It is worth repeating that there is no need for a special concept of ‘protolanguage’. All human languages are full-blown languages. None is inferior in any sense to any other. They each simply use one of three strategies for their grammars, G1, G2, or G3. Thus I find little use for this notion, given the theory of language evolution here.
The question of ‘what evolved’, therefore, eventually gets us back to two opposing views of the nature of language. One is Chomsky’s, which Berkeley philosopher John Searle describes as follows:
The syntactical structures of human languages are the products of innate features of the human mind, and they have no significant connection with communication, though, of course, people do use them for, among other purposes, communication. The essential thing about languages, their defining trait, is their structure. The so-called ‘bee language’, for example, is not a language at all because it doesn’t have the right structure and the fact that bees apparently use it to communicate is irrelevant. If human beings evolved to the point where they used syntactical forms to communicate that are quite unlike the forms we have now and would be beyond our present comprehension, then human beings would no longer have language, but something else.5
Searle concludes, ‘It is important to emphasize how peculiar and eccentric Chomsky’s overall approach to language is.’
A natural reply to this would be that what is one person’s ‘peculiar and eccentric’ is another person’s ‘brilliantly original’. There is nothing wrong per se with swimming against the current. The best work often is eccentric and peculiar. But I want to argue that Chomsky’s view of language evolution is to be questioned not simply because it is original, but because it is wrong. He has continued to double-down on this view for decades. In his recent book on language evolution with MIT professor of Computer Science Robert Berwick, Chomsky presents a theory of language evolution which furthers his sixty-year-old programme of linguistic theorising, the programme that Searle questions above. Chomsky’s view was so novel and shocking in the 1950s that it was initially thought by many to have revolutionised linguistic theory and been the first shot fired in the ‘cognitive revolution’ that some date to a conference at MIT on 11 September 1956.
But Chomsky’s linguistic theory was neither a linguistic revolution nor a cognitive one. In the 1930s Chomsky’s predecessor, and I would say his inspiration, Leonard Bloomfield, along with Chomsky’s PhD thesis supervisor Zellig Harris, developed a theory of language remarkably like Chomsky’s, in the sense that structure rather than meaning was central and communication was considered secondary. And another predecessor of Chomsky’s, Edward Sapir, had since the 1920s argued that psychology (what some today would call cognition) interacted with language structures and meanings in profound ways. In spite of these influences, Chomsky has staked out his claims to originality clearly over the years, reiterating them in his new work on evolution, namely that ‘language’ is a computational system, not a communication system.§§
So for Chomsky there is no language without recursion. But the evidence from evolution and modern languages paints a different picture. According to the evidence, recursion would have begun to appear in language, as we saw earlier, via gestures, prosodies and their contributions to the decomposition of holophrastic utterances.
As speech sounds produced auditory symbols (words and phrases) these symbols would have been used in larger strings of symbols. Gestures and intonation, whether precisely aligned or only perceived to be aligned with specific parts of utterances, would have led to a decomposition of symbols. Other symbols could have been derived from utterances that had little internal structure initially, but were then likewise broken down via gestures, intonation and so on.
The bottom line is that recursion is secondary to communication and that the fundamental human grammar that made possible the first human languages was a G1 grammar.¶¶
Chomsky’s grammar-first theory is disconnected from the data of human evolution and the cultural evidence for the appearance of advanced communication. It ignores Darwinian gradual evolution, having nothing to say about the evolution of icons, symbols, gestures, languages with linear grammars and so on, in favour of a genetic saltation endowing humans with a sudden ability to do recursion. Again, according to this theory, communication is not the principal function of language. While all creatures communicate in one way or another, only humans have anything remotely like language because only humans have structure-dependent rules.##
* This diagram comes from the linguistic theory known as Role and Reference Grammar, one of the only theories of language in existence that concerns itself with the whole utterance, not just its syntax.
† Gestures are also crucial to understanding how duality and compositionality happen.
‡ In Dark Matter of the Mind I offer a sustained discussion of phonology related to Universal Grammar, and I severely criticise the notion that either sonority or phonology is an innate property of human minds.
A commonly used representation of the ‘sonority hierarchy’ is: [a] > [e o] > [i u] > [r] > [l] > [m n ŋ] > [z v ð] > [s f θ] > [b d g] > [p t k].
§ I have written extensively on syllables in Amazonian languages, which are particularly interesting theoretically.
¶ Sign languages do not have phonologies, except in a metaphorical sense, though they do organise gestures in ways reminiscent of sound structures. Fully developed sign languages usually arise when phonologies are unavailable (through deafness or lack of articulatory ability) or when other cultural values render gestures preferable. Since gestures are related to the eyes rather than the ears, their organising principles are different in some ways. Of course, because both phonological and gestural languages are designed by cultures and the minds, subject to similar constraints of computational utility, they will also have features in common, as is often observed in the literature.
