Language and Revolutionary Consciousness

CHRIS KNIGHT

From the outset, ‘spirit’ is cursed with the ‘burden’ of matter, which appears in this case in the form of agitated layers of air, sounds, in short, of language. Language is as old as consciousness, language is practical consciousness, as it exists for other men, and thus as it really exists for myself as well. Language, like consciousness, only arises from the need, the necessity of intercourse with other men.

(Karl Marx and Friedrich Engels 1845-6/1963: 85-6)

7.1 Chomsky’s Model

Replying to his many critics, Chomsky (1979: 57) once accused them of not understanding science. To do science, Chomsky explained, ‘you must abstract some object of study, you must eliminate those factors which are not pertinent…’ The linguist — according to Chomsky — cannot study humans articulating their thoughts under concrete social conditions. Instead, you must replace reality with an abstract model. To deny this is to reject science altogether.

‘Linguistic theory’, in Chomsky’s (1965: 3) well-known formulation, ‘is primarily concerned with an ideal speaker-listener, in a completely homogenous speech-community, who knows its language perfectly and is unaffected by such grammatically irrelevant conditions as memory limitations, distractions, shifts of attention and interest, and errors (random or characteristic) in applying his knowledge of the language in actual performance: In this deliberately simplified model, children acquire language in an instant (Chomsky 1976: 15). The evolution of language is also instantaneous (1998: 17, cited in Botha 1999: 245). Semantic representations are not socially constructed but innate and pre-existent (1988: 191; 1996: 20-30). Humans speak not for social reasons, but in expressing their genetic nature (1976: 57-69; 1980: 229-30). Speech is the natural, autonomous output of a specialized computational mechanism — the ‘language organ’ — unique to Homo sapiens.

In his capacity as a natural scientist, Chomsky (1976: 186) sees people as ‘natural objects’, their language a ‘part of nature’. Linguistics ‘falls naturally within human biology’ (1976: 123). However, this is not biology as normally understood. Discussing the evolution of speech, Chomsky suggests: ‘The answers may well lie not so much in the theory of natural selection as in molecular biology, in the study of what kinds of physical systems can develop under the conditions of life on earth…’ (1988: 167). Language’s features may be ‘simply emergent physical properties of a brain that reaches a certain level of complexity under the specific conditions of human evolution’ (1991: 50). More recently, Chomsky (1998: 17) has speculated that ‘a mutation took place in the genetic instructions for the brain, which was then reorganized in accord with the laws of physics and chemistry to install a faculty of language’.

For Chomsky, linguistics can aspire to the precision of physics for a simple reason — language itself is a ‘natural object’ (2000: 106-33). As such, it approximates to a ‘perfect system’ — an optimal solution to the problem of relating sound and meaning. Biologists, according to Chomsky, do not expect such perfection, which is a distinctive hallmark of physics. He explains: ‘In the study of the inorganic world, for mysterious reasons, it has been a valuable heuristic to assume that things are very elegant and beautiful’. Chomsky (1996: 30) continues: ‘Recent work suggests that language is surprisingly "perfect" in this sense. ..Insofar as that is true, language seems unlike other objects of the biological world, which are typically a rather messy solution to some class of problems, given the physical constraints and the materials that history and accident have made available’. Language, according to Chomsky, lacks the messiness one would expect of an accumulation of accidents made good by evolutionary tinkering. Characterized by beauty bordering on perfection, it cannot have evolved in the normal biological way.

7.2 The Language Machine

On the eve of the computer-inspired ‘cognitive revolution’, Chomsky’s Syntactic Structures (1957) excited and inspired a new generation of linguists because it chimed in with the spirit of the times. The book treated language not as a mass of empirical facts about locally variable linguistic forms and traditions, but as the output of a mechanical device whose design could be precisely specified. In science, according to Chomsky, less is more. If a theory is sufficiently powerful and simple, it should radically reduce the amount of knowledge needed to understand the relevant facts. Syntactic Structures infuriated established linguists — and delighted as many iconoclasts — because its message was that much of the profession’s previous work had been a waste of time. Why laboriously collect and attempt to make sense of concrete, detailed observations as to how the world’s variegated languages are spoken, if a simplifying short cut is available? In an ice-cool, starkly logical argument that magisterially brushed aside most current linguistic theory, Syntactic Structures evaluated some conceivable ways of constructing the ultimate ‘language machine’ :

Suppose we have a machine that can be in anyone of a finite number of different internal states…the machine begins in the initial state, runs through a sequence of states (producing a word with each transition), and ends in the final state. Then we call the sequence of words that has been produced a ‘sentence’. Each such machine thus defines a certain language; namely the set of sentences that can be produced in this way. (Chomsky 1957: 18)

As his argument unfolds, Chomsky rules out this first, crude design for his envisaged machine — it clearly would not work. Then by a process of elimination, he progressively narrows the range of designs which — on purely theoretical grounds — ought to work. Thrillingly, he opens up the prospect of discovering in effect the ‘philosopher’s stone’ — the design specifications of a device capable of generating consistently grammatical sentences not only in English but, with appropriate modifications, in any language spoken (or capable of being spoken) on earth. This ultimate device, Chomsky reasons, must be the very one which, in real life, resides in the brain of every human being. Language, in this new perspective, is no longer a mass of variegated behavioural manifestations or locally observable regularities. It is the device enabling speaking to occur. The aim of linguistics is to expose this hidden object to view.

Chomsky’s (1996: 31) ultimate dream is to integrate linguistics into an expanded version of physics: ‘The world has many aspects: mechanical, chemical, optical, electrical and so on. Among these are its mental aspects. The thesis is that all should be studied in the same way, whether we areconsidering the motion of the planets, fields of force, structural formulas for complex molecules, or computational properties of the language faculty’. Consistently with this project, he defines language as ‘an individual phenomenon, a system represented in the mind/brain of a particular individual’ (1988: 36), contrasting this with the earlier view of language as ‘a social phenomenon, a shared property of a community’. Saussure (1974 (1915]: 14) wrote of langue: ‘It is the social side of speech, outside the individual who can never create nor modify it by himself; it exists only by virtue of a sort of contract signed by the members of a community’. The problem with such usage, Chomsky (1988: 36-7) complains, is that it ‘involves obscure socio-political and normative factors’. His misgivings are clear. Where social factors are concerned, science must inevitably fail us. There simply cannot be a social science. Those who define language in social terms are therefore doomed to find the whole topic ‘obscure’.

