Building a language
It is rather obvious that languages and their local varieties have to be learnt. What is not so obvious is how this happens.
In the 1980s, particularly from work by Noam Chomsky (1981) and Hagit Borer (1984), many linguists came to think of language learning in terms of choices. For instance, there are many languages like Italian, Greek and Spanish in which the equivalent of the pronoun I is routinely not pronounced in the subject of a sentence, with the equivalent of “I love you” said without the I. According to whether this happens or not, the language learner has to do the equivalent of throwing a mental switch one way or the other. Children exposed to English on the one side or Italian on the other have to throw the relevant switch opposite ways. The points around which these choices or settings are made are known as ‘parameters’. From work by Nina Hyams (1986), this seems to happen at around two and a quarter.
Languages like English contrast below and bellow with contrasting levels of stress on the two syllables. Most of the languages of Western Europe use stress, but in different ways from English. Chinese languages, on the other hand, contrast different tones on single syllables. A child learning a European-type language has to set a corresponding parameter one way. From work by Paula Fikkert (1994), it seems that this starts to happen around two and a quarter. From work by Yuen Ren Chao (1973), children exposed to a Chinese-type language start to set the same parameter the opposite way at the same age, rather suggesting that there is one parameter here.
If this approach is on the right lines, does it account for all the details like the English pronunciation of H only at the beginnings of syllables and final NG only at the end, and differences between varieties of English like Southern British English “Hosannah in excelsis Deo” with an R after Hosannah, but not in Scottish English?
The notion of parameters was a great advance on previous notions of rules in systems of great complexity. By the profound insight of Catherine Renfrew, it is very revealing to score children’s ability to detect when English sentences have to be complete. But making the parametric model work for all the details is a non-trivial challenge. There are languages like Finnish and Russian in which subject pronouns are sometimes pronounced and sometimes not. To define all the attested variations across languages and dialects rerquires an implausibly large number of parameters.
In a 2022 talk, Ian Roberts proposes that languages are ‘built’ from ‘blocks’. One block which he examples is the property of some languages, of which the North American language of Mokawk is a strong well-known example, to ‘glue’ together the elements of a sentence in such a way that they cannot be unglued in the minds of native speakers. The simplest test is to ask a native speaker: “What was the last word you said?” Answering, the Mohawk speaker often seems obliged to reply with what would seem to an English speaker like a sentence. English does this very little. One rather trivial example occurs in forms like don’t. If a speaker who has just said: “Don’t!” is asked that question, don’t would seem like the only possible, true response. But in Mohawk this gluing is taken much further.
But the gluing in Mohawk is not plausibly by one parameter. Rather it seems that this building block must itself have a number of components.