192 30 7MB
English Pages 101 [108] Year 1974
APPROACHES TO SEMIOTICS PAPERBACK SERIES
edited by THOMAS A. SEBEOK Research Center for The Language Sciences Indiana University
4
IDEOLOGIES OF LINGUISTIC RELATIVITY by
FERRUCCIO ROSSI-LANDI
1973 MOUTON THE HAGUE - PARIS
© Copyright 1973 in The Netherlands. Mouton & Co. N.V., Publishers, The Hague. No part of this book may be translated or reproduced in any form, by print, photoprint, microfilm, or any other means, without written permission from the publishers.
LIBRARY O F CONGRESS CATALOG CARD N U M B E R : 72—94502
TABLE O F CONTENTS
1 Introduction 2 The experience of one's own mother tongue 3 A look at the formation of the thesis of linguistic relativity 4 Particle and field in Hopi 5 An American Indian model of the universe 6 Human activities and holism in Wintu 7 Navajo, a 'chemical language' 8 Formulations of the thesis of linguistic relativity. . . . 9 Excessive interpretations 10 Sphere and mode of the presumed influence of the language 11 Brief discussion of the question of translating 12 Different thoughts and different languages 13 Self-extensive power of language 14 Social totality and extra-linguistic cultural factors . . . 15 Linguistic relativity and alienation Note Bibliography Subject index Index of names
1 4 10 14 18 22 25 28 35 46 53 56 62 66 71 80 81 96 100
1 INTRODUCTION
1
In this essay a first selection of material concerning 'linguistic relativity' is presented. No preliminary acquaintance with the literature of linguistic relativity is presupposed and the treatment tends to be interdisciplinary rather than specifically linguistic. The phrase 'linguistic relativity' indicates a thesis according to which the global structure of every language exercises a differential influence on the thought of the speaker, on the way in which he conceives reality, and the manner in which he behaves in front of it. Whether we are dealing with a thesis proper or with a mere working hypothesis is precisely one of the questions to be answered. The thesis (let us call it that) can be traced back in the first instance to Humboldt; from Humboldt it was taken up and developed, principally by Weisgerber, as long ago as the 20's, and then by Trier and others in German-speaking countries (the Norwegian Alf Sommerfelt is also to be mentioned); in the United States, without direct reference to Humboldt and independently of European scholars, the thesis was rediscovered (or re-invented) by Edward Sapir (18841939) and, especially, Benjamin Lee Whorf (1897-1941). The most brilliant, stimulating, and fruitful formulations of the thesis, which are still passionately debated, are probably those put forward in the 30's by Whorf. 2 1 First published in Italian in Ideologie 4 (1968), pp. 3-69. Although the Italian text has been revised here and there for the sake of this translation, the English-speaking reader is asked to bear in mind that it was originally intended for a somewhat different public. Cf. footnotes 4, page 3 and 8, page 13. — Translated by Torquil Dick Erikson and revised by the author. 2 Language, thought, and reality (New York: John Wiley, and M.I.T. Press, 1956).
2
INTRODUCTION
The thesis of linguistic relativity exercises great fascination. At times it gives rise to the gravest suspicions; more frequently, it is received with enthusiasm or even taken for granted. If we examine the more recent and considered critical literature, which includes contributions by linguists, ethnologists, anthropologists, sociologists, and psychologists, and which is on the whole unfavourable to linguistic relativity, we shall see numerous difficulties usually overlooked by its supporters; but the very demonstration of such difficulties, by common critical opinion, 3 raises issues which are extremely difficult to unravel. The general impression is that the thesis, in spite of its eel-like nature but partly also because of it, retains considerable importance; not for nothing does it periodically come to life again, just when people think they have buried it. The fact is that the thesis of linguistic relativity is not self-contained; on the contrary, it always leads back to more general problems, which do not usually fall within the province of the various specialists and which seem to demand treatment of a more general kind. It lends itself very well to uncovering the analytical richness and, at the same time, the limits of the investigations of the neopositivistic type which today are wont to be carried out within the various separate disciplines. Anticipating here for a moment one of the conclusive ideas of this essay, I should say that we are dealing with a substantially neo-idealistic thesis: so much so that the new methods of neo-positivistic research (far removed as they are from the pedestrian approach of nineteenth-century positivism), with their requirements of rigour and their demand for verifiability, easily dismantle it. Yet no neo-positivistic dismantling does or can bring the issue to an acceptable end: the analysis must be continued on another ground, introducing new interpretative criteria and changing the points of view in a radical manner. And so we meet again, on a new ground, formulations which recall those which the 3
See, for example, Brown, Words and things (1958), p. 231; Jenkins, Style and language (1960), p. 331 (in the comments on the eighth section of the symposium 'Psychological approaches to the problem of style'); Casagrande, 'Language universals in anthropological perspective', in Universals of language (1963), p. 294; Nunnally, 'Individual differences in word usage', in Directions in psycholinguistics (1965), p. 205.
INTRODUCTION
3
idealists had indeed glimpsed, and which even the most up-to-date positivists seem constitutionally incapable of seeing. In this essay I certainly do not claim to be able to succeed in demonstrating all this; but I hope at least to call the readers' attention to these problems.4
4 The information available to Italian readers on this subject is limited to a few remarks to be found in works dedicated to other subjects, and to some cursory reports contained in popular manuals. Among the former we can list Rosiello's Struttura, uso e funzione della lingua (1965), pp. 103-104; Dorfles' Nuovi riti, nuovimiti (1965), pp. 115-123; Eco's La struttura assente (1968), pp. 224 n. and 368-369. Among the latter, Chase's II potere delle parole (1966), pp. 113-122; La comunicazione di massa (1966; the more accurate original title was Explorations in communication, 1960), pp. 119-173, in which some texts by Lee are to be found which can usefully integrate what we say on Wintu in chapter 6. I for one had raised the problem of objectively non-translatable structures in 'De la communication d'une langue au point de vue épistémologique et au point de vue opératif' (1949), and in'Del non-traducibile' (1952); both papers merged later on in my 1961 volume. But apart from these and other brief references, such as those contained in Pagliaro's essays which appeared in 1952 and 1957, almost nothing had been written on the problems of linguistic relativity in Italy before the present enquiry. It is remarkable, for instance, that the names of Whorf and other ethnolinguists do not figure in Terracini's Conflitti di lingue e di cultura (1957), although almost half of this work is dedicated, and with intelligence, to 'Lingue e cultura' (pp. 123-251). Trier is mentioned once in passing for his theory of 'semantic spheres', but the reference is made through Wartburg's Problèmes. Sapir is spoken of with a certain breadth as a sut generis vindicator of the 'eminently subjective character of the individual activity of the speakers' (p. 127); but this is just one of the aspects of language which is grossly overrated in Continental Europe, especially in Italy where Crocean ideas, finally buried by philosophers, still survive in a number of linguists, with the result of making linguistic relativity impossible to understand. Weisgerber is absent from Terracini's index of names; he is mentioned in passing in a note on p. 129, however, but only for one article from 1933. Of the new Italian journals of pure and applied linguistics, Lingua e stile and Strumenti critici, respectively, six and seven issues have appeared (July 1968), but the problems of linguistic relativity have not as yet been approached in them.
2 THE EXPERIENCE O F ONE'S OWN MOTHER T O N G U E
Let us now enlarge upon some considerations which will help us to approach the problems of linguistic relativity from within our common experience as speakers. We all speak: we have in common ideative, expressive, and communicative techniques which make up language or are acquired through it. But we are also divided into groups called linguistic (or 'lingual' which has the merit of deriving expressly from lingua). For each one of these groups there is a mother tongue in which one learns to speak. A language can be viewed as a stock of wealth, or patrimony, which supplies linguistic materials and instruments common to the group. Following the economic distinction between 'variable' and 'constant' parts of total capital, it can be said that a language is a constant capital while its speakers are a variable capital. By joining constant capital, that is, a language, with variable capital, that is, its speakers, we obtain total capital, that is, SPEECH (or, better, LANGUAGE) in-general. The productive processes practised in common by the speakers constitute collective or common speech. Language-in-general is the dialectical sum of a language and common speech. 1 The experience we have of our mother tongue and the use we make of it present the twofold aspect of our personal participation in repeatable situations and the possibility we have of appealing to the repeatability of certain situations as a means of intersubjective 1
This characterization will be taken up again in chapters 13 and 15. What is lacking here is the current Italian distinction between linguaggio (French langage) and lingua (French langue).
THE EXPERIENCE OF ONE'S OWN MOTHER TONGUE
5
checking. Once we have learned it, we continue to share in the benefits of our mother tongue; we can easily go back to it after a period in which we have been speaking another, and we are always in a position to use it to assess whatever is said in it by others, or to check something that we ourselves are striving to say. The element common to both aspects, as has often been noted for experience in general, 2 is repeatability. And yet, in a third sense, the global experience which we have of our mother tongue is precisely something which can NOT be repeated. We cannot have two mother tongues, and by this I mean that one does not learn to speak twice. In this third sense, the patrimony offered us by our mother tongue can neither be replaced nor repeated. Our experience of it lacks precisely that repeatability which is characteristic of other notions of experience. Without entering here into the far from simple problems of bilingualism, 3 it is clear that it must be mentioned as a possible partial exception. One could possibly say, of those rare individuals who are authentically and perfectly bilingual, that they have repeated the experience of learning a mother tongue. On closer examination, however, one finds that they too have learned only one mother tongue; according to the situation, the mother tongue is the first language learned, or alternatively that which was learned second if it was learned better. And if a child has learned both languages during the same period and constantly mingled together, 4 we could perhaps conclude that he has learned what for him is only 2
Cf. Abbagnano, Dizionario difilosofia (1961), entry 'Esperienza'. After a glance, e.g., at what Martinet says about it in Elementi di linguistica generate (1966), 5.27 and 5.28 (pp. 161-164), one can pass to Weinreich, Languages in contact (1953), Devoto, Profilo distoria linguistica italiana (1953), chapters II and III, and Fishman, 'Who speaks what language to whom and when', La linguistique (1965). Concerning the situation of bilingual writers see the contributions of Kushner and Markiewicz in the Actes du IVe Congrès de l'Association internationale de littérature comparée (1964); also Reimen, 'Esquisse d'une situation plurilingue, le Luxembourg', La linguistique (1965). 3
4
In saying this I limit myself here to a hypothesis, for the discussion of which one may see Tabouret-Keller, 'L'acquisition du langage parlé chez un petit enfant en milieu bilingue', in Problèmes de psycho-linguistique (1963), and the more systematic research of MacNamara, Bilingualism and primary education (1966) and of Stern, Foreign languages in primary education (1967).
