American Scientist

Talking Hands: What Sign Language Reveals about the Mind. Margalit Fox. xii + 354 pp. Simon and Schuster, 2007. $27.

The Gestural Origin of Language. David F. Armstrong and Sherman E. Wilcox. x + 151 pp. Oxford University Press, 2007. $39.95.

Throughout the world, and dating back to antiquity, deaf people have communicated with one another by means of sight rather than sound, using their hands and faces. Signed languages are still often regarded as vastly inferior to speech and are perceived as relying on mere mimicry or pantomime to convey meaning. And historically, the deaf have been treated as though they were mentally disabled. Spurred in part by the late, legendary William C. Stokoe of Gallaudet University, most linguists have now come to accept that sign languages have all of the grammatical and expressive sophistication of true language. Not all linguists have seen the light, though—as recently as late 2005, at the end of a talk in which I made reference to sign language, a prominent linguist stood up and informed the audience that sign language was a primitive pantomime invented in the 18th century and had no relevance to the understanding of true language or its evolution. The two books under review, Talking Hands and The Gestural Origin of Language, are powerful correctives to that antediluvian view.

The author of Talking Hands, Margalit Fox, is a journalist with a linguistic background who joined a team of linguists studying the sign language developed by a Bedouin community known as the Al-Sayyid in the Negev desert in Israel. The community is made up of some 3,500 individuals, and because many of them intermarry, some 150 of the people living there have inherited a condition that has left them profoundly deaf. Although the deaf are in the minority, Al-Sayyid Bedouin Sign Language (ABSL) is widely used in the community, along with a spoken dialect of Arabic.

Al-Sayyid is important both because many members of the community are multilingual in both signed and vocal languages (a circumstance that overcomes many of the problems of translation) and because ABSL is a recent invention, now in only its third generation of signers. It has developed without any influence from the spoken Arabic of the region, or even from Israeli Sign Language, which is also widely used. As one example of its independence, ABSL uses the basic sentence structure of subject-object-verb, whereas the other languages of the region have a subject-verb-object structure, as does spoken English. The Al-Sayyid village has therefore provided a rare opportunity to document the spontaneous emergence, from scratch, of a new language. The rapid development of ABSL shows the extraordinary readiness of people to invent grammatical language, although whether this is evidence for an innate "universal grammar" or is simply testimony to a more general human inventiveness is a matter of some contemporary debate.

The main challenge the investigators faced was not so much linguistic as cultural, and half of the chapters in Fox's book describe her experiences, and those of the research team, in crossing the cultural divide. The community is Muslim, and men typically have several wives and large numbers of children. The houses are small and modest, usually with two rooms, cement floors and tin roofs. It has required considerable diplomacy for the team to gain the respect and confidence of the community. Fox vividly describes her linguist companions, with their various quirks of personality, and her experiences in interacting with the locals. Some of this has a rather contrived feel, but that perhaps is the nature of the genre. One does gain a good sense of what life is like in a community that, to Western sensibilities, is very different indeed.

Interwoven with this material are semitechnical chapters on the nature of sign language. Here, Fox achieves an admirable balance between comprehensibility and technical sophistication. I can strongly recommend the book to anyone who wants to gain a better appreciation of how sign language works without getting too immersed in linguistic technicalities. Fox refers not only to the study of ABSL but also to research based on American Sign Language (ASL) and work carried out in Nicaragua, where Nicaraguan Sign Language has emerged within the past few decades.

Fox points out that our primate heritage has endowed us with hands that provide a natural signaling system, one that is in most respects more natural than that provided by the voice. Nonhuman primates have relatively poor voluntary control over vocalization, but an arboreal life has given them excellent and flexible control over the forelimbs. Vocalization is much better adapted to signaling emotional states, such as fear, territorial claims or aggression, whereas the hands and arms are more readily adapted to conveying information about events that are structured in space and time. One may wonder, then, why haven't all humans—not just the deaf—adopted a full-fledged manual system?

Fox provides two anecdotes that may supply part of the answer. In one, signed communication came to a halt as night fell. Sign languages require visibility, and if our forebears had relied on manual and facial gestures, their communication would have been limited to the period of daylight, although the communal bonfire might have permitted some extension. Speech allows us to communicate all night, if necessary, and in situations in which obstacles block visual access to the speaker. In the other anecdote, Fox describes a hair-raising ride in the backseat of a car that veered alarmingly from side to side as the driver and front-seat passenger carried out an animated signed conversation. Another advantage of speech, then, is that it allows simultaneous use of the hands while one is talking. Although sign language may have all of the expressive potential of speech, and perhaps more, it nonetheless suffers from certain physical, rather than linguistic, restrictions.

One possibility that Fox does not mention, but which seems to follow naturally from her analysis, is that language did indeed evolve from manual gestures, but eventually shifted from a manual to a vocal mode. This theme is taken up in The Gestural Origin of Language, by David F. Armstrong and Sherman E. Wilcox. The idea is an old one, going back at least to the 18th-century French philosopher Étienne Bonnot de Condillac, and it was even hinted at by Socrates. It has been revived many times by many authors, including me, but has never been fully accepted. The linguist Robbins Burling, in his recent book The Talking Ape (2005), noted the main impediment: "the switch that would have been needed to move from a visual language to an audible one."

