The Psycho-Biology Of Language

AN INTRODUCTION TO DYNAMIC PHILOLOGY

George Kingsley Zipf

• Publisher: Routledge
• Year: 2013
• eBook ISBN:
  9781136310539
• Topic: Medicine
• Subtopic: Health Care Delivery

This is Volume XXI in a series of twenty-one on the Cognitive Psychology. Orignally published in 1936, this is a study on the introduction to Dynamic Philology.

III

THE FORM AND BEHAVIOR OF PHONEMES

THE purpose of the present chapter in our study is to investigate the relationship between the form of a phoneme and the relative frequency of its occurrence. In the course of the investigation we shall obtain abundant evidence in support of the following conclusions: (1) that there exists an equilibrium between the magnitude or degree of complexity of a phoneme and the relative frequency of its occurrence, in the sense that the magnitude or degree of complexity of a phoneme bears an inverse ^* relationship to the relative frequency of its occurrence; (2) that, by means of a process partly analogous to the process of abbreviatory truncation in words (see page 30 ff.), the phonemic system of a language is constantly striving to maintain this state of equilibrium; (3) that the preservation of this equilibrium is the probable cause of phonetic changes which lead to dialect cleavages resulting in new dialects and, ultimately, new languages.

For greater clarity of exposition the material of this chapter will be presented in two parts which will differ primarily in point of view. In Part I we shall study the various phenomena of the phoneme primarily as they appear in a cross-section of the stream of speech, that is, in a steady state. In Part II we shall study the various phenomena of the phoneme primarily as they appear in the stream of speech itself. From this latter point of view which is possible only because of the findings obtained from the former point of view, we shall gain considerable insight into the dynamics of the phoneme as revealed by its form and behavior.

PART I

EQUILIBRIUM OR STEADY STATE IN A PHONEMIC SYSTEM

1.THE SPEECH-SYMBOL, THE SPEECH- SOUND, AND THE PHONEME

Before embarking upon our investigation of the dynamics of phonemes, it is well to anticipate a possible confusion by defining in advance the three different terms which will be employed in the course of our ensuing investigation: the speech-symbol,¹ the speech-sound,² and the phoneme.³ Since language consists both of speech-sounds and phonemes, and, when written, is represented by speech-symbols, these three terms will refer to practically identical phenomena in the many instances when we view a record of the stream of speech instead of the vocal stream of speech itself. For example, the record of an utterance of the English word sit may be viewed as consisting of three speech-symbols, and of three speech-sounds, and of what we shall later describe as three phonemes: in each case an s, a short i and a t. Yet in many cases of recorded language these three are not the same, as we shall presently see, nor in the case of merely spoken language do speech-sound and phoneme always refer to identical phenomena. In elucidating the significant differences between the three terms whose usage must be unambiguous in all discussions of the pertinent phenomena in language, we shall first discuss cursorily the general problems and methods of recording any linguistic phenomena, whether graphically or otherwise. This discussion of the question of recording will lead automatically to a discussion of the material to be recorded, a discussion which will provide both the reason for our employment of the two terms (speech-sound and phoneme) and the linguistic criteria whereby the use of these two terms may be clearly defined.

a.The Question of Record and Graphic Representation

The task of inventing a set of symbols to represent the observable or inferable phenomena of nature confronts every field of science and is comparatively simple. The true problem involved concerns the application of a set of arbitrarily devised symbols to the phenomena of nature so that the distinctive features may be recorded, transmitted to others, and scrutinized at leisure. The contrivance of a system of symbols to represent linguistic phenomena is aided, though not accomplished, by the use of mechanical apparatus.

Thanks to the invention and perfection of the phonograph, the gross acoustic effects of the stream of speech of an individual can be recorded with a high degree of accuracy. The phonographic instrument records not only features of the speech which are shared in common with all speakers of the language, but even the peculiar nuances arising from a speaker’s local dialect, or from the peculiar construction or use of an individual speaker’s vocal organs, which distinguishes his voice from that of another; further, it records the variations in pitch and amplitude and speed resulting from the speaker’s temperament and mood, or from the exigencies of the situation, or from the requirements of the context. In short, the phonograph records what is spoken as it is spoken; it records what we shall term the speech-sounds of a discourse, and records them accurately and permanently, and is invaluable for field work. Yet for the purposes of Dynamic Philology the record of the phonograph is (for two reasons) inadequate as the final representation of speech.

