I got a good laugh out of this cartoon on Alex Baker's Cake or Death cartoon file. And it fits in with this post's point: gestures have their limits.
Susan Goldin-Meadow is a hero on this blog because her work is both serious and original. It fills a gap in our understanding. In 2008 she presented a report on gesture that has stayed with me. It made clear that gestures are a natural way of illustrating what is not included in a grammatical structure. For example, a person might say, "The plane ride was very…" and then illustrate the ride by moving the hand horizontally while simultaneously bouncing it up and down. Ever since that presentation I have been of the fixed opinion that gesture has, from speech's beginning, accompanied spoken words. So naturally I was pleased to see that Goldin-Meadow has published a paper titled "Widening the Lens" in the Philosophical Transactions of the Royal Society – B (abstract; paper). It summarizes her work.
The paper also settles a matter I have often wondered about: sign language users also add gestures to their signing. "Like the gestures that accompany speech, the gestures that accompany sign are analogue in form, and thus complement the discrete, segmented categories found in sign (and speech)." [p. 2]
Since apes gesture in the wild, it is often proposed that gesture led to speech, but we need to find a process that might link the two. Apes, after all, don't go on to speak, so why did we? Goldin-Meadow's paper suggests possible processes and their limits.
Gestures among hearing children: Children with normal hearing point before they use words. There is no reason to assume it wasn't always thus, so probably the first time somebody in the human lineage directed a companion's attention to something, it was by pointing. At the same time, there is no reason to assume that the lineage went through some generations of pointing withougt naming. "Gestures pave the way for children's early nouns."
A remarkable observation is that "gesture does not appear to pave the way for early verbs." The gestures for actions come after verbs are used. The story is a little complex: (1) verbs appear before action gestures; (2) once verbs and action gestures are both part of a repertoire, an action gesture for, say, climbing, may appear before the verb climb is spoken.
Gestures are rarely combined. Children are not playing charades. They are trying to communicate. Sentences and phrases are not in their gestural repertoire, but come quickly in speech and signing.
Present behavior does not guarantee historical performance, so the absence of verbs and phrases in modern gesture development does not prove absolutely that the situation was different for Homo erectus. Yet it does put the onus on the gesture-first advocates to demonstrate that the development of gestures in the human lineage used to be different.
Gestures among deaf children: Thanks especially to the study of the emergence of Nicaraguan Sign Language we have a clear idea of the relation between gesturing and signing. Deaf infants naturally make gestures and some of these gestures become standardized so that a deaf child can signal a desire for, say, a cookie, without the need to point to an actual cookie. These standardized gestures are called homesigns, and every family with a deaf child will develop some homesigns that allow minimal communication. Until education programs for deaf children became common, however, most deaf people were condemned to stay at the homesign level. But if deaf children are brought together, over time they will spontaneously transform their homesigns into a language, which will grow richer over the generations.
So for enthusiasts of the gesture-first hypothesis, we have a mechanism for converting standardized gestures into sign language. What this mechanism doesn't explain is why, if language was ever standardized at the signing modality, it switched to an oral modality. The first language-users were not all deaf, so they would be expected to include words from the beginning. Still, the gestures-to-homesigns-to-signing route is there for people who do not want to give up on the gesture-first hypothesis.
The critical point is that whatever modality you favor—speech or signs—the fundamental need in changing homesigns or homegrunts to a language is a sharing community. It is communicating together over generations that makes a language rich enough to serve as the wonder we see today. Archaeologists are wrong to insist that the evidence for language lies in the use of symbols. They should be looking for evidence of communality based on sharing and trust.