Chomsky denies the relevance of social factors even when considering language acquisition by the human child. The infant’s linguistic capacities, he explains, cannot be taught. Instead, they must be ‘allowed to function in the way in which they are designed to develop’ (1988: 173). After briefly discussing this topic, he concludes: ‘I emphasized biological facts, and I didn’t say anything about historical and social facts. And I am going to say nothing about these elements in language acquisition. The reason is that I think they are relatively unimportant’ (1988: 173). Superficial irrelevancies aside, Chomsky views language acquisition as independent of experience: ‘No one would take seriously a proposal that the human organism learns through experience to have arms rather than wings, or that the basic structure of particular organs results from accidental experience. Rather, it is taken for granted that the physical structure of the organism is genetically determined ….’(1976: 9-10). Human mental structures develop in the same way. ‘Acquisition of language’, concludes Chomsky (1988: 174), ‘is something that happens to you; it’s not something that you do. Learning language is something like undergoing puberty. You don’t learn to do it; you don’t do it because you see other people doing it; you are just designed to do it at a certain time’.

7. 3. A Stimulating and Loving Environment

Almost in the same breath, however, Chomsky makes an admission that appears striking in its humanity and self-evident force — and equally in itslack of fit with his other claims. ‘Now a good system of raising children’, he observes (1988: 173), ‘puts them in a stimulating, loving environment in which their natural capacities will be able to flourish:

This innocent-seeming observation in fact poses a puzzle. If experience is of secondary significance, why should it be important to love the child? Of course, without food and protection, the child may die. And without stimulation, its language organ may not develop. But beyond such basic necessities, what does Chomsky mean by a ‘loving’ environment? His thinking is clarified when he imagines the alternative possibility: ‘Now let’s take a human child which is raised in an orphanage, and let’s suppose the child is given the right medical care and food and has normal experience with the physical world. Nevertheless the child may be very restricted in its abilities. In fact, it may not learn the language properly’ (ibid.). Chomsky does not quite specify the problem with this nameless institution. But he leaves us in little doubt. The child is afforded a physically secure yet non-loving, non- stimulating environment. Consequently, its natural capacities may not flourish. Chomsky here comes close to conceding that asocial factor-namely, insufficient caring interaction or ‘love’ — might prevent an otherwise normal child from acquiring language despite its genetic endowment.

Let us agree, for the sake of argument, that language acquisition indeed depends on a ‘loving, stimulating environment’. The suggestion seems hardly controversial. But what are the features required for an environment to count as ‘loving’? Although he assumes some threshold level of linguistic stimulation, Chomsky has shown little interest in such ‘external’ issues. Developmental psychologists and anthropologists, however, have amassed detailed knowledge concerning precisely the conditions under which a human child — under varying cultural conditions — will optimally internalize its natal language (Ochs and Schieffelin 1984).A related body of research has revealed that even apes may acquire limited symbolic competence if cared for in humanlike ways (Savage-Rumbaugh and Rumbaugh 1993; Tomasello et al. 1993). Overall, the evidence suggests that co-operative mind-reading — the solicitous interest of others — is indeed a critical variable in enabling normal linguistic competence to be acquired.

When happy in its social surroundings, a baby’s playful gestures may communicate details of its cognitive states. When unhappy, it expresses mainly emotional and physiological states (Trevarthen 1979: 347n.). A child pointing and gesturing for an apple within its mother’s reach might naturally expect her to respond helpfully, preferably by providing the apple (Lock 1993: Fig. 12.1). But imagine a hungry mother whose response was to eat the desired item herself. The child would soon abandon cognitive transparency in favour of cries, attempts at physical snatching — or eventual silent withdrawal. Lack of a ‘stimulating, loving environment’ would in this way block the child’s linguistic development.

Admittedly, such mother-infant competition would appear unusual, at least among humans. But what about other species? All mammals nurse their young, and most primates carry them, too. But primate mothers — differing in this respect from Kanzi’s loving human carers (Savage-Rumbaugh and Rumbaugh 1993) — rarely offer foraging assistance after weaning. Macaque mothers in South India take food directly out of their infants’ hands and mouths (Simonds 1965). Goodall (1990: 166) describes a chimpanzee mother who not only displaced her daughter from a termite-fishing nest but also seized the young female’s tool for her own use.

Wild-living primates also appear relatively reluctant to share information of obvious value to their young. How should this be explained? Barbara King (1994) addressed this problem in her book, The Information Continuum. Her aim in this volume was to show that ‘no critical watershed in social information transfer separates primates and humans’ (p. 7). As her research proceeded, however, she encountered an unexpected result. Rarely do wild-living apes use volitional signals to donate foraging-related information to one another. Mothers often appear reluctant to give such information even to their own offspring. King comments: ‘Adults do sometimes donate information to immatures …[but] it is puzzling that such instances are so infrequent. If a mother could help her offspring by donating information to it at relatively little cost to herself, why doesn’t she? Primatologists have no true understanding of this situation’ (p. 6).

Such apparent selfishness does indeed seem anomalous — but only if we accept human levels of trust and sociality as the norm. We naturally expect humans to seek social recognition by learning to speak in relevant ways (Dessalles 1998). But an ape intentionally divulging relevant information in public might lose rather than gain in status, as others took advantage of its honesty. Adult chimpanzees, admittedly, do emit co-operative signals — such as food-calls (Wrangham 1977). But, being irrepressible, these excited sounds are interpreted by listeners as audible evidence of the presence of food, regardless of cognitive intentions. The intention behind a food call would be of little interest to others. Ape facial expressions are potentially more transparent in this respect. By human standards, however, they remain inscrutable. In all primates except humans, the sclera — the tissue surrounding the dark-coloured iris of the eye — is dark rather than light or white. The consequent lack of contrast makes it difficult for others to infer direction of gaze (Kobayashi and Kohshima 2001). Like mobsters wearing sunglasses, mature apes are experts at being poker-faced, displaying no enthusiasm at all for having their minds read too easily (de Waal1982 ) .

