The descriptivists

During the years at the end of the nineteenth and beginning of the twentieth centuries when Saussure was working out his ideas in Europe, synchronic linguistics was emerging independently, and in a very different style, in America under the leadership of the anthropologist Franz Boas.

I use the term Descriptivist linguistics for the school funded by Boas, for reasons that will be discussed shortly. Since, throughout the twentieth century, the great majority of synchronic linguistics have been Americans, it has often seemed that Descriptivist linguistics was linguistics.

Franz Boas, born in Westphalia, began his academic career as a student of physics and geography, and it was through the latter subject that he came to anthropology.

Boas specialized in the anthropology of North American, and, after a short period teaching in Berlin, he settled in the USA in the late 1880s. Boas’s introduced to it contains what is still a good summary of the descriptivist approach to language.  

The nature of the language dealt with was one of the chief differences between the Boasian and Saussurean traditions. Saussure had seized the attention of the scholarly world by inventing a new way of looking at phenomena which had been so familiar for so long that it seemed impossible for them still to hold any surprise.

Boas and his colleagues, on the other hand, were faced with the severely practical problem of working out what the current structure of various utterly alien language was like.

The Descriptivist tended to think of abstract linguistic theorizing as a means to the end of successful practical description of particular language, rather than thinking of individual languages as sources of data for the construction of a general theory of language.

Saussure arguing that language imposes an arbitrary structuring on the intrinsically unstructured domains of sound and meaning; Boas showed how this phenomenon produces a false appearance of primitiveness in languages which are in fact fully comparable with our own.

What is true of sound systems is just as true of the syntactic and semantic aspects of language. Two points are often claimed to be characteristics of primitive language. On the hand they are said to be vague; thus, many languages fail to distinguish singular from plural. On the other hand, they are claimed to deal only in the concrete and not to tolerate the information of abstract concepts.

Boas furthermore makes the very apposite point that abstract terms are created when philosophers bend a language to their purpose; since philosophy is a minority this is always a somewhat artificial procedure, but it need be no more artificial for language in which no one has yet philosophized than it is for the classical languages of philosophy. Logical terms such as quality, essence, now commonplace in the language of Europe.

Boas must unquestionably take pride of place in any account of the Descriptivist school; he created the tradition which moulded the work of all other members of the school. But the man who is nowadays taken as principal representative of the Descriptivist school, and is read by many more linguists than read Boans today, is Leonard Bloomfield. He was a nephew of a leading American historical linguist, Maurice Bloomfield.

The logical positivists, there were only two basic kinds of meaningful statements: logical propositions such as Either P or not P, and reports of simple sense-data.

Behaviourism is a principle of scientific method: a rule which says that the only things that may be used to confirm or refute a scientific theory are interpersonally observable phenomena, rather than, say, people’s introspections or intuitions.

Linguistics descriptions was reliable insofar as it was based on observation of unstudied utterances by speakers; it was unreliable if the analyst hand resorted to asking speakers questions such as can you say so-and-so in you language?

The folklorist may be interested in Englishmen´s beliefs about English; the linguist must concentrate rather on how Englishmen speak when they are not thinking about their language.

Many behaviourist psychologists, however, confused the methodological issue with matter of a substantive belief.

The breaches of linguistics description called phonology, morphology, and syntax are all concerned with different types of pattering observable in speech data.

Bloomfield, to analyse meaning in a language is to show what stimuli evoke given utterances s responses, and what behavioural responses are evoke by given spoken stimuli. Thus Bloomfield´s conclusions that semantic analysis is impossible were sound, even if his reasoning was defective.

Bloomfield turned mind and imagination into taboo terms, but he would probably nevertheless have approved of some version of that idea once it had been translated into behaviourist vocabulary.

Descriptivist concern was with the production of correct theories about individual languages.

The Descriptivists, then, thought of general linguistics more as a body of techniques of beliefs about the nature of language.

The descriptivist practice, it seems to me, was essentially what the linguistics ought to be. They were confused about some issues and wrong about others, but their errors were of very little consequence by comparison with the errors of their successors. The descriptivist tradition is that represented by the work of the Summer Institute of Linguistics, under the academic leadership of Kenneth Pike.