6
THE EXPERIENCE OF ONE'S OWN MOTHER TONGUE
ONE mother tongue, but made up of TWO languages usually treated as separate in descriptive linguistics. For the child in question this distinction does not exist, or at least will emerge only later. He has learned to speak THROUGH, BY MEANS OF, two distinct tongues, which together, or in alternation, FUNCTION AS his mother tongue. The presence of these extremely particular cases — whatever their theoretical interest for linguistics — does not prevent us from concentrating on the common and universal case of one single mother tongue, which is both the object and the instrument of a completely unrepeatable learning-process. It is to this that Leo Weisgerber constantly refers in his forty years of analyzing German as a mother tongue. It should be noted that the non-repeatability of every speaker's mother tongue is a factor which conditions the very repetitions he makes IN it and WITH it. This is true both every time we experience again the situations which make up our mother tongue, and every time we appeal to any one of its aspects (meaning of a word, legitimacy of a syntagma, and so forth) in order to exercise any sort of control over the communicative process. In other words, to use a formula, REPETITIONS OCCUR WITHIN THE SPHERE OF NON-REPEATABILITY. In English, for example, we experience the use of the present simple as distinct from the present continuous, and their relationships; the position of adjectives almost always before the noun but very occasionally after it too; the construction of a sentence as subject-verb-complement and the difference between it and an incomplete sentence (phrase, locution, and the like); the semantic fields of adjectives of colour, or of adjectives indicating various degrees and kinds of happiness, and the presence of verbs corresponding to some of them; and so on. There is no particular importance in any such list of examples; they are merely random, and may be common to other languages or not. Any part of the English language, any instance of speech in English, will in fact illustrate the point in question to English speakers. But in any case, all these are aspects of the English language, moments of speech as it works among English speakers. That is, they are aspects of a highly structured and complex social fact, whose fundamental and con-
THE EXPERIENCE OF ONE'S OWN MOTHER TONGUE
7
stituent elements are learned in childhood, a piece of reality which changes more or less rapidly in time and space, and to whose changes, to a certain extent, we adapt ourselves in the course of our lives. It is this social reality, lived in its totality in correspondence with that given piece of history to which we belong from birth to death, that constitutes for each one of us a non-repeatable experience. The repeatable aspects of language, on which we continually fall back, are in all cases aspects of our mother tongue and not of any other tongue; moreover, they are aspects of that tongue at a given moment in its development. The dialectic of repeatability IN, and non-repeatability OF, one's mother tongue is of fundamental importance for the thesis of linguistic relativity — it is indeed one of the possible ways of phrasing it. The non-repeatable experience which we have of, say, English is opposed to that which the Chinese have of Chinese, the Russians of Russian, and so on. There are, of course, further differences among groups of individuals or among single individuals in the non-repeatable experience of their mother tongue; but such further differences are disregarded here for the sake of those common experiences which are inter-subjectively repeatable. A language is a super-individual social reality; in spite of differences among speakers, all individuals can refer to it, and are, as it were, 'spoken by it'. It is with this super-individual reality, and not with individual differences, that we are here concerned (as will be better explained at the end of chapter 8). At this point we can already ask ourselves whether the experience which various lingual groups have of different mother tongues might not exercise an influence on their way of using language in general and, furthermore, on their way of thinking and of relating to the world. Can we say that the fact of speaking English, that is of having learned to speak in English rather than, say, in Italian, gives rise to any noteworthy variations in ideative, expressive, and communicative techniques? And even if in the case of two relatively close and similar languages such as English and Italian such variations were not to be found, or were scarcely noticeable, then what should we say in the case of two languages as far apart and as differ-
8
THE EXPERIENCE OF ONE'S OWN MOTHER TONGUE
ent as, for example, any inflected Indo-European language on the one hand and a language such as Chinese or Navajo or Aranda on the other? The working of the human mind is at the same time individual and social; it is itself a 'working by signs' 5 and can only be fully approached through language. This means that the individual language acting as a mother tongue must be taken into account when studying the human mind. One can ask oneself whether the differences existing among languages have an influence on the ways in which people's experience of the world is structured. Are, for example, some kinds of non-linguistic human behaviour to be considered as the effect of a cause, or at least of past conditioning, the said cause or conditioning being the language spoken? Or vice versa? Are there philosophical, ideological, or other problems which, when expressed and discussed in two different languages, end up by differing to such an extent that one can no longer call them the same problems? All these questions lend themselves to over-hasty answers in both possible directions. Such answers are in fact wide-spread and indeed predominant in the literature on the subject. A preconceived notion of the unitary character of mankind, or only a unilateral consideration of the biological constants of human life, for example, can easily lead to a denial of the differential influence of languages on thought. On the other hand, fear of giving too much importance to the above factors, or understandable respect for the inexhaustible diversities in space and time which are studied in the spheres of 5 Note the order of the terms: instead of saying that the individual and social sign-operating is a projection or externalization of the human mind, it is the human mind which comes to be explained in the terms of that operation. Here the doctrine of consciousness as a social product is made more precise by saying that it is a sign-product. This specification is due particularly to George Herbert Mead. A modern discussion, both of the connected thesis that intelligence is not the cause but the effect of languages, and also of the very interesting thesis that the acquisition of language is to a considerable extent independent of intelligence, can be found in Lenneberg, 'A biological perspective of language', in New directions in the study of language (1966), edited by Lenneberg. A wider version of Lenneberg's views is given in Biological foundations of language (1967).
THE EXPERIENCE OF ONE'S OWN MOTHER TONGUE
9
history, sociology, ethnology, and linguistics can lead to an overemphasis of the said influence, and in certain cases even to an identification of the use of ideative, expressive, and communicative techniques with the use of a particular, historically determined language. In the first case, thought is seen as independent of the languages which it must nevertheless use to exist; whereas in the second, it is seen as existing only inside languages and therefore within this or that individual language. In both cases we are immediately confronted with insurmountable difficulties. What could thought possibly be, detached from the language which it must in any case use? And how on earth could one put one's finger on it, if not by using a language? But if, on the other hand, thought were to be found always, only, and wholly inside a given language, then international thought could not strictly speaking exist. The development of scientific language and the very fact of translating would then be inexplicable mysteries. The fact is that when the alternative is posed in such generic terms, one is condemned to remain in the domain of the generic. To get at the substance of the matter and to begin to clarify the situation, one should examine a clearly identified thesis, the examples brought forward to support it, the criticisms which it has received, and the verifications to which it has been submitted. In the following pages we cannot do all this fully, but we can at least sketch it out.
3
A LOOK AT THE FORMATION OF THE THESIS OF LINGUISTIC RELATIVITY
If any one language has been the protagonist of linguistic relativity as developed in the twentieth century, this is probably Hopi; to this language Whorf dedicated most of his research and drew from it the greater part of his examples. Indeed, undertaking the study of Hopi (at the suggestion of Sapir) was for Whorf a decisive turning-point.1 Hopi is the dialect of a PUEBLO in Arizona (a PUEBLO is a small ethnological residential unit); it belongs to the branch known as 'SHOSHONE' of the Uto-Aztec 'large family' of indigenous languages of North America. Distantly related to Aztec, it is however much more complex, subtle, and to us, strange, especially in respect to grammar. 2 Whorf also drew material to support his thesis from other Amerindian languages: NITINAT, a language of the Nootka group belonging to the Wakash family, spoken by tribes settled on the Pacific coast; and NAVAJO (a name which we give in the titles and quotes also in the Anglicized form 'Navaho'), an Athapaska language similar to Apache, belonging to the Nadenè 'large family', now spoken by nearly 100,000 people in the biggest Indian reservation in America, on the Colorado plateau (between Utah, Arizona, and New Mexico). It should be borne in mind that Amerindian languages — whether now extinct or on the verge of extinction, or adulterated, or spoken 1 The same can be said of Sommerfelt's study of Aranda. On the peculiar character of Hopi culture in general, see Eggan, 'Le rêve chez les indiens hopis' in Le rêve et le sociétés humaines (1967). 2 For these and other descriptive elements I use as a first basis the manual of Meillet and Cohen, Les langues du monde (the second edition of 1952, awaiting publication of the third). For other references see below.
FORMATION OF THE THESIS OF LINGUISTIC RELATIVITY
11
by only a few thousand or a few hundred people (Navajo is an exception here) — often bear no relation to each other. 3 Specialists consider their work to be lightened when, passing from the study of one of these languages to another, they find differences as 'small' as, say, those between Russian and English, or Portuguese and Arabic. Hopi and Navajo, for instance, show not even the remotest connection. 4 This 'archaic particularism' is itself a problem apart, especially if one considers that in the Americas there are, or were until a century ago, more than a thousand extremely different languages. It seems to justify the notion of a Standard Average European language (hence the initials SAE used hereafter) which was counterposed by Whorf to the notion of all those languages representing this 'particularism'. Whorf refers to English, German, and French. Differences like those between Romance and Germanic languages are not even taken into account: from the point of view of the basic structure of a language, such differences are in fact very modest in comparison with those between members of hundreds of pairs of Amerindian languages or between any Amerindian and European language. What immediately strikes one in virtually all the Amerindian languages (as indeed in vast numbers of other 'primitive' languages) is that they are not any simpler, but on the contrary more complex than the SAE languages — as was once noticed by Hegel himself, duly influenced by a reading of Humboldt's essay on the dual (1827). 5 Although he also dealt with various other languages, it was Whorf's long and ardent study of Hopi which caused the following idea to mature in his mind. Not only the basic lexication, as he had already described it with regard to Aztec and Maya, 6 but also, and 3 A presentation in Pinnow, Die nordamerikanischen Indianersprachen (1964); the singularity of the fact that the first comprehensive manual o n AmericanIndian languages is not the work of a North-American scholar has been noticed by C. F. and F. M. Voegelin in their important review of Pinnow's book, in Word (1967). 4 Kluckhohn and Leighton, The Navaho (1946), p. 337, n. 1 in the Introduction. 5 Hegel, Enzyklopädie (1830), § 4 5 9 (S. 370, p. 421). 6 Whorf, 'Aztec linguistics' (1928); cf. Carroll, 'Introduction' to Language, thought, and reality (1956). See also Sommerfelt (1938).
12
FORMATION OF THE THESIS OF LINGUISTIC RELATIVITY
even more so, the grammatical structure can indicate and convey a different mode of feeling and conceiving. Thus the emphasis moves from the content of thought to the modalities of its procedure. In a résumé such as this it is not possible to bring to life the subtlety, the penetration, and the elegance of the analyses made by Whorf and by many other linguistic anthropologists, from J. W. Powell and Franz Boas to Dorothy Lee and Harry Hoijer, to name but a few. Working in the field, inside linguistic communities which almost always had no writing and were in many cases doomed to rapid extinction, they were spurred on by the feeling that they could perhaps save something extremely refined which humanity had produced over the millennia and which was about to disappear forever; and armed with immense patience, these scholars reconstructed little by little, piece by piece, the language and the vision of the world of those communities; in this way they finally reached the stage of looking back, as it were, at their own mother-tongue from the completely new viewpoint of the language which they had studied and acquired. 7 These analyses constitute the basis for the thesis of linguistic relativity, which without them (or with the sole aid of hasty summaries) could even take on from the start the appearance of a pseudo-scientific thesis in which there is an attempt at expressing something vaguely glimpsed or perhaps only desired. But the analyses are real because they were carried out on languages actually spoken by independent human groups; and, in general, it can be said that they were of a high technical level insofar as they did take into account the recent development of linguistics, semiotics, anthropology, and allied disciplines. I say this to avoid the suspicion that the contentions of linguistic relativity derive from the mere relativity of cultures, or from some simpleminded deduction such as (1) languages ARE different from each other, and so (2) it follows that the difference must also obtain for thought, world-view, mentality, and social behaviour. This being so, the problem of exposition for every writer dealing 7 See on this point Whorf himself, 'Some verbal categories of Hopi' (1938), in Language, thought, and reality, p. 112. All quotes from Whorf are from that volume.
FORMATION OF THE THESIS OF LINGUISTIC RELATIVITY
13
with linguistic relativity is that of avoiding, as far as possible, that generic character which impugns much of the current writing on the subject, without, on the other hand, falling into the separate technicalities of a merely linguistic or ethnological or anthropological treatment. Whorf has left us various analyses of particular sectors and some global interpretations of Hopi. We find the same sort of material elaborated by others for various other languages, such as — confining ourselves here to the Amerindian field — Navajo, Wintu, Shawnee, and Eskimo. In this essay I shall have to content myself with a brief discussion of one particular analysis and one global interpretation of Hopi made by Whorf; I shall then present some considerations by Dorothy Lee on Wintu, and some data on the 'chemical type of structure' of Navajo. 8
8 In all these cases, throughout chapters 4 to 7, the Italian text was more a carefully worded periphrasis of English statements than a collection of actual quotes. This was due to the difficulties of presenting the ideas of linguistic relativity in a language different from the one in which it was originally formulated, which involved an additional dimension of linguistic relativity itself. At this stage the main problem was how to make the problems of linguistic relativity palatable to an Italian audience. In the process of translating my own Italian text back into English, however, the original statements showed a legitimate tendency to disentangle themselves from the periphrastic web and to reassume the full status of quotes.This should account for the high number of sentences and phrases the reader will find within inverted commas, and also for all the cases in which some strings of words, actually to be found in the texts by Whorf and others, are handed down to the reader WITHOUT inverted commas. That they belong to the texts summarized should be obvious from the context.