A female gorilla was travelling in a party when she noticed a clump of edible vine. She paused as if intending to groom herself. When the others had unsuspectingly moved on, she took advantage of their ignorance — and ate the whole quantity undisturbed (Savage-Rumbaugh and McDonald 1988: 225, citing Fossey). This ape’s intelligence led her to avoid divulging relevant information to others in her group. In pursuing this strategy, she faced minimal social costs. Even had her duplicity been exposed, she might have lost some share of the food — but not status in the eyes of her community. Primate social status is quite straightforwardly consistent with concealing relevant information. Why, then, should an ape mother foster in her offspring qualities such as honesty and transparency? These are human moral values, not those of Machiavellian primate politics (cf. Byrne and Whiten 1988).

Admittedly, apes are not simply Machiavellian as in ‘completely unscrupulous’ — they may display sociable impulses, including many suggestive of human ‘moral’ sensibilities (de Waal 1996). Still, the global political order remains founded on other principles. In all animals, effective parenting entails equipping the young to reach reproductive maturity, preparing them for the competitive challenges they may later face. To a human parent, donating relevant information to young offspring might appear unconditionally adaptive. But imagine a world in which adult success depended on primate-style competition for dominance. Would it then be adaptive to encourage dependency on easy-to-fake information received second-hand? Where group-level moral principles are non-existent, it may be best to leave youngsters to fend for themselves — as ape mothers indeed appear to do.

7.4 Signal Evolution: The Social Background

Human sobs tend automatically to trigger tears; laughter is intrinsically contagious. In such cases, stimulus and response are not cognitively mediated; little intellectual choice is involved. If only to facilitate deception, however, both primates and humans require at least some element of cognitive control over their species-specific ‘gesture/ calls’ (Burling 1993), and such capacities have in many cases evolved. But there is always a paradox here. What is the value to anyone of a patently false smile? To be cognitively controlled is to be suspect — if a sign can be manipulated, it can be faked. Even where control is theoretically possible, therefore, the need to inspire trust will prompt signallers to fall back on more ‘primitive’ performances that are impossible to fake. Where trust cannot be assumed but must be generated via the signalling activity itself, cognitive control over signal-emission must inevitably be selected against (Knight 1998, 1999,2000).

Much more than other primates, humans rely on signals that are patent fakes. ‘Mimesis’ is the term given by Donald (1991) to a category of signals lying somewhere between hard-to-fake primate gesture/ calls and speechlike arbitrary signs. Mimetic culture, in Donald’s terms, consists of emotionally expressive signals that to an extent can be deliberately manipulated — ‘faked’. Professional actors learn how to do this as part of their trade, but all humans can ‘act’ within limits. There is something socially provocative and risky about such signalling, however. The difficulty is that although fake sobs or tears may be switched on and off at will, human evolved responses remain to an extent emotionally governed and irrepressible. To drop one’s guard when others are faking is to risk being seriously manipulated and controlled.

If ‘information’ is defined as ‘that which permits choices to be made’ (Smith 1977: 2), mimesis is not straightforwardly informative. Music, song, dance, and other forms of art tug at the emotions, seducing and in other ways persuading the target audience in manners that may seem cognitively unfair. To those in the know, however, collaborative faking can build confidence between the actors precisely because of the emotional and energetic costs entailed (cf. Power 1998, 2000). Among human hunter-gatherers, fiction-generating emotional performances — particularly initiation rituals — are staged to cement bonds and generate in-group confidence rather than to widen or enhance cognitive choice (Knight 1999).

Young children become immersed in mimesis from an early age. In middle-class Western nuclear family contexts (Ochs and Schieffelin 1984), a mother’s songlike communication with her infant is not fundamentally informative. Instead, the typically exaggerated prosody has an instrumental function — the sounds have acoustic properties well designed to arouse, alert, soothe, or otherwise directly influence the child’s psychology (Fernald 1992). In hunter-gatherer and other traditional cultures, child-rearing practices may differ substantially, yet functions of social bonding and control remain central to pre-verbal communication. Ritual ordeals experienced later in life — including initiation rites — are in a similar sense meaningful yet linguistically uninformative, being aimed more at narrowing personal choices than widening them (Power 2000; Rappaport 1999).

Speech is quite different, both in form and function. To the extent that infant-directed vocalizing becomes linguistic, something new and distinctively human begins to occur. In addition to alerting, soothing, or otherwise manipulating the infant’s responses, the newly comprehensible words and sentences begin to affect cognition, dispassionately communicating aspects of the speaker’s thought. Assuming the listener can make use of this information, its behavioural choices are thereby widened, not narrowed.

Dessalles (1998) has described how such contributions are valued and rewarded within human speech communities. Listeners motivate speakers to be ‘relevant’ by allocating status on that basis. Competition for linguistic status differs subtly yet fundamentally from primate-style competition for dominance. The key distinction is that linguistic status is assigned by the speaker’s coalition or peer group, not extracted by physical, social, or resource-holding dominance at others’ expense. An inescapably moral distinction is entailed here. ‘When competing for relevance and for status’, Dessalles (2000: 78) writes, ‘individuals behave for the good of the group: The most successful speakers are those best able to contribute — directly or indirectly — to in-group co-operative decision-making. Competition of this kind may be described, paradoxically, as ‘competition to co-operate’.

The human child, correspondingly, acquires speech within a supportive and interactive milieu quite unlike the environment to which a young primate must adapt. In nurturing speech in the growing child, caregivers including playmates must display interests stretching beyond motives of one-sided manipulation or control. Far from striving to limit one another’s behavioural choices, successful conversationalists must learn to widen them. In pursuing this goal, they typically exchange roles, rewarding their mutual co-operation, as each responds informatively to the other’s intentionally transparent moves. The infant’s part in all this begins with gurgling, mutual eye-contact, and smiling, develops into babbling, laughter, and imaginative play — and culminates in syntactically complex speech (Bates et al. 1979; Locke 1993). Such social engagement is not merely an external ‘environment: The intersubjectivity involved here is internal to and constitutive of the child’s developing speech (Tomasello 1996).

