate,
    c) did not collect any of the data, write any portion of the paper or evaluate or review the paper.

    Givón did design research to be carried out with the bonobos at the Language Research in Atlanta, Georgia. The research he was funded to do and which I agreed to conduct with him was radically altered once the funding was in place. Givón did not wish to attend to the fact that Kanzi, Panbanisha and Nyota acquired their language skills sans repeated trial training tasks. He elected to impose didactic repeated trial reward-based presentations upon them as a language training paradigm. Additionally he required that I be absent from all such test sessions, that the data be sent directly to him, and that I have no part in the testing, which was done by Liz Pugh. In brief, Givón asked that the laboratory circumvent and interrupt the normal mode of linguistic expression and comprehension in order to work with him.

    After two years of requiring didactic training, Givón had no success in getting Panbanisha to respond correctly, in a noncommunicative context, to the words he required her to acquire. He did not work with her himself on any occasion. Failure in this didactic training tasks was predicted. It does not speak to Panbanisha’s abilities when engaged in linguistic communication rather than multiple choice noncommunicative right/wrong tasks that are intrinsically devoid of communicative content and which had been steadfastly avoided until this time. Givon’s tasks were nonfunctional, boring, repetitive, and meaningless to Panbanisha. They had no direct relevance to language as a referential communication system, either in terms of how they were presented, or with regard to the implications that allow themselves to be made regarding her capability.


    Dubreuil, having visited the Trust and seen something of the bonobo’s comprehension of language, dares to raise an intriguing question:

    Once we admit that, in eloquent primates, ‘consciousness’ (if we have to stick to this word) and language are closely intertwined (I would agree with this), what is the exact nature of this link? It sometimes seems that Savage-Rumbaugh envisions language as something creating the self, morality, etc. Or could it be (as I sometimes understand it) that language rather expands, shapes, orders and/or channels already existing mental phenomena?

    Does language create the mind — or is it rather an expression of mind that channels our attention in certain ways? This question poses many conundrums if we think about it from the perspective of different cultures and different language users. But the question loses its force if we but reflect upon the fact that many autistic persons do not acquire language. Such children may or may not recognize themselves in mirrors and they may or not may attend to persons who have died, as Dubreuil muses many animals do. For the parents of such children, these matters are irrelevant. What is relevant is whether they are able to communicate with their child. Parents will accept ANY kind of symbolic communication that works. One- and two-word sentences are great to the degree that they constitute real communications which help with the daily negotiation of life. This is because of the great need for the kind of give and take that leads to human normative behavior; i.e. potty training, polite behavior in the presence of others and a meaningful sharing of emotions. Echolalic, grammatically correct long utterances are not helpful. Any language containing propositions and grammar is not useful if it is decoupled from the business of life and the emotional give and take that characterizes the human world.


    Thibault, who has carefully studied and analyzed videos of Kanzi and Panbanisha, observes: it is in this give and take that we locate the real “stuff” of language. As such, what matters most is not what language IS, but what language DOES. Does it construct a particular human kind of morality though human normative relationships made possible by dialogue? The short answer, is “yes it does.” The long answer, which explicates the how and why of this, requires a book.

    Apart from all the observations that have been offered regarding the research with the bonobos at Great Ape Trust, there are some things that all too often go overlooked. These are noted below.

    1. Each different mode of rearing and language presentation has produced a different result in apes reared under those conditions. This fact alone tells us that the science is in the rearing and that method matters. The real question is not about what apes have or do not have. The real questions are about the methods by which various components of human language are brought into self-manifestation in apes and the implications of these findings for humanity’s understanding of self.

    2. The research is only now moving beyond the first generation of linguistically competent bonobos and only now beginning to study these language capacities with a group of competent apes. Critics should not behave as though “the final word is in.”

