‘Madam, how dare you distort, desecrate and defile the English language as you did in your recent Reith lecture…’, ran a letter in my postbag. It was one of many responses to my Reith lectures, some angry, others supportive: ‘I am delighted that you were chosen to give the Reith lectures… those of us who love the language need a champion and now know that we have one!’

Some letters were bafflingly hostile: ‘Well missy, and what are you going to be when you grow up?’ But others were truly heart-warming: ‘Forgive me – a total stranger (unless we met at either L's eightieth or ninetieth birthday parties) taking up your valuable time. It is mainly to thank you for the pleasure you have given two old people with your first Reith lecture.’

Journalists too were divided. ‘Jean Aitchison has hit on precisely the right combination of novelty and controversy’, said Robert Hanks in The Independent. ‘Mind your language, professor’ was the Evening Standard headline, where A. N. Wilson announced that ‘What the professor says is half-baked, half true, and misses the essential point.’

Perhaps I should have been prepared. ‘What Aitchison has embarked upon, she may not realize, is the linguistic equivalent of delivering an anti-war speech in the Pentagon’, said Mark Lawson in The Guardian. For Robert Hanks in The Independent, my first lecture was ‘all perfectly sensible advice, but handing it out to Radio 4 audiences is like trying to set the Inquisition right on theology’.

Dean Farrar, a respected nineteenth-century intellectual, once eavesdropped on some apple-pickers: ‘I once listened for a long time to the conversation of three peasants who were gathering apples among the boughs of an orchard, and as far as I could conjecture, the whole number of words they used did not exceed a hundred’, he guessed. They managed with this low number, he suggested, because ‘the same coarse expletives recurred with a horrible frequency in the place of every single part of speech’. Dean Farrar, like numerous others, apparently grossly underestimated the number of words used and known by native speakers of English.

Words are the topic of this chapter. The language web is the title of this whole book, and the human wordstore with its multitude of links is perhaps the most truly weblike of all aspects of language – even though up till recently, both the size and importance of the internal dictionary, or mental lexicon, have been underestimated.

For centuries, ideas about language origin have frothed up like soap bubbles, then burst into nothing. Over 2,000 years ago, the Egyptian king Psammetichus reportedly gave instructions for two new-born children to be brought up in total isolation by a shepherd in a lonely cottage. The king wanted to find out the words they would first utter, which he assumed would reveal the world's oldest race. Two years later, as the shepherd unlocked the cottage-door one day, the children ran up to him with outstretched hands, pronouncing the word bekos.

Psammetichus discovered with dismay that this was the Phrygian word for ‘bread’, and reluctantly concluded that the Phrygians predated the Egyptians. But according to John Webb, a seventeenth-century writer, Chinese was the original language of humankind.

If “language” were substituted for “organic being” and “natural history” in the excerpt above, Darwin might be expressing the perspective on the origin and evolution of language which we have articulated in this book. We argue that language grows out of a complex of primary human adaptations, including bipedal locomotion, social living, reproduction without an estrous cycle, postnatal epigenesis, postreproductive longevity, and division of labor within the family. In addition, we argue that language grows out of more primitive primate and mammalian neuro-behavioral complexes, including those that govern face-to-face interaction, categorization, and symbolization. Finally, we argue that the key to the transition from primate vocal and visible gesture systems to language (that is, names organized into sentences) is the introduction of iconic, visible gestures at some point in hominid evolution.

Language, like the physical universe, in all likelihood cannot be known fully by an observer in one place at one time. Language theorists and linguists, like physicists, must acknowledge uncertainty. In a masterful summing up of his philosophy of language and much else, Kenneth L. Pike (1993) explains how, as long ago as 1959, he wrote of “Language as Particle, Wave, and Field,” borrowing this three-part label for the principle of complementarity from the work of physicist Niels Bohr.

Complementarity and uncertainty go together: it is only by understanding the limitations on a single point of view or system of mathematics that one can begin to see further. Since Heisenberg, physicists have known that if one can locate a particle precisely, its action (as part of a wave of similar particles) will escape detection, and that if one studies the wave action, individual particles disappear. Field theory states the necessity of looking at both wave and particle.

LANGUAGE FROM A SPECIAL PART OF THE UNIVERSE

If nothing else, language acquisition studies show that language does not develop through an individual's interaction with the natural environment. It emerges only out of social interaction, but social interaction within constrained limits. We would not know what a word means if we had not heard or seen it used by someone else in a context that made the relation between word and meaning reasonably unambiguous. Once language is acquired at a sufficient level, of course, its possessor is able to use language and the aids to thought that language provides to determine the meaning of an unfamiliar word by inference from its context. But the statement still holds. Without the introduction to words and the seminal idea that words symbolize – without the initial acquisition process, which is social – we would have no equipment with which to make linguistic inferences.