7.5 Conventional Signalling among Primates

Animal signals are typically ‘exaptations’ (Gould and Vrba 1982) or ‘derived activities’ (Tinbergen 1952). Over evolutionary time, certain aspects of non-communicative behaviour assume a signalling function, becoming correspondingly specialized through a process known as ‘ritualization’. Signal evolution begins when others read some aspect of normal behaviour as significant. If the subject of such surveillance can benefit from having its mind read or its behaviour anticipated, then over evolutionary time natural selection will accentuate the cues, reducing any ambiguity. By definition, such phylogenetic ritualization entails special elaboration and added costs.

The term ‘ontogenetic ritualization’ (Tomasello and Call 1997) has recently been coined to denote the learning-based process through which two or more individuals, following repeated encounters, may reduce a volitional gesture used in social interaction to a mutually understood shorthand. Since the final version in this case is actually less costly and elaborate than the original, I propose (following Burling 2000) to avoid the potentially confusing term ‘ritualization’ here, referring instead simply to ‘conventionalization’.

Very often, the context of primate conventionalization is play (cf. Knight 2000). A juvenile chimp, for example, may slap an adult on the head while jumping onto its body. Suppose the juvenile regularly starts this game by raising one hand as if preparing to slap. On next recognizing this gesture, the adult may anticipate the whole performance, responding accordingly. Conventionalization is accomplished when, anticipating such comprehension, the juvenile raises its hand in a merely tokenistic way, no longer to perform the physical act of slapping but ‘symbolically’ — to invite participation in the game (Tomasello and Call 1997: 299-300).

In addition to co-operative play, the mother-infant nursing context may allow conventional signals to emerge. The chimpanzee infant ‘nursing poke’ (Tomasello et al. 1994) begins as a functional action: the infant pushes aside its mother’s arm to reach her nipple. As mother and infant interact with one another over time, the poke becomes abbreviated, gradually assuming a unique, idiosyncratic form. Before long, the infant is tweaking or stroking its mother in its own special way, the shorthand now mutually understood as a request to suckle. The relation between the modified signal’s form and its meaning is now partly dependent on agreement. Such conventionalization occurs because each infant has an interest in cutting the costs of requesting a feed, while mothers have a corresponding motive to satisfy their offspring while reducing the amount of poking endured. The outcome is a learned, intentional, discrete shorthand, falling outside the normal species-specific repertoire of emotionally expressive, hard-to-fake gesture-calls.

Conventionalization occurs when interests on both sides coincide, allowing tokenistic signalling to replace costly action or indexical display. Among primates, however, the tokens involved are of restricted scope. The nursing poke has no prospect of becoming circulated beyond the specific mother- infant relationship in which it first emerges. Each mother-infant dyad must re-invent the wheel, as it were, arriving independently at its own idiosyncratic version of the poke. Destined to survive only until that particular infant is weaned, each such token, correspondingly, has no prospect of becoming a cultural replicator or meme (cf. Dawkins 1989).

The nursing-poke and the raised-hand play-invitation are instances of volitional primate signalling displaying an element of conventionalization. Other possible examples are the head nod, head shake, wrist flap, and tap/ poke-cognitively expressive gestures, each with its own meaning, used by immature apes in their playful interactions with one another (Blount 1990: 429). During sex, male bonobos may use a pointing gesture to direct females into the desired position (Savage-Rumbaugh et al. 1977). Arguably, this is shorthand for actually pushing the female into place. More subtle pointing through orientation of the head or eyes may play a role in other intimate relations between apes. But such signals are unlikely to become socially circulated or transmitted between generations in conventional form.

Where trust is not extended throughout a coalition or community, associated conventional signalling cannot be extended either. Instead, conventional signals will be confined within isolated, restricted pockets of social space where sufficient trust momentarily prevails. In their capacity as incipient memes, therefore, such signals lack prospects of achieving immortality. They may evolve, but without leaving any descendants. Consequently, neither linguistic nor more general symbolic cultural evolution can get under way.

7.6 Darwinian Versus Mechanistic Explanations of Language Origins

Why do chimpanzees not capitalize on the potential for symbolic communication which they evidently possess? The reasons are social. With apes as with humans, it is political conflict and anxiety — not individual ‘nature’ — that sets limits on the scope for playfulness and creativity in social interaction. Young apes engaged in playful antics are clearly enjoying themselves. Bonobos of all ages use sexual playas a means of bonding, particularly in relationships between females (Parish 1996). But among primates generally, scheming within a Machiavellian political framework (Byrne and Whiten 1988) sets up anxieties, conflicts, and corresponding defensive alliances which permeate and effectively constitute adult sociality at the global level. In the absence of any public framework for enforcing ‘morality’, the onset of sexual maturity confronts individuals with the Darwinian imperative to engage in potentially traumatic conflict with in-group conspecifics. If playfulness becomes less frequent as ape youngsters mature, it is because autonomous, freely creative expressivity is simply not compatible with a situation in which individuals feel sexually or in other ways anxious or threatened (Knight 2000).

If this is accepted, then the transition from primate gesture/call to linguistic sign was driven less by the emergence of novel mechanisms than by the evolution of novel strategies of social co-operation. Few today would wish to deny Chomsky’s main point — humans, unlike apes, come into the world with genetically determined linguistic potential. But early Homo must also have possessed remarkable symbol-using capacities, long before speech as we know it had evolved. Bonobos such as Kanzi (Savage-Rumbaugh and Rumbaugh 1993) have demonstrated their ability to deploy and act upon conventional signs — if their human caregivers can be relied on to reward such behaviour. In the case of evolving humans, there were of course no external caregivers to give the necessary rewards. Rather, speakers and listeners had to place trust in one another in ways that would be maladaptive under conditions of Machiavellian primate politics.

7.7 The Human Symbolic Revolution

Deacon (1997) argues that language evolution was rooted in novel requirements for communication about social contracts. Non-human primates, in this view, are under no pressure to speak. Needing to communicate only about the currently perceptible world, an ape can always draw attention to some symptom or likeness of the intended referent. But a contract, Deacon observes, exists only as an idea shared among those committed to honouring and enforcing it. How, then, might information about it be conveyed? Since physical correlates — indices — in this case do not exist, Deacon argues that the only way to convey information about a contract would have been to establish a suitable symbol. This symbol — he suggests — would be likely to derive from some ritual action involved in cementing the contract (as a wedding ring might come to symbolize ‘marriage’). With recurrent use, such ritual symbols became reduced to shorthands and eventually to words.

