Roots of Language: The Forbidden Experiment

Summary

The necessary experiments for an empirical approach to the question of language origin are either lacking or impossible to carry out, forcing linguists to extrapolate from simulated and analogical situations, i.e. the acquisition of language by native children and the development of a new language under special historical circumstances, in order to answer the basic question of whether the present state of languages preserves traces of their initial stages.

Child language was chosen by Givón, who observed that sentence structure develops from a one-word utterance in the first stage to a sequence with three basic constituents, Agent-Goal-Verb. Since the reconstructions of protolanguages show that the ancestral word order was predominantly SOV, the historical data confirm the assumption that ontogeny recapitulates phylogeny. The other approach, that of investigating how new languages develop, was chosen by Bickerton, who used the structure of new-born creole languages to argue that the common basic SVO order corresponds to the earliest stage of human language.

These contradictory views can be reconciled by eliminating the extremes, namely Givón’s recapitulationism and Bickerton’s bioprogram, and by reconsidering their data in the framework of the three-level approach to language. The three basic functions of language (referential, illocutive, and pragmatic) can be ranked in an evolutionary order: at the first stage, when language is dominated by emotions, there are two basic functions, referential and illocutive; it is the stage of trained chimpanzees and of children who begin to speak, using one-clause utterances and free word order. The second stage is dominated by pragmatic function, and word order complies with the needs of planned information and multipropositional discourse: this is the evolutionary leap that characterizes the human species.