It may seem that the condition emphasized above is crucial; the association of a word with meaning makes both conversing and verbal thinking possible; but verbal thinking needs language, and language needs the interaction of at least two human beings.

THE SYSTEM OF LANGUAGE

It should be clear by now that we will argue against this hypothesis by Bickerton, an hypothesis which flows from transformational linguistic theory; although, in Chapter 8, we will discuss a candidate gene that has recently been proposed as a possible basis for this brain reorganization, as well as changes in the vocal tract (Greenhood, 1992). We propose, instead, that there are intermediate stages between non-syntactic communication and fully syntactic language. The essential vehicle of communication that must be understood according to our argument is the iconic, visible gesture. The key problem, according to Bickerton, is getting from nonhierarchical strings of symbols to hierarchical structures, with embedding of phrases. We have already presented alternatives to the rules and representations system that this implies (Chapter 5). Here we will argue that the key to building syntax incrementally is the discovery of relationships within symbols, and that embryo sentences are already inherent in simple visible gestures.

MODULAR VERSUS ASSOCIATIONIST THEORIES OF LANGUAGE

There is a recent version of the nativist theory of language that makes use of the modern concept of modularity, a concept derived from the construction of electronic devices, especially computers. According to modular theories, the brain can be understood as a processing device that contains a number of innately differentiated components, modules, each of which is responsible for a separate subroutine or type of computing activity. Modular theories are in contrast with associationist theories, which assume that the brain is relatively homogeneous and its interconnections are relatively unconstrained but become differentiated primarily through the organism's interactions with the environment. Modular theorists ordinarily cite regularities in human behavior, especially language, as evidence for innately determined mechanisms; while associationists have traditionally cited the great diversity of human languages and cultures as evidence for plasticity.

Precursors of modular as well as associationist theories of the causation of human behavior have very long histories in Western thought. Modular theories have been associated with the idealist tradition in philosophy (largely French), and associationist theories have been linked with the empiricist tradition (largely English). The fundamental issue is the opposition of the notion that all ideas (concepts) must be pre-programmed or built in, as against the observable plasticity and flexibility of the organism.

It is apparent that the question in the title of this chapter is closely related to the question of linguistic modularity, and it is also apparent that in its most general sense the answer to it, as was apparent to Amos Kendall, must be yes, since all children under normal circumstances do, more or less, acquire the language of their community. The principal problem for modularists is to show how an innate program could lead to such a variety of languages, now known to include signed languages. More particularly, how, if language were genetically programmed, is it that an infant from any social group can easily learn the language of any other social group? Is there a general program that operates from global rules, or a series of modules related to specific aspects of language? Or, as associationists would insist, is the brain extremely plastic and languages based entirely upon negotiated conventions within social groups?

One way in which the central role of gesture in language can be understood is to consider gesture as a critical link between our conceptualizing capacities and our linguistic ability. This chapter will begin by exploring two opposing approaches to the study of language with special significance for how we might unify our understanding of spoken and signed languages. Next, we will explore the nature of gesture and how it is involved in the evolution and structure of human cognition. Then, we will demonstrate how visible gesture might link cognition to language.

In Chapter 7, we will develop the idea that visible gestures hold the seed of syntax. Here, we suggest that visible gestures played a pivotal role in the evolution of the cognitive capacities underlying linguistic competence. A critical function of the early conceptual abilities of hominids was to categorize an essentially unlabeled world of objects and events. Once we were able to categorize and conceptualize, visible (primarily manual) gestures – hands acting – were themselves categorized as prototypical objects and actions in the world. These capabilities led to the development of language and, once again, enabled us to marshall these concepts and articulators as linguistic symbols.

LANGUAGE FROM A DIFFERENT PART OF THE BODY

Shortly after a signed language used by deaf people was first studied as a language and not as a speech surrogate or secondary code (Stokoe, 1960), it became obvious that information about non-vocal languages might shed light on the nature of language in general. Questions about the possible origin and development of language, whether it evolved from general cognitive capacities or without need of them, whether it resulted from unique and identical neural structures in every brain – these and other questions might well be reexamined in the light of information about language without speech. Many studies of ASL, however, have been based on the untested assumption that language structure is everywhere the same – the same no matter where in the world, the same now as when language began, the same in every brain, the same whether expressed vocally or visibly.