Using different lines of reasoning, Maynard Smith and Szathmáry (1995: 279-309) are persuasive in linking the evolution of language with the emergence of social co-operation at the level of the group, viewing this in turn as based on ritually cemented contractual understandings. Drawing on these insights, can we construct a model testable in the light of archaeological, palaeontological, genetic, and ethnographic data? The ancestors of modern humans had long monitored events in the physical and biological worlds, and additionally must have possessed sophisticated ‘mind-reading’ powers. But no amount of mind-reading ability will enable one to see a ‘spirit’. How and why, then, did our ancestors shift to a novel cognitive preoccupation with ‘institutional facts’ (Searle 1996) — that is, with a domain of contractual fictions such as ‘gods and underworlds’ (Chase 1994)? According to one recent Darwinian model (Power and Aiello 1997; Power 1998, 1999), ‘sham menstruation’ (more usually known as ‘female initiation’) was a fiction-generating strategy in which newly fertile females were bonded in coalitionary alliance with their pregnant and nursing mothers, sisters, and other kin. I favour this model because, unlike some others that have been proposed, it sets out from premises in modern Darwinian theory.

As brain size increased dramatically during the later phases of human evolution, the heavily child-burdened human female needed to avoid being made pregnant and then abandoned by her mate. Females who secured increasing levels of continuous investment from males had improved fitness. A popular hypothesis in this context is that females solved their problems through ‘sham oestrus’ (Hill 1982). The evolving human female ceased to restrict sexual activity to the period around ovulation, instead dampening ‘oestrus’ and extending her receptivity beyond the fertile period. By with-holding precise information about her true periods of fertility, the human female kept her mate sexually interested over an increasingly extended period.

Sham menstruation builds on this idea, but gives it an additional twist. Even where females have evolved continuous sexual receptivity, they still risk being exploited by philandering males. Once ovulation signals have been phased out, menstruation becomes one of the few indicators of fertility to remain externally detectable. This may help explain the extraordinary attention focused upon it in virtually all hunter-gatherer and other traditional cultures (Knight 1996). The human female menstruates considerably more copiously than any other primate. Unless countermeasures are taken, this risks divulging information that would allow philanderers to discriminate against pregnant or nursing mothers in favour of females who are visibly cycling — hence potentially available to be impregnated in the near future. Unless cultural factors intervene, we would expect a philanderer to monitor such information, attempting to seduce an imminently fertile (cycling) female at the expense of his already impregnated mate. This threat — according to sham menstruation theory — provides the stimulus in response to which symbolic culture emerges. Recall that by starting to menstruate, a young female divulges her imminent fertility, attracting corresponding male attention. We would predict that pregnant and nursing females in the vicinity might perceive this as a threat. In the counter-strategy envisaged by Power and Aiello (1997), these females take collective action to deal with this threat. It would be difficult to hide the young female, denying her condition or existence. Hiding her would also mean failing to exploit her attractions-her potential value in extracting additional mating effort from males. Instead of hiding her, therefore, the young woman’s real and/or fictional mothers, aunts, and sisters do just the reverse. Publicly advertising her condition, they seek by proximity to identify with her ‘imminent fertility’ and corresponding attractions. Bonding together closely, the young woman’s female kin temporarily bar male sexual access to her and scramble her signal, packaging the released information in a form that philanderers cannot use.

Painting up with ‘blood’, the kinswomen dance and in other ways act in synchrony, asserting themselves as inseparable from their menstruating ‘sister’. Defending against the threat of harassment, they draw jointly on their own and one another’s male kin for support. As ‘brothers’ consequently become involved in physically defending their ‘sisters’, would-be philanderers are deterred from picking and choosing between one female and the next on the basis of biological signals. The fiction is broadcast that everyone just now is imminently fertile. Sexual harassment or violence remains, of course, an option for would-be philanderers. But under the new conditions, sexual rewards can more easily be earned by going away hunting and returning to camp with supplies of meat. In the arrangement that now begins to emerge — known ethnographically as ‘bride-service’ — meat from the hunt is handed to senior figures within the bride’s coalition. They then redistribute it among close and distant kin. To maximize provisions thereby obtained, kin-bonded females should logically maximize joint access to multiple in-married males. This in turn means ensuring that any male who already has a sexual partner is barred from monopolizing access to additional mates. Ideally, each young woman as she comes of age should bring in at least one additional ‘bridegroom’ — preferably a young man with a good reputation as a hunter.

For intrinsic reasons — since outgroup male harassment may at any time be attempted — this strategy must involve coalitionary control over women’s bodies and availability (cf. Knight 1991: 122-53). This in turn depends on one fundamental precondition. When any female is signalling sexual resistance, her kin — both male and female — must support her in this and ensure that her message is understood. Females must schedule their action to maxi- mum effect, appealing to male kin as necessary, synchronizing with sisters and ensuring that signals are salient and unambiguous. As each young female comes of age, she must be decisively and permanently initiated into the same coalition-based strategy.

Against this background, we can derive a prediction about the precise form of the basic cultural replicator or meme (cf. Dawkins 1989). The expected signature of sexual resistance can be inferred by recalling its theoretical antithesis in the form of primate sexual soliciting. We may conduct this reasoning in two steps:

A chimpanzee in oestrus signals her availability with a prominent genital swelling. This is a competitive signal, advertising to males in the vicinity that she is of the right (same) species, the right (different) sex — and that this is the right (fertile) time.
On this basis, sexual resistance should be displayed by collectively signal- ling wrong sex (male), wrong species (animal) and wrong time (menstruating).

Matching these predictions, gender- and species-ambivalent performances are in many traditions intrinsic to women’s ‘rituals of rebellion’ (e.g. Gluckman 1954/1963). Throughout sub-Saharan Africa , females on such occasions paint up with cosmetics — especially brilliant reds — simultaneously playing male and/or animal roles (Power 2000). Cross-culturally, hunter-gatherer initiation ceremonies establish sexual inviolability in comparable ways (Knight et al. 1995; Power and Watts 1997).