    3. Kanzi and family are now being contrasted with a control group, all of whom share familial and cultural ties. The control group is difficult and costly to maintain. They cannot share directly in the bicultural world of Kanzi and his family, because they construct concepts that are monocultural. They have as much difficulty as do monocultural humans, in relating to Kanzi’s world.

    4. No replications of the Rumbaugh, Savage-Rumbaugh and Fields paradigms have yet been attempted. Therefore, any conclusions which disagree with the findings of the authors are premature.

    5. The fact that these apes must be held captive in a setting which requires them to be treated in certain nonhuman ways, and to be exposed to living conditions that are not at all conducive to language acquisition and/or production, has gone unrecognized. These apes have acquired language in circumstances that often force them to forgo the language knowledge they have.

    6. The research is difficult and demanding. Publications emphasizing the linguistic limitations of apes are easy to get published and bring attention. The truth is far more complex to get across. The living environment of a group requires constant linguistic input and attention, and the personal needs and feelings of the bonobos which have come to ride upon their language skills must be acknowledged and appropriately managed in a world that discounts them of personhood and linguistic competency. The moral responsibilities to beings who have such capacities are significant and occupy much of the time of those working with them. Were it not for the fact that the language competencies of the bonobos far exceeds anything even claimed for them to date, let alone the few skills acknowledged by critics, the effort would have been abandoned long ago. The effort has been and will remain a simple one, to tell truth of what has been learned and to find means of enabling others to know that truth as well.

    Finally, on a personal note, I shall state very simply: Had I any idea how much language these bonobos would be able to acquire I would never have been able to undertake the work itself. It began as an interesting challenge and as a puzzle. Why were beings who seemed at least as intelligent as two or three year olds not acquiring language? The research has taken me far beyond that question. It has produced beings whose intelligence, patience and insight are able to challenge my own in some dimensions. It has produced beings who know that they are subjects, that they are on television, and that they are studied for their “apeness.” It has produced beings who realize that most humans cannot understand them and that most humans see them as oddities. It has produced beings that know humans discriminate against them. Language has a way of giving those who possess it, the ability to think about the world around them and reflect upon their situation within it — as well as the fairness of it all.

    This has happened in the “Experimental Group.” As a scientist, to date, I have failed these beings, and science, in that the truths which I know about them are not those that I have been able to enable the rest of the world to grasp. This provokes a feeling in discordance between the truth and the perception of the truth. The capacities that I have encountered in these apes have so raced ahead of accepted paradigms to explicate what is happening that I quickly reached the limits of what science found persuasive.

    The simple fact is that most people are not willing to risk getting to know apes without a wire to protect them. Thus the truth of what apes are will continue to be overlooked. I understand why people are skeptical. I entered the field as a skeptic of the findings of ape language. When I began research with Lucy, Booee, Bruno, Lana, Sherman and Austin I had no concept that apes could even begin to take language ability where the EG has taken it. The fact that this has happened is not an accident, for I have sought assiduously the conditions that foster language emergence.

    Yet I did not anticipate the degree to which self-understanding and self-reflection would emerge. As one sees this sort of mind arise in ape, the reason for continuing acquires an entirely new dimension. The effort no longer becomes one of understanding language (that is done), or “convincing others” (that can only be accomplished as more people seek to have real experiences of their own),or about professional recognition (that has been achieved). The effort is about the moral and ethical steps now required to protect these beings and buffer them against a world that cannot yet understand them for what they are. As a scientist, I have an obligation to describe the truth as I encounter it and then to find methods to prove the truth. This takes time because the truths I am encountering are counterintuitive and there is not agreed-upon standard of proof. As humanist, I have a moral obligation to make the lives of these beings as dignified, meaningful and rational as possible.


    Fields, W. M. (2007). Ethnographic Kanzi versus empirical Kanzi: On the distinction between “home” and “laboratory” in the lives of enculturated apes. Rivista di Analisi del Testo, 21.

