The paper which opens this section on the nature of language was written as an article for the Encyclopedia of the Social Sciences. Its scope encompasses the whole field of language, as did Sapir's earlier book Language, although the article is necessarily much more concise and tautly written. The basic contributions of the book, such as the concept of linguistic drift and the outline for a structural classification of languages, are included. But this article also deals with some of the ideas and interests which Sapir had developed in the twelve years between the publication of the book in 1921 and the appearance of the encyclopedia article in 1933. Thus there is use of the term "phoneme" in the discussion of speech sounds, there is consideration of the psychological aspects of language, and there is concern with the matter of an international language.
Two papers which have contributed significantly to the development of the phonemic approach in linguistics follow next, Sound Patterns in Language (1925) and The Psychological Reality of Phonemes (1933). Both stress the importance of a configurational or field approach to understanding the elements of language, and underscore the fallacy of purely mechanistic or atomistic analyses of linguistic phenomena. The first paragraph of the former paper states that the author's purpose is "to indicate that the sounds and sound processes of speech cannot properly be understood in such simple, mechanical terms." And the same paper ends with the note that the discussion is an illustration of the necessity of "getting behind the sense data of any type of expression in order to grasp the intuitively felt and communicated forms which alone give significance to such expression."
The latter, and later, of these two papers similarly begins with the theme that no entity in human experience can be defined adequately as the mechanical sum or product of its physical properties. Sapir's demonstration in this paper utilizes examples from the phonemic systems of Southern Paiute, Sarcee, Nootka, and English. This article was first published in a French translation; the present version is from Sapir's original manuscript.
The next paper, A Study in Phonetic Symbolism (1929), also deals with the sounds of speech, but does so by means of experiment, and indicates the "tendency of symbolisms to constellate in accordance with an unconscious or intuitive logic which is not necessarily based, on experience . . .