Associated with the earliest appearance of anatomically modern Homo sapiens in sub-Saharan Africa — dating to between 100,000 and 130,000 years ago — archaeologists report evidence for the deliberate mining, selection, treatment, and artistic application of red ochre pigments. Arguably, this is evidence for the world’s first ‘art’ (Watts 1999). Drawing on ethnographic data from the same region, the red ochre pigments have been interpreted as connoting blood, fertility, and the supernatural potency so widely attributed to menstruation (Power and Watts 1997; Watts 1999). Examples of more recent local rock art include therianthropes — creatures in which human and animal features are combined (Lewis-Williams and Dowson 1989). Khoisan traditions are of particular significance, since in their case strands of cultural, including ritual, continuity evidently stretch back into the Middle Stone Age (Watts 1999). In the Kalahari, a Ju/’hoansi maiden still celebrates her first menstruation by ceremonially bonding with female kin. While bleeding, she adopts the identity of the Eland Bull. Central to belief about this mythical creature is that he enjoys sexual relations with his multiple ‘wives’ — the maiden’s kinswomen acting the part of eland cows. During the Eland Bull Dance — the ceremony held to celebrate a girl’s first menstruation — such fictional sex is acted out by all concerned (Lewis-Williams 1981). Other hunter-gatherer representations of divinity — such as the Australian Aboriginal Rainbow Snake — illustrate the same logic of gender and species reversal (Knight 1983, 1988).

The ‘human revolution’ became consummated as coalitionary resistance to philandering drove up the costs of ‘selfish’ male strategies to the point where they were no longer affordable. With this source of internal conflict removed, enhanced community-wide trust transformed the context in which communication occurred. We have seen that signals may become conventionalized wherever trusting listeners can be assumed. The establishment of stable, ‘blood’-symbolized kin-coalitions allowed ‘brothers’ and ‘sisters’ to trust one another as never before. Signallers no longer needed to ground each communicative performance in hard-to-fake displays whose intrinsic features inspired trust. Trust, in other words, no longer had to be generated signal by signal — it could be assumed. With this problem removed, even patent fictions could now be valued as evidence from which to reconstruct others’ thoughts. Language consists entirely of fictions of this kind. Humans who had undergone the revolution, then, no longer had to stage a ‘song and dance’ each time they needed to appear persuasive. Costly ritual performance remained necessary, but only because each individual’s initiation into and subsequent commitment to the speech community could be signalled in no other way. Once such commitment had already been displayed, coalition members could cut their costs, replacing indexical display with a repertoire of conventionally agreed shorthands (see Knight 1998, 1999, 2000). Since these low-cost abbreviations — ‘words’ or ‘protowords’ — were tokens in the first instance of group-level contractual phenomena, they could be honest without having to be grounded in anything real. Reality-defying performances upholding community-wide moral contracts are familiar to anthropologists as ‘religion’ (Rappaport 1999). Once humans had established such traditions, they found themselves communicating within a shared moral universe — a socially constructed virtual reality — of their own making.

7.8 Summary and conclusion

Chimpanzees have significant untapped potential for symbolic communication. During infancy, rudimentary symbolic skills may serve purposes in playful and mother-infant contexts. But with the approach of sexual maturity, such skills become increasingly marginal. Only if trusting, playful, and solicitous relationships were central in the struggle for adult reproductive success would mothers enhance their fitness by fostering sophisticated symbolic dependency in their offspring. As things are, chimpanzee mothers refrain from any such inappropriate parenting. Instead, chimpanzee infants are taught that most valuable of all lessons — how to fend for themselves.

Human children learn a different lesson. In their case, the importance of play is not that it prepares for an adult life of competitive conflict. It prepares instead for a life of engagement with symbols. In the human case, it is as if communal pretend-play were not confined to childhood but — in novel forms such as art, ritual, and religion — had become extended into adult life. Creative, playful, and imaginative symbolic performance is central to human social competence, playing a key role in alliance- formation, sexual courtship and reproductive success (Knight 1999; Miller 1999; Power 1999).

Social theorist Pierre Bourdieu (1991: 430) derides Chomsky’s model of a ‘homogenous speech community’, terming it the ‘illusion of linguistic communism’. But this may be too dismissive. Chomsky identifies human linguistic creativity with social and political freedom. He is entitled to argue that where performance is distorted by external factors such as conflict or social inequality, these aberrations should not distort our picture of human speech as such. Anthropologists have confirmed that structures of one-sided political dominance indeed obstruct syntactical creativity, driving speakers to resort to syntactically impoverished, repetitive verbal clichés (Bloch 1975). During social revolutions, by contrast, as authoritarian pretensions are punctured through gossip, humour, and political conspiracy, the full creative possibilities of speech become explored.

The scenario I favour is that speech emerged out of a social revolution (Knight 1991, 1998, 1999, 2000; Knight et al. 1995). Consummated in sub- Saharan Africa some 130,000 years ago (Watts 1999), it inaugurated the most successful and stable system of human social organization to date — the egalitarian hunter-gatherer lifestyle. ‘Communism in living’ (Morgan 1881; Lee 1988) afforded ideal conditions for language’s rapid evolution. In freeing our ancestors from their former competitive anxieties, it liberated human potential. Far from being marginal or external, as Chomsky suggests, distinctively human social strategies in this way gave rise to distinctively human speech.

REFERENCES

BATES, ELIZABETH, BENIGNI, L., BRETHERTON, L, CAMAIONI, L., and VOLTERRA, L. (1979), The Emergence of Symbols: Cognition and Communication in Infancy (New York : Academic Press).

BLOCH, MAURICE (1975), Political Language and Oratory in Traditional Society (London : Academic Press).

BLOUNT, B. G. (1990), ‘Spatial Expression of Social Relationships among Captive Pan paniscus: Ontogenetic and Phylogenetic Implications’, in S. T. Parker and K. R. Gibson (eds.), ‘Language’ and Intelligence in Monkeys and Apes: Comparative Developmental Perspectives’ (Cambridge : Cambridge University Press), 420-32.

BOTHA, R. P. (1999), ‘On Chomsky’s "Fable" of Instantaneous Language Transmission’, Language and Communication, 19: 243-57.

BOURDIEU, PIERRE (1991), Language and Symbolic Power (Cambridge : Polity Press).

BURLING, ROBBINS (1993), ‘Primate Calls, Human Language, and Nonverbal Communication, Current Anthropology, 34:25-53.