    Fields, W. M., Segerdahl, P., and Savage-Rumbaugh, E. S. (2007). The Materical Practices of Ape Language Research. In (Eds.) J. Valsiner and A. Rosa, The Cambridge Handbook of Sociocultural Psychology. (pp. 164-202). Cambridge: Cambridge University Press.

    Greenfield, P. M. (2008). Protolanguage in ontogeny and phylogeny: Combining deixis and representation. Interaction Studies, (34-50).

    Lyn, H., Franks, B. and Savage-Rumbaugh, E. S. (2008). Precursors of morality in the use of symbols, “good” and “bad” in two bonobos (Pan paniscus) and a chimpanzee (Pan troglodytes). Language and Communication, 28: 213-224.

    Perry, G. F., Yang, F., Marques-Bonet, T., Murphy, C, Fitzgerald, T., Lee, A., S., Hyland, C., Stone, A. C., Hurles, M. E., Tyler-Smith, C., Eichler, E. E., Carter, N. P., Lee, C.,Redon, R. (2008). Copy number variation and evolution in humans and chimpanzees. Genome Research, 18, 1689-1710.

    Rumbaugh, D. M. (1977). Language learning by a chimpanzee: The LANA project (pp. 287-309). New York: Academic Press.

    Savage-Rumbaugh, E. S. (1994). Hominid evolution: Looking to modern apes for clues. In Hominid Culture in Primate Perspective., D. Quiatt and J. Itani (Eds.). Niwot, CO: University Press of Colorado.

    Savage-Rumbaugh, E. S. (1999). Ape Language: Between a rock and a hard place. In The Origins of Language, (Ed). Barbara King. Sante Fe: School of American Research Press.

    Savage-Rumbaugh, E. S. (1997). Why are we afraid of apes with language? In The Origin and Evolution of Intelligence, (Eds.) A. B. Scheibel and J. W. Schopf. Sudbury MA: Jones and Bartlett.

    Savage-Rumbaugh, E. S., Fields, W. M. (2000). Linguistic, Cultural and Cognitive Capacities of Bonobos (Pan paniscus). Culture and Psychology 6:131-153.

    Savage-Rumbaugh, E. S., and Fields, W. M. (2007). Rules and tools: Beyond anthropomorphism. In The Oldowan: Case Studies into the Earliest Stone Age, ed. N. and Fields Toth, M. Goport IN: Stone Age Institute Press.

    Savage-Rumbaugh, E. S., Fields, W. M. and Spircu, T. (2004). The emergence of knapping and vocal expression embedded in Pan/Homo culture. Biology and Philosophy. 19:541-575.

    Savage-Rumbaugh, E. S., McDonald, K., Sevcik, R. A., Hopkins, W. D. and Rubert, E. (1986). Spontaneous symbol acquisition and communicative use by pygmy chimpanzees (Pan paniscus). Journal of Experimental Psychology: General. 115:211-235.

    Savage-Rumbaugh, E. S., Murphy, J., Sevcik, R. A., Brakke, K. E., Williams, S. L., and Rumbaugh, D. M. (1993). Language Comprehension in Ape and Child. Monographs of the Society for Child Development. 58:1-256

    Savage-Rumbaugh, E. S. and Fields, W. M. (in press-a). The Evolution and Rise of Human Language in Homo Symbolicus and Pan Symbolicus.

    Savage-Rumbaugh, E. S. and Fields, W. M. (in press-b). Human uniqueness: Constructions of ourselves and our sibling species: Pan troglodytes and Pan paniscus.

    Segerdahl, P., W. M. Fields and E. S. Savage-Rumbaugh. (2005). Kanzi’s Primal Language: The Cultural Initiation of Primates into Language. New York, NY: Palgrave Macmillan.

    Thompson, C. R. and Church, R. M. (1980). An explanation of the language of a chimpanzee. Science, V. 208, no. 441, pp. 313-314.

  • This conversation, while ending here, continues on Facebook. Join us there by logging on to your Facebook account and proceeding to our group: On the Human.