4 PARTICLE A N D FIELD IN HOPI
An excellent example of a particular analysis of a sector of the Hopi language is the article 'The punctual and segmentative aspects of verbs in Hopi' (1935). Let us summarize it in Whorf's own terms. The Hopi verb has NINE VOICES, which can be explained as 'categories considered from the point of view of the role of the subject' (intransitive, transitive, reflexive, passive, semipassive, resultative, extended passive, possessive, and cessative); NINE ASPECTS, which can be explained as 'categories considered from the point of view of the type of action described by the verb' (punctual, durative, segmentative, punctual-segmentative, inceptive, progressional, spatial, projective, and continuative); THREE TENSES or rather WAYS OF CONVEYING MEANING, that is 'categories which distinguish different kinds of information' (113) and not properly referring to time and duration (factual or reportive, which is understood by us as a present-past; expective, which is understood by us as a future or even as a 'future in the past'; and generalized or usitative, also called nomic, which affirms a general type of truth, independent from any given situation [in the interpretation of 'tenses' as 'ways of conveying meaning' I am also taking into account Whorf's subsequent article, 'Some verbal categories of Hopi']); and finally, SEVEN CONJUGATIONS. Without counting these last, or indeed the numbers and persons, a Hopi verb can therefore IN THEORY have 343 different forms. It may be noted that among the nine aspects of the verb there are no perfective and imperfective aspects; Hopi does not formalize the contrast between completion and incompletion of actions; but its verbal aspects formalize instead different
PARTICLE A N D FIELD IN HOPI
15
varieties of the contrast between point-locus and extent-locus (forming a field or succession) of phenomena (cf. p. 51). In the article in question Whorf deals only with the punctual and segmentative aspects of the verbs of the first conjugation which is the richest and most creative. The simple expression is given by the third person singular of the intransitive voice, punctual aspect, and factual or reportive tense; it is represented by a root of the form CVCV (C = consonant, V = vowel). The segmentative aspect is formed by final reduplication of this root plus the durative suffix -ta. The change in meaning thus produced consists in this: 'the phenomenon denoted by the root, shown in the punctual aspect as manifested about a point, becomes manifested as a series of repeated interconnected segments of one large phenomenon of a stretchedout segmental character' (52). For example, if the phenomenon requires a rigid or semi-rigid substance for its field of manifestation, we are presented with a static tableau disposed in space, and we have the following differences between punctual and segmentative: ha'ri — it is bent in a rounded angle; harfrita = it lies in a meandering line, making successive rounded angles.
If, however, the phenomenon denoted by the root requires a mobile or nonrigid substance for its field of manifestation (such as a liquid or a swarm of particles), the punctual will denote a single pulse or movement, while the segmentative will refer to the entire train or field of pulsations, both as extending in space and continuing in time: wcila = it (e.g. a liquid) makes one wave, gives a slosh; wala'lata — it is tossing in waves, it is kicking up a sea.
If again the phenomenon be one resulting from the type of force known as torque, it will require a body with a certain degree of rigidity but at the same time capable of certain degrees of motion relative to other bodies (53). Here the punctual will indicate a single oscillation or a partial rotation; the segmentative, a train of oscillations or a continued rotation, in other words a vibratory or rotatory motion:
16
PARTICLE AND FIELD IN HOPI
ro'ya — makes a turn or twist; roycCyata — is rotating. Another class of phenomena is that in which the punctual indicates a shock, a blow, or other sudden disturbance of a momentary nature; and the segmentative, a pulsative phenomenon such as a rapid succession of shocks (54): tfrt = he gives a sudden start; tlrVrtta = he is quivering, trembling. Or, for rhythmical movements of the body and limbs: wVki = he takes a step without moving from place; wikVkita = he is doing steps, or dancing in one place. In yet other cases, the phenomenon may be one of disturbance at a point in a MEDIUM which our science would define as gaseous or etheric. A medium of this kind gives little or no evidence of either motion or extension in space. The segmentative here indicates only pulsation IN TIME: rVpi — it gives a flash; rlpVpita = it is sparkling. There is finally a class of events to which the segmentative, which concerns only the world of what we call 'external' observation, is NOT APPLIED in Hopi: it is the class of events which we call 'mental', 'emotional', or 'psychological' (55). All this, Whorf maintains, illustrates the way in which language organizes experience. We have seen how all sorts of vibratile phenomena are classified by being referred to elementary types of deformation processes. Hopi terminology is so rich, precise, and capable of systematic extensions that, if one wanted, one could use it to describe vast numbers of the phenomena of our technical world, such as the movements of machinery and wave processes — things which the Hopi have never known and for which, on the other hand, a definite nomenclature is lacking in European languages. This perfect functionality of Hopi is due to the fact that it establishes between two large categories of experience a general
PARTICLE AND FIELD IN HOPI
17
contrast corresponding to that between particle and field, considered by modern physics more fundamental than the contrasts between space and time or between past, present, and future, which are however imposed on us by the SAE languages which we speak. This contrast, Whorf concludes, 'being obligatory upon their verb forms, practically forces the Hopi to notice and observe vibratory phenomena, and furthermore encourages them to find names for and to classify such phenomena' (56).
5 AN AMERICAN INDIAN MODEL O F THE UNIVERSE
This conclusion ties in with the global description we find in the essay 'An American Indian model of the universe', published posthumously in 1950, but dating from 1936. Here too we shall adhere strictly to Whorf's own formulations. The Hopi •— or those of them who know only their own language and the cultural ideas of their own society—do not possess the NOTIONS of space and time that we do; even less do they have an INTUITIVE AWARENESS of them which could be considered necessary and universal. In particular, they do not conceive of time as a uniform CONTINUUM in which everything in the universe travels at an equal rate out of the future through the present and into the past, that is to say (looking at it from the opposite standpoint) in which the stream of duration continuously draws the observer away from the past and into the future. The Hopi language 'is seen to contain no words, grammatical forms, constructions, or expressions that refer directly to what we call "time" or to past, present, or future, or to enduring or lasting, or to motion as kinematic rather than dynamic . . . or that even refer to space in such a way as to exclude that element of extension or existence that we call "time" and so by implication leave a residue that could be referred to as "time" '(51). Thus there is nothing here that we could identify negatively as an implicit temporal dimension. It would, however, be gratuitous to assume that the notion of time is present in Hopi THOUGHT, independently of the language, in an intuitive manner, as an unanalysable datum. The Hopi language is by itself perfectly capable of describing precisely and exhaustively all observable phenomena in the universe from a pragmatic and
AN AMERICAN INDIAN MODEL OF THE UNIVERSE
19
operational point of view; and there is no reason why we should not admit a different description of the universe from our own, simply because it does not contain our familiar contrasts of time and space. Modern relativistic physics gives us such a description mathematically; Hopi gives us yet another, only not mathematically, but linguistically. In the Hopi world-view, time as an independent entity disappears and space ceases to be the uniform and temporal expanse of our supposed intuition (and of classical mechanics). Their regulatory function is taken over by other abstractions, for which there are precise terms in Hopi, or indeed which are implicit in the structure of the language; these different abstractions can also be deduced from an observation of the Hopis' cultural behaviour. We are dealing here with expressions neither more nor less mysterious than those which govern our European languages — of a three-dimensional space and of a time which flows uniformly in one direction and which is in its turn tripartite. Whorf calls them GRAND COSMIC FORMS and maintains that the Hopi speaker imposes these f o r m s on the universe. They are the MANIFESTED and the UNMANIFEST.
The manifested is that which has (already) manifested itself, revealed, shown itself; it thus falls into the category of the objective. The unmanifest is that which has not (yet) manifested itself, is hidden, not shown; it falls into the category of the subjective. The second cosmic form is by no means a negation of the first, even if numerous negations have had to be used to present it in English. Both forms are real, and indeed activity belongs to the second form. Let us examine them, if only for a brief moment, more closely. The OBJECTIVE of MANIFESTED 'comprises all that is or has been
accessible to the senses, the historical physical universe, in fact, with no attempt to distinguish between present and past, but excluding everything that we call future' (59). The idea of causation is absent from the manifested, for, if a cause were in act, its effect would not yet be realized, not yet delivered, and thus not yet manifested. The manifested is extension, with all the extensional aspects of existence; it comprises intervals, distances, seriations, and numbers. Time and movement here are dependent on the size and complexity
20
AN AMERICAN INDIAN MODEL OF THE UNIVERSE
of the operations by which events are connected. 'Events at a distance from the observer can only be known objectively when they are "past" (i.e. posited in the objective) and the more distant, the more "past" (the more worked upon from the subjective side)... If it does not happen "at this place", it does not happen "at this time"; it happens at "that" place and at "that" time' (63). The Hopi describes these relationships in his language simply as 'intervals of magnitude'. At a certain point, however, beyond certain boundaries which derive from the practical limitations of our human capacities, 'extension in detail ceases to be knowable and is lost in the vast distance' (63). In view of what has been said above, it can be understood how this unfathomable distance is, for the Hopi, at the same time the extremely remote past: in other words, the 'abysm of antiquity' described by Hopi legends — of which, it must be understood, the Hopi language carefully distinguishes the degree of reality, differentiating it from that of every-day practical affairs. At that distance, within that abysm, the objective merges with the subjective. The UNMANIFEST or SUBJECTIVE comprises in the first place everything we call 'future'; this is not a contradiction of the abysm of antiquity in which it becomes merged with the objective, for just as much future would be needed to reach the abysm. Also, part of the unmanifest is everything which we call 'mental' and think of as animation or as the efficient cause issuing from the nature of a thing; that is, everything which exists or appears in the mind, or (as a Hopi would say) in the 'heart', and not only in the heart of man, but also in the heart or essence of all other beings which we think of as animate or inanimate, and in Nature herself beyond her external forms and appearances (59). The essence and typical form of the unmanifest, of the subjective, is the TENDENCY TO BECOME MANIFEST, which is continuously pressing forward and emerging — to such an extent that, without wanting to create a confusion of terms, one could even oppose the MANIFESTING to the manifested. The unmanifest or subjective is the manifesting. We have here a gradual process of evolution, not however to be understood as physical movement, FROM the subjective TOWARDS a result, which, when
AN AMERICAN INDIAN MODEL OF THE UNIVERSE
21
achieved, will be a definite objective result. This evolution also includes that part of the present which is 'the moment of inception' (60), the BEGINNING of the manifestation; and Hopi has extremely precise terms to indicate this. All the rest of the present, insofar as it is undistinguishable from the past because in any case already manifested, belongs instead to the other cosmic form (6061). The world of the subjective has 'no space in the objective sense'; it gives more the idea of a vertical dimension, of an 'inner axis', 'like the growth-axis of a plant' (62). And around this axis the objective world extends in all directions. Thus 'the Hopi do not need to use terms that refer to space or time as such'; if anything, 'such terms . . . are recast into expressions of extension, operation, and cyclic processes provided they refer to the solid objective realm', or into 'expressions of subjectivity' if they refer to the world of the not-(yet)-manifest (64). One senses, as it were, 'echoes', which might seem to originate in a highly technological, decidedly post-Newtonian, operational, and pragmatic world, a world indeed with a touch of Heidegger about it. One cannot but suspect that all these things were seen in Hopi by Whorf because of his scientific preparation as an engineer and as a linguist, and his fervid imagination as an investigator; but the fact remains that he did not invent Hopi. The structures of Hopi are those which he describes and other specialists confirm this. 1 And in any case, such suspicions should be advanced only after examination of the methodology followed by Whorf and others in their descriptions of Hopi and of other languages so far removed from us.
1
Yet it is imperative to mention that C. F. and F. M. Voegelin have traced in Hopi also words for temporal units, apparently missed by Whorf: 'Hopi domains: a lexical approach to the problem of selection' (1957).
6 H U M A N ACTIVITIES A N D HOLISM IN WINTU
Another Amerindian language, Wintu, was studied at length by Dorothy Lee, an anthropologist of linguistic origin. Wintu is the Californian branch of the Penutian 'large family' and is spoken by a small tribe not far from the Oregon state boundary, in the area of Mount Shasta. Here I shall condense the exposition considerably; for an analytical examination of the Wintu language as such, Mrs. Lee's essay 'Conceptual implications of an Indian language'(1938) is indispensable. The noun and verb categories of Wintu express the basic conviction that 'reality — ultimate truth — exists irrespective of man. Man's experience actualizes this reality, but does not otherwise affect its being...'. 1 The Wintu actualises a scheme of reality which is already present in that reality, and thus renders it temporal in his own experience; but he neither creates nor changes anything. Wintu nouns refer to generic substances; the given is not a series of particulars to be classified into universals, but rather an undivided mass, within which the speaker delimits a particular, which consequently does not exist in nature but only in the consciousness of the speaker himself. 'What to us is a class, a plurality of particulars, is to him a mass or a quality or an attribute', united in a single concept expressed by a single word. The plural does not interest him, and in any case it is not derived from the singular. There are no plural forms; to say, for example, 'men' instead of 'man', a completely different root is used ( q ' i s instead of wi'Da). Objects appear in 1
Lee, 'Linguistic reflection of Wintu thought' (1944), in Freedom and culture (1959), p. 121.