– (2000), ‘Comprehension, Production and Conventionalisation in the Origins of Language’, in Knight et al. (2000:27-39).

BYRNE, RICHARD, and WHITEN, Andrew (1988), Machiavellian Intelligence. Social Expertise and the Evolution of Intellect in Monkeys, Apes, and Humans (Oxford : Clarendon Press).

CHASE, P. (1994). ‘On Symbols and the Paleolithic’, Current Anthropology, 35/5: 627-9.

CHOMSKY, NOAM (1957), Syntactic Structures (The Hague : Mouton.

– (1965), Aspects of the Theory of Syntax (Cambridge , Mass. : MIT Press).

– (1976), Reflections on Language (London : Temple Smith ).

– (1979), Language and Responsibility (Hassocks: Harvester).

– (1980), Rules and Representations (Oxford : Basil Blackwell).

– (1988), Language and Problems of Knowledge: The Managua Lectures (Cambridge , Mass. : MIT Press).

– (1991), ‘Linguistics and Cognitive Science: Problems and Mysteries’, in A. Kasher (ed.), The Chomskyan Turn: Generative Linguistics, Philosophy, Mathematics, and Psychology (Oxford : Blackwell), 26-55.

– (1996), Powers and Prospects: Reflections on Human Nature and the Social Order (London : Pluto Press).

– (1998), ‘Language and Mind: Current Thoughts on Ancient Problems’, Parts 1 and 11; lectures presented at Universidad de Brasilia, published in Pesquisa Linguistica, 3, 4. Page references are to the English manuscript.

– (2000), New Horizons in the Study of Language and Mind (Cambridge : Cambridge University Press).

DAWKINS, RICHARD (1989), The Selfish Gene, 2nd edn. (Oxford : Oxford University Press).

157 Chris Knight Language and Revolutionary Consciousness

DEACON, TERRENCE (1997), The Symbolic Species: The Co-evolution of Language and the Human Brain (London : Penguin).

DESSALLES, JEAN-LOUIS (1998), Altruism, Status and the Origin of Relevance’, in Hurford et al. (1998: 130-47).

– (2000), ‘Language and Hominid Politics: in Knight et al. (2000: 62-80).

DONALD, MERLIN (1991), Origins of the Modern Mind. Three Stages in the Evolution of Culture and Cognition (Cambridge , Mass. : Harvard University Press).

DUNBAR , R., KNIGHT, C., and POWER, C. (1999) (eds.), The Evolution of Culture (Edinburgh : Edinburgh University Press).

FERNALD, ANNE (1992), ‘Human Maternal Vocalizations to Infants as Biologically Relevant Signals: an Evolutionary Perspective’, in J. H. Barkow, L. Cosmides, and J. Tooby (eds.), The Adapted Mind: Evolutionary Psychology and the Generation of Culture (New York and Oxford : Oxford University Press ), 391-428.

GLUCKMAN, MAX (1954/1963), ‘Rituals of Rebellion in South-East Africa ’, in M. Gluckman (ed.), Order and Rebellion in Tribal Africa (London : Cohen and West), 110-36.

GOODALL, JANE (1990), Through a Window: My Thirty Years with the Chimpanzees of Gombe (Boston : Houghton Mifflin).

GOULD, STEVEN JAY, and VRBA, E. S.(1982), ‘Exaptation: A Missing Term in the Science of Form’, Paleobiology, 8: 4-15.

HILL, KIM (1982), ‘Hunting and human evolution’, Journal of Human Evolution, 11: 521-44.

HURFORD, J. R., STUDDERT-KENNEDY, M., and KNIGHT, C. (1998) (eds.), Approaches to the Evolution of Language: Social and Cognitive Bases (Cambridge : Cambridge University Press).

KING, BARBARA J. (1994), The Information Continuum. Evolution of Social Information Transfer in Monkeys, Apes, and Hominids (Santa Fe : School of American Research Press).

KNIGHT, CHRIS (1983), ‘Lévi-Strauss and the Dragon: Mythologiques Reconsidered in the Light of an Australian Aboriginal Myth’, Man, NS 18: 21 -50.

– (1988), ‘Menstrual Synchrony and the Australian Rainbow Snake’, in T. Buckley and A. Gottlieb (eds.), Blood Magic: The Anthropology of Menstruation (Berkeley and Los Angeles : University of California Press), 232-55.

– (1991), Blood Relations: Menstruation and the Origins of Culture (New Haven and London : Yale University Press).

– (1996), ‘Menstruation’, in A. Barnard and J. Spencer (eds.), Encyclopedia of Social and Cultural Anthropology (London and New York : Routledge ),363-4.

– (1998), ‘Ritual/Speech Coevolution: A Solution to the Problem of Deception’, in Hurford et al. (1998: 68-91).

– (1999), ‘Sex and Language as Pretend-Play’, in Dunbar et al. (1999: 228-9).

– (2000),’Play as Precursor of Phonology and Syntax’, in Knight (2000: 99-119).

– POWER, CAMILLA, and WATTS , IAN (1995), ‘The Human Symbolic Revolution:

158 Chris Knight Language and Revolutionary Consciousness

A Darwinian Account’, Cambridge Archaeological Journal, 5/1: 75-114.

KNIGHT, CHRIS, STUDDERT-KENNEDY, M., and HURFORD, J. R. (2000) (eds.), The Evolutionary Emergence of Language: Social Function and the Origins of Linguistic Form (Cambridge : Cambridge University Press).

KOBAYASHI, HIROMI, and KOHSHIMA, SHIRO (2001), ‘Unique Morphology of the Human Eye and Its Adaptive Meaning: Comparative Studies on External Morphology of the Primate Eye’, Journal of Human Evolution, 40/5: 419-35.

LEE, RICHARD B. (1988), ‘Reflections on Primitive Communism’, in T. Ingold, D. Riches, and J. Woodburn (eds.), Hunters and Gatherers, i. History, Evolution and Social Change (Chicago: Aldine), 252-68.

LEWIS-WILLIAMS, DAVID (1981), Believing and Seeing: Symbolic Meanings in Southern San Rock Paintings (London : Academic Press).

– and Dowson, T. A. (1989), Images of Power: Understanding Bushman Rock Art (Johannesburg : Southern Book Publishers).