# One example of a a nonconcatenative process comes from English, where ‘foot’ is singular and ‘feet’ is plural, with one vowel substituting for another. Another famous set of cases comes from Semitic languages, such as Hebrew, as seen in the example below:
‘he dictated’ (causative active v.) is TB?n ‘hih’tiv’, while ‘it was dictated’ (causative passive v.) is nrqri ‘huh’tav’
** There is some overlap between the account presented here and an independently developed set of proposals by Erkki and Hendrik Luuk: ‘The Evolution of Syntax: Signs, Concatenation and Embedding’ (Cognitive Systems Research 27, 2014: 1–10), a very important article in the literature on the evolution of syntax. Like the account below, Luuk and Luuk argue that syntax develops initially from the concatenation of signs, moving then from mere concatenation to embedding grammars. My differences with their proposals, however, are many. For one thing, they seem to believe, which is common enough, that compositionality depends on syntactic structure, failing to recognise that semantic compositionality is facilitated by syntactic structure, but not dependent on it in all languages. Further, they fail to recognise the cultural context and why therefore modern languages do not need embedding. They do seem to embrace my view that language is a cultural tool.
†† See Marc Hauser, Noam Chomsky and Tecumseh Fitch, ‘The Faculty of Language: What Is It, Who Has It, How Did It Evolve?’ Science 298, 2002: 1569–1579. Though the authors use the term ‘recursion’, they now claim that they do not mean recursion as understood by people doing research outside of Chomsky’s Minimalism, but they actually intend Merge, a special kind of grammatic operation. This has caused tremendous confusion, though, ultimately, the issues do not change and Merge has been falsified in several modern grammars (see my Language: The Cultural Tool, among many others).
‡‡ This is not to say that animals could not use symbols. I just am not aware of any well-supported or widely accepted claims that they can, in the wild. I certainly accept the idea that gorillas and other creatures have been taught to use symbols in the lab, and there are cases of animals, such as Koko the go
rilla, using symbols after instruction.
§§ More technically, language is nothing more nor less than a set of endocentric, binary structures created by a single operation, Merge, and only secondarily used for storytelling, conversation, sociolinguistic interactions and so forth.
¶¶ Recursion simply allows speakers to pack more information into single utterances. So while ‘You commit the crime. You do the time. You should not whine.’ are three separate utterances we can say the same thing in one utterance using recursion: ‘If you commit a crime and do the time, you should not whine.’
## This is circular in the sense that Chomsky takes a feature that only humans are known to have, structure-dependency, and claims that, because this defines language, only humans have language.
10
Talking with the Hands
Small gestures can have a big impact.
Julianna Margulies
JUST AS TACIT CULTURAL KNOWLEDGE shapes grammars, it is also important in each of the latter’s components – words, gestures, phonology, syntax, discourse and conversation. However, for many linguists and anthropologists, gestures are often omitted from discussion, judged too quickly to be secondary accoutrements of speech, a separate, independent facet of human behaviour. But researchers from various theoretical perspectives have shown, to the contrary, that an intimate set of connections exists between hand movements, linguistic structure and cognition, held together by tacit, cultural knowledge. An analysis of the symbiosis between the hands, the mouth and the brain and how these evolved has to round out any theory of language evolution.1
Additionally, some have argued that this all shows something more, namely that the highly language-specific components of gestures are innate. This research, pioneered in the work of Susan Goldin-Meadow, examines the ‘spontaneous emergence’ of hand movements in children who have otherwise no access to linguistic input, as in the deaf children of hearing, non-signing parents. She calls these gestures ‘homesigns’ and the gestural systems she studies might indeed be crucial to our quest to tease apart native vs cultural or a priori vs a posteriori perspectives on the origins of (some) dark matter.
To understand the role of gestures in language, it is important to grasp how they work together with intonation, grammar and meaning. It is possible to get some idea of how these different abilities combine by focusing on gestures and intonation as ‘highlighters’, helping hearers pick out new or important information from old information that the speaker assumes to be shared with the hearer. One way to do this is to examine the evolving research into gestures and human language, from the ancients through the current and very important research of contemporary scientists. Without understanding gestures there is no understanding of grammar, the evolution of language, or the use of language. Gestures are vital for a fuller understanding of language, its origins and its broader role in human culture, communication and cognition.
Language is holistic and multimodal. Whatever a language’s grammar is like, language engages the whole person – intellect, emotions, hands, mouth, tongue, brain. And language likewise requires access to cultural information and unspoken knowledge, as we produce sounds, gestures, pitch patterns, facial expressions, body movements and postures all together as different aspects of language. I want to begin here with an overview of the functions and forms of gestures in the world’s languages, including most likely the language(s) of early Homo species. Gestures can be complex or simple. But they can be learned.
The gestures that accompany all human speech reveal an intersection of culture, individual experience, intentionality and other components of ‘dark matter’, or tacit knowledge. There are two kinds of knowledge of human grammars, as there are of most things: the static and the dynamic. These are quite possibly related to declarative and procedural memories, but they do seem a bit different. Static knowledge is a list of the things we know. Rules for telling stories are static knowledge. Dynamic knowledge, however, is understanding that things change and knowing how to adapt to changes in real time. If static knowledge is knowing the rules for telling a story, dynamic knowledge is telling the story. Gestures are crucial components of our multimodal languages. They are themselves intricate in structure, meaning and use. Contemporary research makes it clear that gesticulations are as analytically challenging and as intricate in design and function as any other part of language. But, to reiterate, these are not simply add-ons to language. There cannot be a language without gestures. Most of these are used unconsciously and employ tacit knowledge. They are shaped by the needs of the language they enhance and the cultures from which they emerge.