HUMAN ACTIVITIES AND HOLISM IN WINTU
23
nature in a generic way, and their form is imposed by man through language (or rather, and this is the point, through that particular language which is different from all the others), by means of a specific act of choice or volition. A Wintu speaks in the first instance of 'deer-ness' (of being deer; of something which is deer, but without the idea of number), no B; he transforms it into 'a deer', noBum, when he needs it for some specific purpose. But as soon as he no longer needs it in that particular way, he makes it disappear as an individual thing, and allows THE deer to merge back into undifferentiated deer-ness. This procedure is the opposite of ours, as can be seen from the fact that no Bum is obtained in Wintu by adding a suffix onto the simple form, and not vice versa. Following an analogous type of construction, the individual is for the Wintu a delimited piece of society; society comes before the plurality of individuals into which it can be divided. While in our languages we use mutually exclusive dualistic categories, in Wintu they use categories which are inclusive but not mutually so: object A is included in object B, but not vice versa, and one then proceeds to distinguish B by means of various linguistic devices. For example, man is included in nature, which proceeds independently from him and so carries him forward within its motion; all the same, he in his turn does enjoy a certain independence inside the limits of natural law. Of particular interest is the Wintu concept of the person, which Lee defines as 'holistic, or psychosomatic, but without the suggestion of synthesis' of elements previously separate. 2 There are no words in Wintu for 'body' or 'corpse'; and the so-called parts of the body are aspects or localizations. Nor is there even a word for the self, which is indicated contextually by means of grammatical instruments. A study of the grammatical expression of identity, relationship, and otherness, shows that the Wintu conceives of self not as strictly delimited or defined, but as a concentration, at most, which gradually fades and gives place to the other.... For example, the Wintu do not use 'and' when 2 Lee, 'The conception of the self among the Wintu Indians' (1950), in Freedom and culture, p. 134.
24
HUMAN ACTIVITIES AND HOLISM IN WINTU
referring to individuals who are, or live, or act together. Instead of analysing the 'we' into 'John and I', they say 'John we' using the John as a specification. In this way, 'grammar contains in crystallized form the accumulating experience, the Weltanschauung of a people'. 3
3
Lee, 'Conceptual implications of an Indian language' (1938), p. 89.
7 NAVAJO, A 'CHEMICAL LANGUAGE'
In Navajo an immense precision in pronunciation is required to avoid continual misunderstandings caused by differences of sound which, to start with, the European ear can at first hardly distinguish. This is why Kluckhohn and Leighton say that Navajo is the most delicate language we know with regard to its phonetic dynamics.1 According to the common opinion of all those who have done field-work in Navajo, its vocabulary is extraordinarily rich. For example, there are a thousand RECORDED plant-names; each occupation or trade has its own complete terminology; and the terms for ceremonial techniques alone add up to at least five hundred. But what makes Navajo surprising and unique is that, even more than Hopi, it consists almost entirely of verbs. Its verb system is the most complex, rich, and precise ever to have been studied. Besides the few 'real' nouns, there are some substantival verbs which can also be conjugated like neutral verbs. Articles do not exist. In a formal sense there are no adjectives either, for those words which appear to be adjectives are in fact third persons of neutral verbs indicating qualities, states, or conditions. Many pronouns are absorbed by verbs; but there are also some which are used independently, either as prefixes to nouns, or as suffixes. Navajo nouns in any case have no gender and, with few exceptions, have the same form in the singular and plural. Finally, adverbs and conjunctions are extraordinarily precise and idiomatic. There are, for instance, enclitic particles known as 'directional', which divide space into zones and circles or into lines and directions from the point of view 1
Kluckhohn and Leighton, The Navaho, p. 258.
26
NAVAJO, A 'CHEMICAL LANGUAGE'
of the speaker or of the hearer, or of both together. Navajo verbs derive quite definite meanings from the assembling of elements that are generalised and colourless in themselves. Indeed, it might be called a chemical language. T h a t is, the basic process is that of utilizing the varying effects of small elements in different combinations. Syntax, to the N a v a h o
consciousness, is locked up, confined within the verb (Kluckhohn and Leighton, 263; my italics). The root of a verb bears an 'image' which remains constant — a nuclear notion much more specific than that of the vast majority of English verbs. To translate 'I give', one must in Navajo choose from among more than twenty forms which differ according to the nature of the object which is given. From the start distinctions are made between round objects, long objects, living objects, groups of objects, rigid containers with their contents, woven objects, bulky objects, groups of parallel objects, objects resembling a pile of wool, rope-like objects, mushy objects. 2 For each of these classes a different root must be used. But one cannot decide which root to use in Navajo on the basis of the nuclear idea present in an SAE language, because the difference between the structures is too great; furthermore, an SAE language is too vague and imprecise for this to be possible. To be able to say that it has begun to rain, a Navajo is constrained by his language to distinguish, amongst other things, the following circumstances: if he himself was aware of the beginning of the[jrain, or if he has reason to believe that it had been raining for^some time before he noticed it; if the rain is falling all around him, occupying his whole field of vision in an undifferentiated way,"or if, the rain still falling all around, there may not be some'sign of an abatement of the storm. Nor can a Navajo simply 'move', he can only move-in-a-certain-way; he cannot simply 'go', he must go-somewhere-specific; he cannot even just 'go-to-town', he must go-there-on-foot or go-there-by-some-means-of-transport (which^must'jbe specified); and so forth. 2
Hoijer, 'Classificatory verb stems in the Apachean languages' (1945), pp. 13-23.
NAVAJO, A 'CHEMICAL LANGUAGE'
27
In the few syllables which make up a single verb form, Navajo constructs a microcosm of meanings which often requires a long phrase when rendered in a European language. For example, 'eeshdeel 'I have eaten one by one some separable objects' (e.g., berries); hahlcoos 'you take a fabric-like object out of an enclosed space'; ni'a 'a group of round objects extends off in a horizontal line'; and so on. 3 We shall return to these aspects of Navajo and to the consequences which have been drawn from them with regard to linguistic relativity, at the beginning of chapter 9.
3
See chapter 8 of Kluckhohn and Leighton's book, various writings of Hoijer, as well as Young and Morgan, The Navaho language (1943).
8 FORMULATIONS O F THE THESIS O F LINGUISTIC RELATIVITY
Now that we have some descriptive material at hand, let us examine more closely how this thesis came to be formulated. It will be best to refer at some length the words with which Sapir and, especially, Whorf have expounded its various aspects. Sapir, who in Language (1921) tended to exclude the thesis of linguistic relativity, 1 becomes its advocate in an essay of 1928 (which appeared in 1929). Human beings do not live in the objective world alone, nor alone in the world of social activity as ordinarily understood, but are very much at the mercy of the particular language which has become the medium of expression for their society. It is quite an illusion to imagine that one adjusts to reality essentially without the use of language and that language is merely an incidental means of solving specific problems of communication or reflection. The fact of the matter is that the 'real world' is to a large extent unconsciously built up on the language habits of the group. No two languages are ever sufficiently similar to be considered as representing the same social reality. The worlds in which different societies live are distinct worlds, not merely the same world with different labels attached .... Even comparatively simple acts of perception are very much more at the mercy of the social patterns called words than we might suppose .... We see and hear and otherwise experience very largely as we do because the language habits of our community predispose certain choices of interpretation.2
1 On the differences and similarities between Sapir and Whorf see Guchman, 'È. Sèpir i "ètnograficeskaja lingvistika" ' (1954). 2 Sapir, 'The status of linguistics as a science' : in Culture, language and personality, p. 69; in Selected writings, p. 162.
FORMULATIONS OF THE THESIS OF LINGUISTIC RELATIVITY
29
Whorf put forward very similar views. To a certain extent he was influenced by and followed the example of Sapir, who collaborated with him in his work; but principally he was following up, generalizing at a theoretical level, the conclusions reached in the course of his own brilliant researches. When they began to examine critically a large number of widely differing languages, Whorf says, linguists found that the fundamental system of every language — its grammar — is not only a reproductive instrument for communicating ideas, but also is itself the shaper of ideas, the program and guide for the individual's mental activity 3 .... We are inclined to think of language simply as a technique of expression, and not to realize that language first of all is a classification and arrangement of the stream of sensory experience which results in a certain world-order, a certain segment of the world that is easily expressible by the type of symbolic means that language employs (55). The formation of ideas is not an independent process, rational in an absolute sense, as was once believed; but 'is part of a particular grammar' (212) and differs to a greater or lesser extent within different grammars. N o individual is therefore free to 'describe nature' with absolute impartiality: quite the opposite, he 'is constrained to certain modes of interpretation even while he thinks himself most free'. We thus arrive at a new principle of relativity, which holds that all observers are not led by the same physical evidence to the same picture of the universe, unless their linguistic backgrounds are similar, or can in some way be calibrated (214). Indeed, users of markedly different grammars are pointed by their grammars toward different types of observations and different 3 Language, thought, and reality, p. 212; we will be giving other page references in the text as we go along. The essays from which quotes are taken, are, in order, 'Science and linguistics' (1940, pub. 1942), the already mentioned 'The punctual and segmentative aspects of verbs in Hopi' (1935, 1936), 'Linguistics as an exact science' (1940, 1943),'Language, mind, and reality' (1942), 'Gestalt technique of stem composition in Shawnee' (1940), 'Languages and logic' (1941, 1943).
30
FORMULATIONS OF THE THESIS OF LINGUISTIC RELATIVITY
evaluations of externally similar acts of observation, and hence are not equivalent as observers but must arrive at somewhat different views of the world (221). Whorf specifies that every language is a vast pattern-system, different from others, in which are culturally ordained the forms and categories by which the personality not only communicates, but also analyzes nature, notices or neglects types of relationship and phenomena, channels his reasoning, and builds the house of his consciousness (252). The categories and types that we isolate from the world of phenomena we do not find there because they stare every observer in the face; on the contrary, the world is presented in a kaleidoscopic flux of impressions which has to be organized by our minds — and this means largely by the linguistic systems in our minds. We cut nature up, organize it into concepts, and ascribe significances as we do, largely because we are parties to an agreement to organize it in this way—an agreement that holds throughout our speech community and is codified in the patterns of our language. The agreement is, of course, an implicit and unstated one, BUT ITS TERMS ARE ABSOLUTELY OBLIGATORY; we cannot talk at all except by subscribing to the organization and classification of data which the agreement decrees (213-214). Two other passages f r o m Whorf give us an idea of the richness of his approach, of his methodological caution (seldom taken into consideration by his favourable critics and ignored by his adversaries), but also of the tangle of problems connected with the thesis of linguistic relativity: To compare ways in which different languages differently 'segment' the same situation or experience, it is desirable to be able to analyze or 'segment' the experience first in a way independent of any one language or linguistic stock, a way which will be the same for all observers. This cannot be done by describing the situation in terms of subject-predicate, actor-action, attribute-head, etc., for any scientific use of such terms contemplates that they shall have a variable meaning as defined for each particular language, including the possibility that for some languages their meaning shall be nil. Neither can it be done wholly by familiar terms ranging from the common-sense type to the quasiscientific, as by trying to break up the situation into 'things, objects, actions, substances, entities, events'. Cautious use of such terms may be helpful, perhaps unavoidable, but it must be remembered that in their ranges of meaning they are but the
FORMULATIONS OF THE THESIS OF LINGUISTIC RELATIVITY
31
creatures of modern Indo-European languages and their subsidiary jargons, and reflect the typical modes of segmenting experience in these tongues. They do not become scientific for linguistics because they may happen to be used in physics or chemistry (162).