LOCK, ANDREW (1993), ‘Human Language Development and Object Manipulation: Their Relation in Ontogeny and its Possible Relevance for Phylogenetic Questions’, in K. R. Gibson and T. Ingold (eds.), Tools, Language and Cognition in Human Evolution (Cambridge : Cambridge University Press), 279-99.

LOCKE, JOHN L. (1993), The Child’s Path to Spoken Language (Cambridge : Cambridge University Press).

MARX, KARL, and ENGELS, FRIEDRICH (1845-6/1963), The German Ideology; extract trans. in T. Bottomore and M. Rubel (eds.), Karl Marx: Selected Writings in Sociology and Social Philosophy (Harmondsworth: Penguin), 85-7.

MAYNARD SMITH, JOHN, and SZATHMÁRY, EÖRS (1995), The Major Transitions in Evolution (New York : W. H. Freeman).

MILLER, GEOFFREY (1999), ‘Sexual Selection for Cultural Displays’, in Dunbar et al. (1999: 71-91).

MORGAN, LEWIS HENRY (1881), Houses and House-Life of the American Aborigines (Chicago and London : University of Chicago Press).

OCHS, ELINOR, and SCHIEFFELIN, BAMBI B. (1984), ‘Language Acquisition and Socialization: Three Developmental Stories and Their Implications: in R. A. Shweder and R. A. LeVine (eds.), Culture Theory: Essays on Mind, Self and Emotion (Cambridge : Cambridge University Press), 276-320.

PARISH, AMY (1996), ‘Female Relationships in Bonobos (Pan paniscus): Evidence for Bonding, Cooperation, and Female Dominance in a Male-philopatric Species’, Human Nature, 7: 61-96.

POWER, CAMILLA (1998), ‘Old Wives Tales: The Gossip Hypothesis and the Reliability of Cheap Signals: in Hurford et al. (1998: 111-29).

– (1999), ‘Beauty Magic: The Origins of Art’, in Dunbar et al. (1999: 92-112).

– (2000), ‘Secret Language Use at Female Initiation. Bounding Gossiping Communities: in Knight et al. (2000: 81-98).

– and AIELLO, LESLIE C. (1997), ‘Female Proto-Symbolic Strategies’, in L. D. Hager

159 Chris Knight Language and Revolutionary Consciousness

(ed.), Women in Human Evolution (New York and London : Routledge), 153-71.

– and WATTS , IAN (1997), ‘The Woman with the Zebra’s Penis: Gender, Mutability and Performance’, Journal of the Royal Anthropological Institute, NS 3: 537-60.

RAPPAPORT, ROY A. (1999), Ritual and Religion in the Making of Humanity (Cambridge : Cambridge University Press).

SAUSSURE, FERDINAND DE (1974 [1915]), Course in General Linguistics, trans. W. Baskin (London : Fontana/Collins).

SAVAGE-RUMBAUGH, E. SUE, and McDONALD, K. (1988), ‘Deception and Social Manipulation in Symbol-Using Apes’, in R. W. Byrne and A. Whiten (eds.), Machiavellian Intelligence (Oxford : Clarendon Press), 224- 37.

– and RUMBAUGH, D. M. (1993), ‘The Emergence of Language’, in K. Gibson and T. Ingold (eds.), Tools, Language and Cognition in Human Evolution (Cambridge : Cambridge University Press), 86-108.

– WILKERSON, B. J., and BAKEMAN, R. (1977), ‘Spontaneous Gestural Communication among Conspecifics in the Pygmy Chimpanzee (Pan paniscus)’, in G. Bourne (ed.), Progress in Ape Research (New York: Academic Press), 97-116.

SEARLE, JOHN R. (1996), The Construction of Social Reality (London : Penguin).

SIMONDS, P. E. (1965), ‘The Bonnet Macaque in South India ’, in I. DeVore (ed.), Primate Behavior (New York : Holt, Rinehart & Winston), 175-96.

SMITH, W. JOHN (1977), The Behavior of Communicating: An Ethological Approach (Cambridge , Mass. : Harvard University Press) .

TlNBERGEN, NIKOLAAS (1952), ‘"Derived" Activities: Their Causation, Biological Significance, Origin and Emancipation During Evolution’, Quarterly Review of Biology, 17: 1-32.

TOMASELLO, MICHAEL (1996), ‘The Cultural Roots of Language’, in B. M. Velichkovsky and D. M. Rumbaugh (eds.), Communicating Meaning: The Evolution and Development of Language (Mahwah , NJ : Erlbaum), 275-307.

– and CALL, J. (1997), Primate Cognition (New York and Oxford : Oxford University Press).

– SAVAGE-RUMBAUGH, SUSAN, and KRUGER, A.C. (1993), ‘Imitative Learning of Actions on Objects by Children, Chimpanzees, and Enculturated Chimpanzees’, Child Development, 64: 1688-705.

– CALL, J., NAGELL, K., OLGUIN, R., and CARPENTER, M. (1994), ‘The Learning and Use of Gestural Signals by Young Chimpanzees: A Trans-generational Study’, Primates, 35: 137-54.

TREVARTHEN, C. (1979), ‘Communication and Co-operation in Early Infancy: A Description of Primary Intersubjectivity’ in M. Bullowa (ed.), Before Speech (Cambridge : Cambridge University Press), 321-47.

DE WAAL , FRANZ (1982), Chimpanzee Politics. Power and Sex among Apes (London : Unwin).

– (1996), Good Natured: The Origins of Right and Wrong in Humans and Other Animals (Cambridge , Mass:: Harvard University Press).

160 Chris Knight Language and Revolutionary Consciousness

WATTS , IAN (1999), ‘The Origin of Symbolic Culture’, in Dunbar et al. (1999: 113-46).

WRANGHAM, RICHARD W (1977). ‘Feeding Behaviour of Chimpanzees in Gombe National Park , Tanzania ’, in T. H. Clutton-Brock (ed.), Primate Ecology: Studies of Feeding and Ranging Behaviour in Lemurs, Monkeys and Apes (London : Academic Press), 504-38.

From The Transition to Language, ed Alison Wray. 2002.

Oxford University Press, Oxford, UK. ISBN 0 199 25066 9.