Kenneth Pike saw gestures as evidence for the idea that language should be studied in relation to a unified theory of human behaviour:
In a certain party game people start by singing a stanza which begins Under the spreading chestnut tree … Then they repeat the stanza to the same tune but replace the word spreading with a quick gesture in which the arms are extended rapidly outward, leaving a vocal silence during the length of time which would otherwise have been filled by singing. On the next repetition the word spreading gives place to the gesture, as before, and in addition the syllable chest is omitted and the gap is filled with a gesture of thumping the chest. On the succeeding repetition the head is slapped instead of the syllable nut being uttered … Finally, after further repetitions and replacements, there may be left only a few connecting words like the, and a sequence of gestures performed in unison to the original timing of the song.2
Pike concludes from this example that gestures can replace speech. Later researchers, however, have shown that the gestures he refers to are a very limited type out of several kinds that are possible. Language is just a form of behaviour, as gestures also are. Still, Pike’s basic point is valid – language and its components are human behaviour guided by individual psychology and culture, dark matter of the mind.
All human behaviour, including language, is the working out of intentions, what our minds are directed towards. Language is the best tool for communicating those intentions. Communication is a cooperative behaviour. It follows cultural principles of interaction.
Pike raised another question: why don’t people mix gestures or other noises with speech sounds in their grammars? Why is it that only sounds made by the mouth can be used in syllables and speech more generally? Why couldn’t a word like ‘slap’ be [sla#], where [#] would represent the sound of someone slapping their chest? It may sound easy, but really this is not a possible word or syllable in anyone’s language. As a beginning linguistics student, I thought this question was interesting but did not appreciate adequately the degree to which it impinges on the understanding of language.
Gestures aim towards what linguists and philosophers call ‘perlocutionary effects’, the effects that a speaker intends her language to have on a hearer. Speakers use highlighters in order to help the listener use or react to the information in the way the speaker hopes they will.
To more fully illustrate the need for a single theory of culture and language, indeed all human behaviour, one might contemplate a scene like the following. Two men are watching other men move some heavy furniture down the stairs in their apartment building. One man passing on the stairway landing is huffing and puffing and concentrating solely on his heavy load. His wallet is hanging loosely from his back pocket, about to fall out. He clearly wouldn’t notice if someone relieved him of this burden. The first observer looks at the second observer with raised eyebrows, looking at the wallet. The second one sees him and simply shakes his head to indicate ‘No’. What happened here? Is this language? It is a form of communication that is parallel to language. Certainly shared culture and conventions are necessary for this kind of exchange. Just about anything two members of a culture wish to exploit can be used to communicate.
There is a broad popular interest in gestures, though people often fail to recognise how fundamental they are to language. They formed the basis of a 2013 article in the New Yor
k Times by Rachel Donadio titled ‘When Italians Chat, Hands and Fingers Do the Talking’.3 Italians do indeed stand out gesturally, but so do we all. Even in the seventeenth century northern European Protestants disapproved of Italians’ ‘flamboyant’ hand movements. But the first person to study Italians’, or any other language’s, gestures from a modern scientific perspective was David Efron, a student of twentieth-century pioneer in anthropology and linguistics Frans Boas. Efron wrote the earliest modern anthropological linguistic study on cultural differences in gestures more than seventy years ago. He focused on the gestures of recent Italian and Jewish immigrants and later compared those with the gestures of second- and third-generation immigrants.
Efron’s study, Gesture, Race and Culture, was simultaneously a reaction against Nazi views of the racial bases of cognitive processes, a development of a model for recording and discussing gestures and an exploration of the effects of culture on gesture. The core of Efron’s contribution is his description of the gestures of unassimilated southern Italians and east European Jews (‘traditional’ Italians and Jews), recently emigrated to the United States and mainly living in New York City (though some of his subjects also came from the Adirondacks, Saratoga and the Catskills). According to Efron, Italians use gestures to signal and support content. For example, a ‘deep’ valley, a ‘tall’ man, ‘no way’. The Jewish immigrants of Efron’s study, on the other hand, use gestures as logical connectives, that is to indicate changes of scene, logical divisions of a story and so on. These uses of gesture underscore the fact that language is triple-patterned (symbols, structure and highlighters, such as gestures and pitch) and shaped by culture.
Efron wanted to know two things about gestures. First, are there standardised group differences in gestures between Italian and Jewish immigrants? Second, he wanted to understand how gestures change when an immigrant is socially assimilated. He discovered a strong cultural effect. An ‘Americanisation’ of gestures occurred in each group over time. The initially strong differences between Jewish vs Italian immigrants grew less pronounced until they were identical to any other citizen of the United States.
How Language Began Page 26