Of the relationships among language in general, thought, and individual tongues, Whorf writes: Moreover, the tremendous importance of language, cannot, in my opinion, be taken to mean necessarily that nothing is back of it of the nature of what has traditionally been called 'mind'. My own studies suggest, to me, that language, for all its kingly role, is in some sense a superficial embroidery upon deeper processes of consciousness, which are necessary before any communication, signaling, or symbolism whatsoever can occur. ... It may even be in the cards that there is no such thing as 'language' (with a capital L) at all! The statement that 'thinking is a matter of LANGUAGE' is an incorrect generalization of the more nearly correct idea that 'thinking is a matter of different tongues'. The different tongues are the real phenomena and may generalize down not to any such universal as 'Language', but to something better — called 'sublinguistic' or 'superlinguistic' — and NOT ALTOGETHER unlike, even if much unlike, what we now call 'mental' (239).
As can be seen, the term 'linguistic relativity' does not refer to the diversity of languages as languages. That would be only banal. What the thesis does say is that the diversity of languages also makes itself felt in other fundamental fields of human activity. Languages, in their systematic diversity, represent diverse ways of thinking and bind their speakers to them. At the level of signantia we can neither go against the structure of the language nor go outside its system; for example, we cannot say 'that roses am viel schon': we HAVE TO say 'those roses are very pretty', or 'jene Rosen sind sehr schon'. Now the thesis of linguistic relativity posits analogous obligations at the level of signata. And since the signata are organized in organic semantic fields, which, taken together as a whole, make up the semantic totality of a language, whoever speaks that given language finds himself as it were enmeshed in it. The situation offers no alternative; or better, the alternatives it offers consist not in getting out of the entanglement but rather in changing it. This means that when we pass from one language to another
32
FORMULATIONS OF THE THESIS OF LINGUISTIC RELATIVITY
we change to another kind of entanglement; and the greater the difference between the two languages involved, the more remarkable the change will be. What is universal therefore is not the MANNER in which ANY PARTICULAR TONGUE exercises an influence on thought; on the contrary, the denial of this universality is a part of the thesis of linguistic relativity. What is affirmed as universal is the fact that every tongue influences thought, that is, that influences of this kind OCCUR. It immediately follows that as languages are in fact different from each other, the influences will also be different. If this is so, linguistic relativity is a new universal discovered by human research in the sense indicated by the Italian epistemologist Colorni. 4 The differences among various languages as studied by Whorf and the others do not regard those particular historical situations which may rightly be called cultural or literary. They do not regard things like not being able to translate Erlebnis (as used by Dilthey and Husserl) by a single word in English or Italian; or realizing that when a contemporary English philosopher speaks of 'experience' it is often difficult for Italians to indicate what he means by simply translating it as esperienza; or discussing how best to render in a Western language certain basic terms of Oriental systems of thought (an excellent example here is the tao which emerges as the bearer of such diverse meanings as 'road', 'principle', 'cosmic order', 'nature', but eludes the expression it requires). The differences and difficulties which are dealt with by linguistic relativity are part and parcel of the practical mechanics of any language in the way that it functions for all those who speak it at an everyday level; that is, at the level in which, for example, we say that in Italian there are two genders of article, in German three, and in Russian none, or that the temporal scansion of the Slavonic verb is substantially different from that of the Western European verb, or that the relationships among the words in a Chinese sentence are indicated not by modifications of the words themselves but by their position within the sentence (and by the presence of 'empty words', the function of which is supposed to be MERELY syntactic 4
Particularly in 'Critica filosofica e fisica teorica' (1948, posthumous).
FORMULATIONS OF THE THESIS OF LINGUISTIC RELATIVITY
33
and metalinguistic). The doctrine of untranslatability which idealist philosophers used to trace back to metaphysical ideas like 'the manifold differences of the expression of the Spirit in its inscrutable freedom' here assumes the character, which we could consider technical in its way, of a DESCRIPTION OF CODIFYING PROCEDURES which cannot be reduced one to the other. What we understand by 'style' and 'idiolect' does not concern the thesis of linguistic relativity which instead deals with each language as a social machine inasmuch as it is intersubjective or supra-personal. The principal case in which the thesis applies remains that of the mother tongue. But the formulations given above do not insist exclusively on this. They leave it to be understood, rather, that the influence of a language on thought can manifest itself also in cases when the language is simply taken up at any moment, even by someone who has another language as his mother tongue. Whoever learns a new language becomes a different person, as was also admitted by Mead in Mind, self, and society (283). An Englishman, for example, who as an adult learns to speak Chinese will begin to see the world through the structures of Chinese, even if he will still be conditioned by the fact of having originally learned to see it through the structures of English. This corresponds to the common experience of one who, gifted with Sprachgefühl, has the sensation of no longer thinking in exactly the same way when he adopts another language. The idea of a direct and immediate influence of a language which is learned and spoken in any given moment clearly constitutes a stronger and more binding statement of linguistic relativity. It is easier to accept the idea that everyone, to some extent, builds his own world-view by learning to speak in a given language and therefore within the realm of the structures of that language, rather than the idea that world-view and mode of thinking change ipso facto when one passes from one language to another. In the second case the question also arises of the basic operational residue of one's mother tongue, something which is also felt by the polyglot endowed with Sprachgefühl. If we are no longer using our mother tongue, and feel conditioned both by the language we are speaking, and, in the background, by our mother tongue, then
34
FORMULATIONS OF THE THESIS OF LINGUISTIC RELATIVITY
questions arise as to which are the status, the locus, the modus of the residual conditioning. A first reply would be that it is to be sought at a 'lower' level because it is 'deeper', where thought 'becomes' language. But this is just another way of saying that languages do have some sort of influence on thought.
9 EXCESSIVE INTERPRETATIONS
At the end of the second chapter we alluded to excessively favourable or unfavourable reactions to the idea of a differential influence of languages on thought and on world-view. Now that we have a modest stock of examples at our disposal, we can develop the discussion further. One could be strongly tempted to say straight off that Whorf is right when he maintains that only equally diverse world-views can accompany or correspond to languages as diverse as those he describes; or that Weisgerber is right when he elaborates Humboldt's notion of the 'intermediate world' to the point of giving priority to an infralinguistic concept of reality. For example, it seems one should say that the Navajos look always or predominantly at concrete particulars. Since they concentrate their attention on only some aspects of every situation (especially, as we have seen, those regarding the type of activity performed), this would prevent them from using general concepts and from assuming the existence of anything substantial and permanent. These differences of attitude, behaviour, and thought would seem to be crystallized and perpetuated in the very forms of Navajo grammar. Speaking in English (or in Italian or in any other SAE language), we subsume under some more general concept the particulars on which a Navajo would concentrate his attention. For the Navajo speaker, it is precisely this more general concept which is not made available to him by his language, so that for him these particulars are not part of something larger, but appear as something primary. According to this interpretation, it is not enough to say that a
36
EXCESSIVE INTERPRETATIONS
Navajo's spontaneous way of thinking is more concrete, specific, particular, precise to the point of everyday fussiness, than the European way of thinking, or that it is organized linguistically according to complex 'geometries' completely unknown to us; and it is not even sufficient to say that his language makes him classify the elements of experience in a different way from ours. One has also to establish HOW that which a Navajo considers essential varies. According to Sapir, Navajo is nearer than English or Latin 'to a freshly objective view of the nature of events' (1931: 578. Reported in Kluckhohn and Leighton, The Navaho, p. 284). One can add that, in Navajo, everything which in the development of thought is founded on the general would have been impeded by the immense network of distinctions imposed by the language; or even that in such a language, based on verbs and on the modalities of actions, a metaphysics of the Aristotelian type cannot be expressed and THEREFORE never came into being. From the texts examined, for example, it would appear that a Navajo could not deal with movement-in-general. Were he to contemplate even the sole idea of going-in-general, which linguistically he cannot express, he would have to do it with a thought independent of language. Now, even if we can say that one can think also without speaking (both in the sense of keeping quiet while thinking and in the sense of carrying out intelligent behaviour of a non-linguistic character), the fact remains that ideas are formed linguistically. Without actually using a language it is not possible to communicate ideas in a complex and worked-out manner, and without the EXISTENCE of a language a good part even of non-verbal communications would not exist either. Speaking in a language in which expressions for the general do exist — so this interpretation goes on — we believe that we see a deficiency in the Navajo approach to the universe. But the Navajo does not feel this deficiency in the least. He is completely enclosed within the mental curvature imposed on him by his language. It therefore makes no sense to suppose that he can think the general apart from his language, just as it makes no sense to presume that we have truly grasped in an SAE language the differences present
EXCESSIVE INTERPRETATIONS
37
in Navajo. When, in translating, we think we have grasped them, we are underrating enormously the incomplete and imprecise, indeed merely allusive character of our translations, which cannot render the implications of the original. Even if we can indicate in an SAE language a difference not present in its structure, the SAE text cannot tell us how and in what respect the translated form is distinguished from all the other possible forms; that is to say, what are the series of oppositions which depart from that form in the various directions existing in the language. A complete translation would need the help of long theoretical explanations about the classificatory systems on which that form rests and would therefore end up by being equivalent to at least a theoretical command of a sector of the very language from which the translation is being made; or rather, the command should also be practical, spontaneous, unconscious: based on the simultaneous presence — actual or immediately potential — of all those series of oppositions. But such a command is possible only of a mother tongue (and, only very secondarily, of a thoroughly learned foreign tongue). One should conclude from this that, not having learned, for example, Navajo as a mother tongue, we are all deaf to certain distinctions; whereas the Navajo, not having learned an SAE language as a mother tongue, is deaf to certain others. In an analogous manner, an adequate presentation of the Hopi conception of the universe could only be made in Hopi (suffice it to recall that what for the Hopi are sentences-and-therefore-propositions about happenings or events are for us sentences-and-therefore-propositions about things): and vice versa, a Hopi succeeds in glimpsing the universe according to our conception only to the extent in which he learns an SAE language. It is clear that according to such interpretations of linguistic relativity as these we would find ourselves facing separate universes, between which no communication is possible. Every speaker would be a prisoner of the universe constituted by his own language. To the extent in which we speak different languages, that is different languages whose structures do not overlap (it is evident that the structures of, for example, Italian and Spanish, despite seemingly
38
EXCESSIVE INTERPRETATIONS
important differences, overlap considerably, if we just compare them to languages of different stocks), to that extent we Would be DIFFERENT MEN.