Armstrong and Wilcox offer a solution to this problem. Like Fox, they stress the expressiveness of signed languages as evidence for this, but they note one important difference: Signed languages are heavily iconic, in that many signs bear a visual relation to what they represent. They cite a study of Italian Sign Language showing that 50 percent of the hand shapes and 67 percent of the bodily locations of signs are iconically motivated. This is not to say that signed languages are simply pictorial, or even transparent to a naive viewer. But the iconic element nevertheless flies in the face of what Ferdinand de Saussure called the "arbitrariness of the sign"—the widespread view that an essential property of language is that it is composed of symbols that are arbitrary and bear no physical relation to what they stand for.

Armstrong and Wilcox, building on their earlier work with Stokoe, get around this problem by redefining language itself. In their hands, as it were, language is considered an embodied system whereby bodily gestures become ritualized and conventionalized into an accepted communication system. Given that our ancestors were tree-dwelling primates, our hands are well adapted to create four-dimensional space-time representations of the four-dimensional world. This ability was especially amenable to exploitation once our hominin forebears became bipedal and gained additional freedom of hand movement. With conventionalization, gestures become simplified and may lose their iconic aspect, but they are readily maintained through cultural transmission.

In this view, speech itself is a gestural system, composed of movements of the lips, velum and larynx, and the blade, body and root of the tongue.

• Facebook
• Twitter

In this view, speech itself is a gestural system, composed of movements of the lips, velum and larynx, and the blade, body and root of the tongue. This is consistent with the so-called "motor theory of speech perception" developed at the Haskins Laboratories (a private research institute in New Haven, Connecticut) during the 1960s, which holds that the perception of speech is not so much an acoustic phenomenon as the recovery, through sound, of speech gestures. The arbitrary nature of speech sounds is not a fundamental property of language but is rather the consequence of the medium through which the gestures are expressed. The authors aptly quote the linguist Charles Hockett: "When a representation of some four-dimensional hunk of life has to be compressed into the single dimension of speech, most iconicity is necessarily squeezed out." The concentration on speech may have created a myopic view of what language is really all about.

Once it is understood that speech is gestural, the notion of a switch from manual to vocal language becomes easier to comprehend. Indeed, language may have always involved vocalizations and movements of the face as well as of the hands, and signed languages are as much facial as manual. Conversely, spoken language is characteristically accompanied by manual gestures. Evidence from primate vocalization and hominin fossils indicates, though, that the anatomical and neural changes necessary for the intentional production of articulate sounds took place late in hominin evolution; the ability to sustain autonomous speech may not have developed until Homo sapiens emerged about 200,000 years ago, or it may be an even more recent phenomenon than that.

Armstrong and Wilcox refer to evidence that the FOXP2 gene, known to be involved in vocal articulation, underwent a mutation within the past 100,000 to 200,000 years, and they suggest that this genetic alteration may have been a final, crucial step on the path to spoken language. The advantages that may have led to the selection of vocalization as the dominant mode include a couple that have already been mentioned—the ability to communicate at night or when obstacles intervene, and the freeing of the hands for manufacture and other purposes—as well as the development of pedagogy, lower energy requirements and the fact that acoustic signals command attention more readily than do visual signals.

The view that language is an embodied system is finding increasing support from neurophysiology, and especially from the so-called "mirror system" in the primate brain, which is activated both when such an animal performs an action and when it observes the same action being performed by another individual. This system also links the sounds of actions to their production. In humans, it includes brain areas involved in language. These facts imply that the evolution of language, far from being a "big bang" at the dawn of our own species, developed out of the mirror system and so has deep roots in primate behavior. Nevertheless it remains true that human language has a complexity and expressive power not observed in other species. Armstrong and Wilcox suggest that the critical steps from raw gesture to a gestural language, perhaps somewhat comparable to modern signed languages, probably began more than two million years ago, with the emergence of the genus Homo and the ensuing large increase in brain size. This development may have been a response to the dramatic ecological changes during the Pleistocene, driving an enhanced dependence on cooperation and social communication.

The idea of embodiment effectively removes language from its pedestal as an encapsulated, symbol-manipulating system and returns it to the general provenance of human cognition and of biology. Language, then, is not special. Of course there is still the problem of explaining how gestures came to communicate abstract ideas or how syntax evolved.

Armstrong and Wilcox suggest that we can understand abstract ideas through metaphor, which is grounded in bodily dimensions and movement. For example, the notion of understanding can be represented metaphorically by the action of grasping, and in spoken English the word grasp is often taken to mean "understand." Syntax is increasingly viewed as a natural process of grammaticalization, whereby some signs or words lose their meaning and serve purely functional roles.

This book appears at a time when theories about language and its evolution are in flux, and only time will tell whether the notion of language being embodied will gain general acceptance. There will of course be resistance, given the strong tradition of basing the properties of language on those of speech.

Although The Gestural Origin of Language is not always an easy read—its ideas sometimes become swamped in jargon—it is an important book. The authors, who have added solidity to the gestural theory of how language first evolved, are part of a sea change in the way we view language and indeed ourselves.