Linguists commonly maintain that language is a system for translating the hierarchically organized contents of the mind into linear strings of arbitrary symbols. While this may (or may not) be an appropriate way to characterize speech, it is clearly not an appropriate way to characterize signed languages. The notion that language employs only a single channel has been carried over into the study of signed languages, and this is a misconception that may have been aided by the direct transfer of linguistic methods to the study of signed languages. By maintaining these positions, linguists have boxed themselves into a position from which it becomes extremely difficult to imagine how language (or the biological capacity for language) might have evolved according to the same processes as other biological systems. This book will suggest a different approach to understanding this process of evolution that takes into account the full range of human communicative behavior, by employing a more expansive notion of language. In this book, no strict separation will be made between language and human communication generally. That is, no strict separation will be made between language and gesture. In this view, human communication will be considered all-inclusive.
One could argue that the position taken here is wrongheaded in that the true field of study for linguistics is limited to speech, that what needs to be explained is speech and speech alone. Arguments will be presented here that this position is no longer tenable and needs a complete overhaul. Indeed, each of the assertions of traditional linguistics, presented above, are open to challenge. It is by now well known and generally accepted that when deaf people communicate using signed language, they are using a well-formed human language. However, the importance of this insight beyond the interests of the deaf community is not generally recognized. This simple observation should call attention to the fact that human beings regularly and habitually communicate linguistically using methods other than speech.
Stemming from the notion that languages are closed, formal systems is a corollary—that the best way to represent the data of linguistics is through tree diagrams. These diagrams, familiar from pedigree analysis and then from representations of the relationships among biological species, first appear in philology to represent the historical relationships among various languages. The model is one of parent languages giving birth to daughters and sons that may then be subject to further reproduction with modification (see fig. 2). These diagrams have also been used to represent grammatical relationships among the elements of sentences and have appeared most recently in the syntactic theory of Chomsky and other generative grammarians to represent the hierarchical relationships among the elements of phrases and sentences, in the manner of the data structures of the computing sciences. According to Steven Pinker, these tree structures constitute more than simply an attempt to represent grammatical relationships—they represent a theory about the way in which the brain actually does its linguistic computations. Thus, anyone seriously considering the study of language today is confronted with a virtual forest of trees on which are hung the symbols of linguistics.