For philosophy and for theoretical thought in general, these are devastating propositions. The Kantian flavour of Whorf's own statements, as quoted above, gives rise to the suggestion that, transferred to language and further historicized, Transcendental Aesthetics and Logic must be multiplied by the number of languages which exist or indeed have existed on the planet; and even if the bourgeois ideology of philosophy as a general or 'first' science has been abandoned, it is still startling to hear that there must be thousands of Transcendental Aesthetics and Logics. The a priori turns out to be confined not only to language in general, but indeed to the various languages; changing language, one changes a priori. However, it is not only an eighteenth century approach like Kant's which becomes relativized and which crumbles away. Husserl's idea of the grammatical categories common to all languages also collapses. In his fourth 'logical research', dedicated to the idea of pure grammar, Husserl in fact maintains that there are das Apriori, das rein im gattungsmässigen Wesen der Bedeutung als solcher wurzelt (336) ... apriorischen Fundamente ... betrifft die wesentlichen Bedeutungsformen und die apriorischen Gesetze ihrer Komplexion, bzw. Modifikation, und keine Sprache ist denkbar, die durch diese Apriori nicht wesentlich mitbestimmt wäre (338) .... Alle in der reinen Formenlehre herausgestellten, nach Gliederungen und Strukturen systematisch erforschten Bedeutungstypen — so die Grundformen der Sätze, der kategorische Satz mit seinen vielen Sondergestaltungen und Gliederformen, die primitiven Typen propositional komplexer Sätze, wie die konjunktiven, disjunktiven, hypothetischen Satzeinheiten, oder die Unterschiede der Universalität und Partikularität auf der einen, der Singularität auf der andern Seite, die Syntaxen der Pluralität, der Negation, der Modalitäten usw. — all das sind durchaus apriorische, im idealen Wesen der Bedeutungen als solcher wurzelnde Bestände ... [und gleichen] einem absolut festen, sich in empirischer Umkleidung mehr oder minder vollkommen bekundenden 'idealen Gerüst' (339). 1 1 Husserl, Logische Untersuchungen, vol. II, first part: 'An a priori rooted solely in the essence of meaning as it is generically understood (336) ... aprioristic foundations ... concerning the essential forms of meaning and the a priori laws of their combination or of their modification ... an ideal frame-
EXCESSIVE INTERPRETATIONS
39
It is precisely propositions like these of Husserl's that are contradicted one by one by an extreme interpretation of the thesis of linguistic relativity. In his essay 'The relation of habitual thought and behavior to language' (1939), Whorf denounces the assumption that a 'natural logic' exists disconnected from the particular language that we happen to speak. If his criticism is correct, philosophical analysis of a language in its everyday usage 'becomes' — or rather, shows that it has always been — only a description of what happens linguistically within a given community. In this sense ANALYSTS are discovered to have always been ANNALISTS: compilers of linguistic annals for local use, recorders of small linguistic habits in which are reflected the ideologies, customs, fashions, or even idiolectic crazes which flourish within a given jargon community. Notwithstanding a few scattered hints at such a difficulty by the always acute Waismann, and later, for example, also by Flew 2 — so far behind linguistic and ethnological research in any case — it can be said that the problem has never been felt by analytical philosophers. By contenting themselves with analysing their own mother tongue in their own mother tongue, they have shown their conviction that this was sufficient for the clarification of human problems in general. work which every language ... fills and clothes with empirical materials in various ways (so that) no language is thinkable which is not codetermined by such a priori (338) ... All types of meaning put in evidence by pure morphology ... like the fundamental forms of sentences, the categorical sentence with its numerous particular figures and forms of articulation, the primitive types of propositionally complex sentences, such as the conjunctive, disjunctive, hypothetic sentence units, or the differences between universality and particularity on the one hand and singularity on the other, the syntaxes of plurality, of negation, of modalities, etc. — all these are wholly aprioristic entities ... [and constitute] an absolutely stable 'ideal framework', which manifests itself in a more or less complete manner under an empirical clothing' (339). [The above translation can be compared with that by J. N . Findlay which has now appeared (1970).] 2 For Waismann, see the series 'Analytic-synthetic' (1949-1952), especially the fifth essay, and 'Verifiability' (1945), pp. 137-139 (on p. 138 the principle of linguistic relativity is almost stated); for Flew, his introduction to the second series of Logic and language (1953), p. 3, and 'Philosophy and language', in Essays in conceptual analysis (1957), on pp. 4-7.
40
EXCESSIVE INTERPRETATIONS
T h e i r w o r k r e v e a l e d t h e c o n s t a n t a s s u m p t i o n t h a t there w a s s o m e t h i n g b e y o n d l a n g u a g e s that c o u l d be r e a c h e d i n d e p e n d e n t l y o f t h e s i n g l e l a n g u a g e in w h i c h it w a s b e i n g e x a m i n e d ; a n d in this s o m e t h i n g it w a s difficult t o a v o i d s e e i n g t h o u g h t in general — o r at least ( t o p a y t r i b u t e t o the d e s c r i p t i v e p r e c i s i o n o f t h o s e p h i l o s o p h e r s ) this o r t h a t a s p e c t o f that w h i c h is usually referred t o w h e n t h e t e r m ' t h o u g h t ' is u s e d i n a n u n s p e c i f i e d m a n n e r . T y p i c a l o f w h a t w e are s a y i n g is t h e s t a t e m e n t t h a t t h e y w e r e n o t l o o k i n g , f o r e x a m p l e , f o r t h e USE o f c a u s e as distinct f r o m that o f
Ursache
(these are USAGES, m e r e l y l i n g u i s t i c a n d social institutional f a c t s ) ; b u t rather t h a t t h e y d i d l o o k f o r USE AS FUNCTION (USE as o p p o s e d t o USAGE), as s u c h s o m e t h i n g c o m m o n t o b o t h t e r m s ( t h e distinct i o n b e t w e e n USE a n d USAGE c o r r e s p o n d e d t o s o m e extent t o that, n o t f o u n d i n E n g l i s h , b e t w e e n langage were c o n c e r n e d w i t h le langage
3
a n d langue;
a n d l i n g u i s t s w i t h les
philosophers langues).3
There is no doubt that at Oxford the intention was to obtain results of not only local but also international value. This need was continually expressed, particularly in the form of an opposition between linguistico-philological research (understood as historical and individualising) and logico-philosophical research (understood as structural and universalising). On the other hand, the analyses were carried out in only one language, English; and the technique consisted precisely in appealing to the structure and to the lexicon of that language in order to decide logical and philosophical questions. Great difficulties could have been smoothed out, or at least brought to light, through an acquaintance with the literature — especially the linguistic and ethnological literature — on linguistic relativity. But no one, not even the acute and welllearned Austin — although he openly lamented his and his colleagues' ignorance of the 'facts of language' — was in a position to furnish any aid in thai direction. As a result of this, having busied myself uselessly for a while with enterprises like a awatjfiaaia of my own invention, in 1953 I tried to tackle the difficulties I encountered in a practical manner, and in general to verify in another language the validity of philosophical investigations centered around common speech, by means of an adaptation (with cuts, additions, and substitutions of examples ex novo) of Gilbert Ryle's very typical book The concept of mind (the Italian edition appeared in 1955 under the title Lo spirito come comportamento). Synsimasia, as opposed to synonymity, i.e. to the study of words bearing similar meanings, was intended as a study of the diverse meanings of the same words. As such it pointed towards (i) a 'syntactics of meanings' constructable at a level below the level of glottic articulations, and (ii) towards a comparison of semantic spheres connected to key-terms considered as correspondent in two or more 'natural' languages (e.g., the semantic sphere of
EXCESSIVE INTERPRETATIONS
41
In this way a DETERMINISM of languages on (let us say) thought was affirmed, while the differences among the various languages continued to be simply ignored, or else were grotesquely shunted off into another brand of academic specialization which, it turned out, had been discarded in advance as irrelevant. The problem of whether different languages exercised a different determinism was not posed at all; and even when it was touched upon was never faced. It seems that there are three pillars on which any attempt to construct a nonglotto-centric analysis of language must rest: the mutual determinism of the language and thought, the differences among the various languages and the consequent variations of the determinism itself, and the further mutual influences between the linguistic and the non-linguistic. Of these three pillars only the first had been erected, between the Cam and the Isis, to the detriment of the other two; and of the first, only the direction from language to thought was considered. The glottocentrism and sociocentrism of the whole approach could not have been put into a clearer light by their own victims. Nor could it be argued that, after all, at Oxford the idea of thought as independent from the language was current (a thesis not to be confused with that of the ATTAINABILITY of thought INDEPENDENTLY FROM the language in which the research is being made). On the contrary, typical of Oxford was a kind of unconscious renewal of a fundamental Marxian intuition, which in its turn rested on the dialectical structures discovered by Hegel in the 'theoretical spirit'. It was maintained that a thought independent from the language simply does not exist (apart, naturally, from the already mentioned cases in which the presence of thought is shown by forms of behaviour other than linguistic). In this way, the glotto- and sociocentrism of the whole approach was made worse; for if instead of a mere determinism OF the language ON thought one had to go so far as to speak in terms of identity of the two, then the very structures of the single language under examination were to be considered the Italian sentire compared with that of the English 'feeling'). Of these investigations, which remained unpublished, there is some trace in the note 'About the present version' which follows the 'Introduction' to the Italian edition of Ryle's book.
42
EXCESSIVE INTERPRETATIONS
as bearers of wisdom-and-truth — and this wisdom-and-truth, therefore, varied automatically and irremediably from language to language. It must be recognized that the success of the Oxonian analysis of common speech was ALSO international (in the sense of interlinguistic). And this fact, that is that translations were made of some of their works, implicitly demonstrated, or bore factual witness, that the main tenets of the analytical school could not be completely restricted within the confines of the structural and idiomatic characteristics of the language in which those analyses took place. 4 It remains to be seen up to what point the restriction was valid: namely, the extent to which the structural characteristics of the various languages were a determining factor for those analyses, in the sense of limiting their translatability and thus their exportability. Was the use of that linguistic capital such as to condition the market? Were linguistic commodities produced at Oxford only for internal consumption? These and other questions of a like nature bring us once again into the midst of the difficulties raised by the thesis of linguistic relativity; and one can conclude that it was, in any case, a grave shortcoming on the part of philosophical analysis of common language not to have taken it into account, indeed not to have made it, right from the beginning, one of its central problems — were it not for the suspicion that had this happened, that kind of philosophical analysis would never have been able to materialize. Accepting Whorf's disavowal of 'natural logic', the psychologist Smythies also argues against the philosophical cult of common linguistic usage within a given language; and concludes that 'for the physiologist... the standards of correctness in common English usage represent only one of a number of possible equally valid ways 4
Whoever wishes to develop this line of thought, that is to see to what extent Oxonian analysis made speech its object and did not limit itself to examining the structures of usage within one language, can begin with Ryle's essays from 'Categories' (1938) to 'Ordinary language' (1953) and 'The theory of meaning' (1957); cf. also the posthumous works of Wittgenstein. For the possibility of interpreting the second Wittgenstein in a non-sociocentric way, see 'Per un uso marxiano di Wittgenstein' (1966), now also in II linguaggio come lavoro e come mercato.
EXCESSIVE INTERPRETATIONS
43
of analysing experience'. It will therefore always be permissible to violate the normal usage of this or that language in order to study 'the natural processes of perception'. 5 For Smythies linguistic relativity is a reality; but it is a reality which sends us back to the prelinguistic level and which demands the formation of a special terminology or even of a technical language, capable of describing univocally what is happening. However what are the 'natural processes' which Smythies invokes as something independent of our speaking about them, and what are the operations for identifying them, operations to be performed in any case THROUGH language; and how are we to be successful in this venture despite the recognized variety of languages — these too are difficulties which, rather than removing the problematic of linguistic relativity, bring us right back into the heart of it. It can be said that the more directly interested parties, the linguists, have in general accorded to linguistic relativity an always joyous, and at times a triumphant reception. There has been a sort of Academic orgy about it. But one should add that this was the case up to some years ago. The more recent trend is instead that of looking for 'linguistic universals' — that is for elements which remain constant in varying contexts — and for 'deep structures': both of them independent (in degrees still to be ascertained) from those elements which differentially constitute the various languages in their relatively 'superficial' aspect. For linguistic universals the reader is referred in particular to the work of Greenberg, for deep structures to that of Chomsky. 6 5 Smythies, Analysis of perception (1956), p. 105, and all of Par. 12.8; in this context, also Yinger, Toward a field theory of behaviour, and Werner and Kaplan, Symbol formation. These are all instances of psychologists who, in their way, accept linguistic relativity. Yinger even states that 'the fact that we see the world through lingual glasses is now well ascertained. What Hallowell calls 'the ubiquitous role of symbolic mediation of the world' results from the way in which reality is defined by the concepts, categories, and descriptive terms of the language' (p. 56). [A COMPLETE acceptance of linguistic relativity is in Shands (I960).] 6 Greenberg is the editor of Universals of language (1963), in which, together with others, he contributed to the initial 'Memorandum concerning language universals' (pp. XV-XXVII) and with an essay of his own on 'Some universals of
44
EXCESSIVE INTERPRETATIONS
These new trends are prevalent mainly in the English-speaking world, and especially in the United States (it should be added that they do not exclude the thesis of linguistic relativity in a global manner, but tend rather to reformulate it with moderation or to recognize its existence with a certain residual perplexity). But in spite of them it can be said that, on the whole, the joyous and triumphant reactions are still the most widespread. I recall in passing some important examples, chosen from non-English languages in order to broaden the discussion and afford a glimpse of some variations in approach. André Martinet strongly recalls Whorf's thesis when he attributes to a language complete freedom to choose and delimit not only its signantia but also its signata;1 Georges Mounin and Emile Benveniste take it for granted — it is not even a subject for discussion; it is enough to mention it, as being obvious, against those who persist in not understanding it. 8 Helmut Gipper devotes the whole of the fifth chapter of his Bausteine to it with loud praises (297-366); Julius Stenzel (who as far back as 1921 had upheld the direct influence of the Greek language on Greek philosophy) devotes a section of his Philosophic der Sprache to the 'language as articulation fGliederung] of the objective world and of the spirit which reflects and conceives it' (35-38), and the last chapter, after the manner of Weisgerber, to 'mother tongue and conception of the world'. Even Hjelmslev's distinctions, founded as they are on the sign-function conceived as internal rather than external to the sign (a choice in which Hjelmslev openly follows Saussure and Weisgerber), make the 'purport' — that is, the factor
grammar with particular reference to the order of meaningful elements' (pp. 73-113); later he wrote the monograph Language universals (1966). For Chomsky's work, see the well-known series of monographs listed in the references. The problem had already been posited in various guises, for example, by B. W. and E. G. Aginsky in one of their articles in Word (1948). See also Word classes, a special issue of Lingua (1966), reprinted in 1967. 7 See the second part of Chapter 8 here above. By Martinet, see 'Arbitraire linguistique et double articulation', pp. 115-116. 8 Mounin, Les problèmes théoriques de la traduction, ch. IV. Benveniste, 'Tendences récentes en linguistique générale', pp. 133-134; 'Catégories de pensée et catégories de langue', pp. 419-421, 426, 428-429.