The first inadequacy of phonographic recording for our purposes is that the record is not in a form immediately available for statistical analysis. Before the record can be analyzed statistically, it must be transcribed into written symbols. Hence the value of phonographic recording over an immediate written notation of the stream of speech is chiefly that the phonographic record of a given sample of the ephemeral stream of speech is permanently available for graphic transcription irrespective of time or place.

The second inadequacy of phonographic recording is that in many respects the record is too detailed in its recordings. That is, in recording all the nuances which occur in the stream of speech, it records many personal features of a speaker which from the point of view of the entire speech- community are unessential if not indeed fortuitous. A record of these fine nuances, though possibly of considerable importance in studying the psychology and anatomy of the individual or his personality, is by no means prerequisite to an understanding of the basic linguistic forms and patterns of linguistic behavior which are common to all speakers in the entire speech-community; in fact, a preliminary study of the common basic fabric of speech, if possible, is a better approach to an understanding of peculiarly individualistic elaborations and distortions of the common basic fabric than a reversal of the procedure. That is, an understanding of the phoneme which is common property leads more quickly to an understanding of the peculiarities of individual speech- sounds than will be the case if the order of procedure is reversed. Yet there are more fundamental differences between speech-sounds and phonemes than the mere fact that the former are more individualistic than the latter.

b.The Phoneme as a Norm Approximated by Speech-Sounds; Variant Forms of a Phoneme

The chief feature of a phoneme is that it represents a norm which the individual speech-sounds may be said to approximate. To objectify this relationship, we might, with considerable appropriateness, use the target as an analogy ⁴: the bull’s eye of the target represents, say, the phoneme k in English, and the shots aimed at the target are the speech- sounds. That is, the actual articulations of k in the stream of speech by speakers of the language are shots aimed at the norm which is the phoneme. The speech-sounds, distributed about the norm of the phoneme, give significance to the phoneme, just as the norm of the phoneme gives meaning to the speech-sounds which approximate it. The one without the other is unthinkable.

However, it is often observable in language that individual phonemes may possess several subsidiary norms instead of merely one norm, and that the speech-sounds approximate one or another of these subsidiary norms according to conditions prevailing in the stream of speech at the time. These subsidiary norms of a given phoneme may be termed the variant forms ⁵ of that phoneme, and the speech-sounds, which constitute the occurrences of the phoneme in the stream of speech, approximate one subsidiary norm or another according to the particular conditions under which the speech-sound is produced. For example, the initial k-sounds as pronounced by English speakers in the English words keel, call, and cool are neither acoustically nor phonetically identical in the subsidiary norms which they approximate. In English the velar k is closed farther forward in the mouth in keel than in call, and farther forward in call than in cool : the k of keel is palatal, the k of cool is velar, and the k of call falls in between. The norms which these three different k-sounds approximate are conditioned in each case entirely by the nature of the following vowel sound. Indeed it is practically impossible for the average English speaker to pronounce the palatal k of keel before the vowel of cool, or the velar k of cool before the vowel of keel, nor will this alternate pronunciation be necessary for the proper pronunciation of any word in English. As a parallel example, the spirant in German, usually written as ch, has two variant forms: (1) the palatal, as pronounced in German ich, and (2) the velar, as in German Buch. These two forms of ch are not identical in position of articulation and each consistently approximates its own norm. But in the case of German ch, the determining factor is the preceding vowel sound and not the succeeding vowel, as was the case above with k in English. The German speech-sound, written ch, when following an a or a back vowel, is automatically and invariably velar in German, and approximates what may be termed the velar norm of the phoneme (e.g. Bach, hoch, Buch); yet when following a front vowel in German, the ch is automatically and invariably palatal, and approximates what may be termed the palatal norm of the German phoneme ch (e.g. sich, Blech).

These two typical examples from English and German illustrate the manner in which specific surrounding sounds may sufficiently influence the articulation of a phoneme as to make it deviate consistently under specific conditions in the direction of one of several variant forms. The variant form of a speech-sound is not of random occurrence in the stream of speech; wherever the environmental conditions are correct for the occurrence of a particular variant form, that variant form appears, and from its appearance we may in turn be certain that the prerequisite environmental conditions are present.

C.The Phoneme as the Smallest Unit of Distinctive Significance

One may wish to inquire into the particular criteria whereby it is determined whether a given form is to be viewed as an independent phoneme or as a variant form...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Table Of Contents
• I. Introduction
• II. The form and behavior of words
• III. The form and behavior of phonemes
• IV. Accent within the word
• V. The sentence: positional and inflectional languages
• VI. The stream of speech and its relation to the totality of behavior
• References
• Index
• Notes