EXCESSIVE INTERPRETATIONS
45
common to all languages — something which can be interpreted only in the different organisations and articulations made of it in the various languages in which it is realized'. 9 Further 'confirmations' of linguistic relativity can be found also in scholars of diverse cultural origin, who deal with very specific subjects. For example Elders supports it vigorously in his studies on the relationships between Japanese language and thought; Luther traces and confirms it in his analyses of the history of the Greek mind from Homer to Aristotle. In the second volume of his tetralogy Weisgerber himself now speaks with satisfaction and approval of the theories of linguistic relativity of American origin as a confirmation of his own theories. 1 0 Besides — getting back to the English speaking world — as far back as 1900 Bertrand Russell had linked the idea of substance to the belief that every proposition must necessarily consist of a subject and predicate {Leibniz, 14-15); in 1921, following the Orientalist and comparativist A. H. Sayce, he repeated that from Aristotle's day on, Western philosophy had been dominated by the belief in a world which was of necessity divisible, like sentences, into subjects and predicates, that is, into substances and attributes (The Analysis of Mind, 212); and in 1927 he suggested that someone should write 'a big book' about the influence of the predicative structure on European thought and, in particular, on the notion of substance.
9
See in particular the Prolegomena to a theory of language, pp. 47 ff. (It. 52 ff.), 50-57 (It. 55-62); but also his essays of 1943, 1947, 1954, and 1957, now united in Essais linguistiques (1959). For the conception of 'the sense of what is said' as something internal to words and to the structure of the phrase, see also De Waelhens, Existence et signification, pp. 136-137. 10 Weisgerber, Die sprachliche Gestaltung der Welt, pp. 44-46. Of course it is impossible to understand the works of either a Trier or a Wartburg without preliminarily accepting linguistic relativity; a recent monograph on the relationships between Weltbild, Denkform and Sprachgestalt is that of Tschirch, and bears a title composed of these terms.
10 SPHERE A N D MODE OF THE PRESUMED INFLUENCE O F THE LANGUAGE
Things are not so simple, however: one would need to know IN WHAT WAY and IN WHICH SPHERE precisely the conditioning action of a language on thought takes place. In this section some aspects of a possible conceptual systematization are put forward. 1 The first thing to be considered is that the notion of reality as opposed to the language — the reality which thought, under the sway of the language, supposedly recognizes or constructs in different manners — is a notion which is anything but univocal. Sometimes it is a question of a reality tout court, pregnant with unresolved philosophical difficulties; sometimes of the reality of physical objects and of their arrangements and articulations; sometimes of social reality in one or another of its aspects, that is of a reality which is in its turn made up of non-verbal signs or indeed even of linguistic material. Everything on which the language exercises an influence is, in turn, the thought, the mentality, the experience, the attitudes, the mode of behaviour, the world-view, the basic ideology of a 1
This sketch is independent of the important works by Fishman, 'A systematization of the Whorfian hypothesis' (1960),'Who speaks what language to whom and when' (1965), and the review of Hertzler's volume A sociology of language (1967); Hymes', 'Two types of linguistic relativity (with examples from Amerindian ethnography)' (1966); Seiffert's 'Neo-Humboldtian semantics in perspective: "Sprache und Gemeinschaft"' (1968); andTrager's 'The systematization of the Whorf hypothesis' (1959). It seems to me that Fishman and Hymes in particular have clarified the situation in an objective and systematic manner instead of contenting themselves with advancing destructive criticism, like, for example, Feuer (1953), Lenneberg (1953), and Black (1959); or instead of being too inclined to acceptance, like Basilius (1952). This is also the place to mention the symposium Language in culture (1954), wholly dedicated to the examination of 'Whorf's hypothesis'.
SPHERE AND MODE
47
community. This influence is covertly or overtly described in various ways as causal, or as conditioning (wholly or in part) its object, or as accompanying its formation. Here and there the dialectic emerges between the topicality of the various languages and interlingual universality. As for the language, its influence is localized sometimes in its grammatical structure, sometimes in its vocabulary, and occasionally in certain syntactical and semantic characteristics; and it is either limited to partial aspects, or extended to global features, which, constitutive of the language in question, make it different from others. It is impossible to unravel this tangle here (and it is perhaps useless to attempt it anywhere): an interminable casuistry would result — some sort of gigantic table of all linguistic factors that could exercise any type of influence on all the non-linguistic factors conceivable, as well as on other linguistic factors. Here we shall make a paradigmatic outline for the reality of physical objects only, a paradigm which with the necessary completions could also be repeated for other types of entities, taking into account the level and tenor of reality attributed to them. Starting from the premise that there must be some sort of influence of the language upon the determination of the reality of physical objects (were we to deny ANY kind of influence, this discussion would not even be worth starting), it seems that one can distinguish immediately between the following two notions at least. (1) What has been collectively elaborated within a given mother tongue is real; we consider real, for example, a TREE, which for an Italian is an albero, for a German a Baum, for a Chinese a shit, with all the different syntactical, semantic, and pragmatic associations which the word brings with it and gives rise to in the different languages. In this sense everything that is real belongs to, or cannot leave out of account, the 'intermediate world' [Zwischenwelt] of our mother tongue. Here the stress is laid on the differences to be found between tree, albero, Baum, shü and so forth; not only on the differences in the ways in which words like these are received and positioned within the systems of their respective languages, but also on the differences which obtain at the underlying level of signata,
48
SPHERE A N D MODE
that is, in the last instance, at the level of the meaning-conferring behaviour of those who, within those languages, use those words in their speech. The extreme version of this consists in saying that a tree is not an albero, and so on; that an Italian will never see a tree as an Englishman does, for the simple reason that an Italian does not see a tree but an albero. This point of view can already be traced in Humboldt, and is now typical of Weisgerber. (2) However, that which lies BEYOND the intermediate world of the mother tongue is also real, or beyond the language tout court (whatever it is, maternal or not) which one happens to be using, or indeed beyond any symbolism. It is not with language alone that something real can be reached. In more than one important sense physical objects 'are there'. It is true that if we begin to speak about them, then switch to another language, it CAN HAPPEN that the way in which we speak about them changes and that the subject-object relationship is thereby altered to some extent and in that particular instance even though it can be assumed that the object remains what it is. But it is equally true that there are many other ways of reaching physical objects: they can be touched, moved, manipulated in various ways. As Rothacker, a recent and favourable critic of Weisgerber says, Die ganze Welt, in der wir leben und auf deren anschauliches Bild sich unsere Handlungen beziehen, ist Sprachinhalt, d.h. erwortet.2 Alle Phänomene, die wir überhaupt kennen, sind erwortete Phänomene ... Anderes als Erwortetes kennen wir anschaulich ja gar nicht. Alles was wir 'kennen' muss etwas Zwischenweltliches sein. Tatsächlich ist es ja auch immer lebensbezogen und 'bedeutungsvoll'.... But on the other hand the same author also puts forward the hypothesis that wir primär PRAKTISCH in die Wirklichkeit eindringen und dass wir SPRACHLICH nur synthetisieren und stabilisieren, was uns zunächst dort in der Reibung mit Wirklichem bedeutungsvoll, wichtig oder auch beunruhigend wurde ... , 3 2
'Spoken' or better 'put into words', but still better (if one could say it) 'inworded'; we shall use this last term in the English translation given in note 3 below. 3 Rothacker, 'Ontologische Voraussetzungen des Begriffs Muttersprache' (1959), pp. 44-45. The following is a translation of the two passages: 'The
SPHERE AND MODE
49
So it would seem that even among the supporters of a very extreme form of linguistic relativity there is a definite awareness that the object can also be reached by non-linguistic means and that, whatever the means, it does to some extent remain what it is. We could therefore correct the formulation given in the second chapter with regard to experience OF one's mother tongue, and say instead that it is a matter of experience OF THE WORLD, which takes place THROUGH the mother tongue (and can take place through other means too). But who can give us the criterion for making the distinction? When, and in what manner will the experience of the language begin, and when the experience of the world? And in short, how can one distinguish between linguistic and non-linguistic if not insofar as the linguistic has already been established? The thesis asserts that it is precisely in having experience of the world that we have experience of our mother tongue, and vice versa; in other words, that the two processes cannot be separated. It seems possible at this point to gather together in schematic form some of the uncertainties which have been raised by our examination of the thesis of linguistic relativity. It is not clear which of the following propositions the thesis states, or which it prefers: (a) The DIRECT INFLUENCE on thought of certain characters of language, as they appear embodied in a given language. It is always a matter of characters which belong to a given language [langue] and only as such belong to the more general notion of language [langage] which is the dialectical sum of THE language and speech [of langue and parole]. Furthermore in some cases it seems that this influence is CAUSAL, in others only DETERMINING, and this to an extent which can be more or less complete and intense; and as whole world in which we live and to whose intuitive image our operations are referred, is linguistic content, that is 'in-worded'. All the phenomena that we usually know are 'in-worded' phenomena.... Intuitively we do not know anything which is not 'in-worded'. Everything that we 'know' has something of the intermediate world about it. In fact it is already something always referred to life, something 'significant'....' 'We enter into reality primarily IN A PRACTICAL WAY; IN A LINGUISTIC WAY, we only synthetize and stabilize that which, at that point of our friction with reality, stands out as significant, important or even disquieting.'
50
SPHERE AND MODE
for the problem of a RECIPROCAL ACTION, it is left hanging in mid-air and never properly faced. Subordinately it is never really clear what one ought to understand by 'thought': the opinions of the various authors differ, or pile one on the other from page to page. It is certainly possible to distinguish AT LEAST the following items: (1) the fundamental categories of thought formally considered: the traditional categories, like unity, plurality, affirmation, negation, modality; or also those which we can call 'new', not because they did not exist before, but because research has concentrated upon them over the last few decades, meaningfulness, significance, reference, various types of duty, various types of possibility, and so on; (2) the contents of thought which are actually present in the form of ideas, images, intuitions, representations; (3) collective psychological habits, that is, things like the tendency to interpret certain phenomena in one way rather than another, to have certain relations with the rest of the world (objects, animals, people, natural phenomena, supernatural beings); (4) a prevalence of each one of the above, or of a particular combination of them. — At which of these levels of thought, on which of its various senses and aspects, is it maintained that the mother tongue exerts its influence? (b) The CONCOMITANCE of certain linguistic [lingual] characters with certain characters of thought, which here too could be any one of those listed under a. In this case no causal or determining influence is affirmed. But by affirming the concomitance one is also affirming the hermeneutic value of the linguistic characters alone in relation to the others; so that in theory it would be possible to reconstruct a past civilization from its language alone, if this were the only thing to survive. In these terms the whole thing may sound rather like science fiction. For this claim to have enough sense to be a fit subject for criticism, it should be borne in mind that if, on the one hand, this reconstruction would not be restricted to the existence of objects and institutions which are actually named and described in a recognizable manner in the language (which is obvious), on the other hand it would include the way in which the speakers thought as well, and therefore, TO SOME EXTENT, also the
SPHERE AND MODE
51
institutions which were built up by that thought. That is, this concomitance would permit one to affirm in principle that 'given a certain linguistic structure, the speakers must have thought in a certain way'. The deception which this statement can contain is the very measure of truth which may be attributed to it, since it is founded on the relative IDENTITY OF THOUGHT AND LANGUAGE [Ilangage]; except that this identity is not necessarily that BETWEEN THOUGHT AND THE LANGUAGE [langue]. This is one of the basic cruxes of the whole matter. (c) The STATISTICAL PREVALENCE of certain modes of thought in communities speaking a given language and thus also in the individuals belonging to that community when compared to individuals belonging to other communities. The difference between this statement and the preceding one is that here the question is that of a relative concomitance: i.e. it is admitted here that the concomitance also may not occur. In neither (b) nor (c) is the causal or determining nexus affirmed; but in both, with an approach which is apparently purely descriptive, it is affirmed that certain phenomena accompany certain others. Still, not even here is the approach purely descriptive, as we can infer from the fact that it has isolated two classes of elements — those of the language and those of thought — from a background which contains both and which is considerably more complex than either of them, and also than their sum. The development of this last remark will bring us to a new field of discussion, of a much more decisive importance than any of those so far touched upon. In conclusion, it would seem that to say that the language exercises an influence on thought is to say almost nothing. As soon as we get out of the feeling of agreement and intellectual contentment which the thesis of linguistic relativity may produce, and we ask ourselves in what exactly does this influence consist, those same fascinating formulations of Whorf's begin to appear rather ambiguous (although a squarely faced study of Whorf's entire work would bring to light many subtleties which in my opinion have escaped not only his critics, but also his disciples and advocates). If we distinguish (1) the factors of the language which exert
52
SPHERE AND MODE
an influence, (2) the manner of such an influence (for example, as we have seen, direct influence, concomitance, statistical prevalence), and (3) the linguistic or non-linguistic factors upon which the influence is supposedly exerted (for example, various types and levels of reality), 4 we then see that a large part of the work of verification has not been done, that it is extremely difficult to do, and that it is necessarily confined to very precise and limited influences, capable of being subjected to experimental methods. This is the case, for example, of numerous researches carried out on the differing perception of colours by subjects speaking different mother tongues. 5 In this way the apparently general and profound idea of an intimate relationship between language and world-view seems to dissolve before our eyes into small, modest, irritating laboratory researches. Unfortunately, it is on these very researches that our assessment of the general importance of the thesis is often presumed to depend; they supposedly bear the burden of its verification, deciding whether we are to accept it or discard it as a mere hold-over from speculative philosophy. Clearly there is something wrong here, and wrong at a deep level. But before expressing this discontent in a clearer form and looking to see if there are different lines of research which may attenuate it, it would be as well to go through, more thoroughly, the destructive criticisms to which the thesis of linguistic relativity has been subjected. A démystification of both the ideology of linguistic relativity and the ideology of its neo-positivist critics, together with certain aspects of the separatistic sciences they practice, can only be attempted after we have better understood the arguments and doubts on both sides.
* Fishman (1960), e.g., distinguishes four levels singled out by the relationships between semantic or lexical characteristics and grammatical characteristics with linguistic data and non-linguistic data. 5 See, e.g., Lenneberg and Brown,'A study in language and cognition' (1954); Landar, Ervin, and Horowitz, 'Navaho color categories' (1960); Lenneberg, 'The relationship of language to the formation of concepts' (1961) and Biological foundations of language (1964).
11 BRIEF DISCUSSION O F THE QUESTION O F TRANSLATING
Neither Whorf nor other upholders of linguistic relativity seem to have given all the necessary importance to the fact, apparently so simple, that after all WE DO SUCCEED IN TRANSLATING. If we compare — not just French with Italian, or Danish with Norwegian, or even Spanish with Polish — but even two very different languages such as an Amerindian language with any European language, we do obtain, even if at times with great difficulty, the result of transferring identical or at least extremely similar propositions from one to the other. Now, were such propositions not sufficiently similar to be subjected to a decree of 'sameness', one could not even make the comparisons on which the whole thesis stands. Or else, by translating we would have already changed the first propositions (expressed in the original language) into the second ones, with the result that at the end of the process we would, in fact, be comparing not two different objects with the aim of discovering their similarity, but rather two specimens of the same object, the similarity, and indeed in this case, the identity of which would have been there waiting for us throughout the whole operation. In their study on the Navajos Kluckhohn and Leighton write: almost anything which can be said in Navaho can be said in English and vice versa, though a translation which gets everything in may take the form of a long paraphrase which sounds strained and artificial in the second language (275). I interpret in this way: in the process of decoding and recoding not all the decoded material can be recoded DIRECTLY; a part of it must
54
BRIEF DISCUSSION OF THE QUESTION OF TRANSLATING
be recoded in the form of a metalinguistic description of a codification belonging to the language being decoded, which thus remains partly as an object-language INSIDE the recoding language. Codes as such cannot reproduce each other exactly because their structures are diverse, or because their vocabularies do not overlap precisely everywhere, or for both of these reasons. But this fact is not an exception. We continually put forward foreign terms, locutions, and constructions and explain in our own language what they mean. This does not make them incomprehensible to us. Every culture is capable of being informed IN its language as to the limitations of its own experiences, including those OF the language itself; and it is also possible to complete any language and thus any culture with certain aims in view, abandoning the level of mere meta-linguistic description of what takes place in another language to move to that of the re-elaboration, or, at least, the correction and enlargement of the very code which is being used. Had this not happened, modern science and the connected technical and non-technical awareness of its results and of its general significance (as is generally met with in 'educated people') would not have spread all over the planet. Today modern scientific treatises are published in Chinese and in Hebrew. And as the Navajos, unlike nearly all the other surviving Indian tribes in the two Americas, for reasons which have nothing to do with their language as such, are still fairly numerous and compact today, it may happen that eventually such treatises will also be written in Navajo. If the linguistic universes were mentally incommensurable, none of this would be possible; indeed, in principle, a mentally incommensurable linguistic universe would be simply beyond our ken. This would look like a solipsism transferred from the individual to the linguistic community. Instead, precisely what Whorf and the others have done has been to give us, in English, linguistic universes very different from our own; were their own theses completely correct, they could never have done this. So to some extent what happened is that a kind of total ethno-linguistic separatism was being maintained in the same breath in which, to maintain it,
BRIEF DISCUSSION OF THE QUESTION OF TRANSLATING
55
it was shown to be nonexistent. From this point o f view, the very methodological precautions of a Whorf (which can be glimpsed here and there in the passages cited, but are still more evident to those who study his writings) can be interpreted as defences and complications which prevented him from seeing certain fundamental difficulties of his approach. Two main points emerge and must be clearly distinguished: (1) two languages can differ so greatly that their codes are not able to reproduce each other directly; (2) nevertheless we always, or nearly always, succeed in RECONSTRUCTING one language in another, and this is done either by describing the first metalinguistically or by expanding the second. It is at this point permissible to ask oneself the real extent to which (1) impedes (2) (what is the value o f the phrase 'nearly always'?), as well as the extent to which (2), methodologically and in p r i n c i p l e , r e d u c e s ( 1 ) t o A MERE DESCRIPTION OF THE DIFFERENCES
IN QUESTION, that is, to a description from which nothing can be drawn as to diversity of world-view, perception of things, content of thought, and forms of thought itself. In this dialectic the ambiguous (or, as previously said, 'eel-like') character o f the thesis o f linguistic relativity comes to the fore, or at least the ambiguity of many o f the principal propositions in which it consists. Such propositions either refer us back to reproductive capacities inherent in every language and largely independent o f its superficial structure, or confine themselves to describing such differences o f superficial structure — but instead of insisting on them as obvious they make o f them the bearers of mysterious messages regarding deeper levels.
12
DIFFERENT THOUGHTS A N D DIFFERENT LANGUAGES
But what happens is that when we start to look for what these 'messages' situated at a deeper level may be, we find that there are both different thoughts 'in' the same language, and the same thought 'in' different languages. Let us look at the two cases separately. 1 Completely different conceptions, world-views, and philosophies have developed within the same linguistic culture, often cohabiting cheek by jowl 'inside' the same language. Had this not happened, the history of ideas and of thought in general would have no meaning and would indeed lack an object of its own. By affirming this we do not contradict the criterion that linguistic relativity concerns the relatively constant structure and the relatively common usage of a language rather than the doctrinal elaborations which are made in it. Here the discussion is centered on the idea that such structure and such usage can hinder some conceptions and favour others. Linguistic relativity regards, we repeat, the 'machine' of a given language as compared with at least one other language. We 1 I mention here collectively in chronological order some other works to bear in mind: Vygotsky, Thought and language (1934); La Barre, The human animal (1954); Gleason, An introduction to descriptive linguistics (1955; significant and almost unexpected, in the second edition [1965], the addition of some precautionary judgements in favour of Whorf); Bertalanffy, 'An essay on the relativity of categories' (1955); Brown, 'Languages and categories' (1956); O'Connor, review of Whorf's Language, thought, and reality (1958); Gastil, 'Relative linguistic determinism' (1959); Diamond, 'Anaguta cosmography: the linguistic and behavioral implications' (1960); Quine, Word and object (1960); Cohen, The diversity of meaning (1962, 1966); Language in culture and society (1964); Landar, Language and culture (1966); Deese, The structure of associations in language and thought (1966).
DIFFERENT THOUGHTS AND DIFFERENT LANGUAGES
57
mention this once again to avoid that the thesis be interpreted as a sort of philological or stylistic study of the language of a particular literary school or school of thought. Once this point is clear, the POSSIBILITY that various schools of thought and of literature reach expression in this or that language, are communicable and open to development in it — this possibility continues to constitute an object of enquiry which concerns the language as such. Thus we shall say not only that, for example, the conceptions contained in Galileo's Nuncius Sidereus were profoundly different from those in the contemporary Disputatio de Coelo by Cesare Cremonini; we shall also say that the Latin in which they were written was NOT different; and we shall add that the Dialogo dei massimi sistemi certainly did not contradict the Nuncius for having been written in Italian rather than in Latin. We shall speak not so much of the different philosophical schools in Denmark in the last century, but rather of the fact that Kierkegaard has managed to communicate to us his metaphysical anguish regardless of the decadence, already marked in his day, of the subjunctive mood in Danish, and this in spite of the fact that the subjunctive mood is precisely the one which ought to express doubt and uncertainty on the part of the speaker (Feuer 1953). In the same way we can observe that Hegel has taught us Becoming notwithstanding the periphrastic heaviness of the German gerund; or that Bergson introduced intuition as against the intellect, and duration as against mechanistic time in spite of the spatio-temporal structure and substantival-adjectival character of the French language, through the meshes of one of the most rigid syntaxes of Europe. The theory of physical relativity was developed in German, and the theory of linguistic relativity itself in German and in English; that is, within typical European languages, and by scholars who spoke them as native speakers. The tendency towards double or multiple negation, notoriously present in some languages and absent in others, does not seem to have brought forth the use of plurivalent logics by the speakers of those languages where it is endemic. As Feuer observes, with a rather gloomy shrewdness, in his 1953 essay against linguistic relativity, the widespread usage in American
58
DIFFERENT THOUGHTS AND DIFFERENT LANGUAGES
English of double or multiple negation has by no means placed the American proletariat in a dialectically negative position with regard to capitalism. 2 Evidently, if the structure and vocabulary of a language limited the thoughts that can be formed in it, the aforementioned developments would not have taken place in their respective languages; or at any rate they would have taken place much more slowly and with much greater difficulty. On the other hand, similar conceptions are expressed in different languages; this is a complementary fact. To provide evidence for it any manual of the history of philosophy will suffice: of Western philosophy, as long as we keep to European languages together with Arabic and Hebrew; but also of Chinese and Indian philosophy, with which other still more different languages are included. That the languages of the red and black races have not developed real and proper philosophies is PERHAPS a fact which we should take into account. However, they have obviously developed conceptions of the world which are often very rich and precise; and there is a literature which includes experiences like those of white philosophy teachers in Africa, of missionaries more or less all over the world, and of the scholars of ethnolinguistics themselves. 3 As Mohrmann explains, Christianity created a Latin expressly for its own specific purposes. 4 The salient feature is not so much that those purposes required a new Latin, but that this new Latin could have been formed. A classic example of the development of a philosophy in languages which are fairly different from the one in which it originated is the example of the Judaic, Arabic, and Latin 2
Feuer, 'Sociological aspects of the relation between language and philosophy' (1953), p. 88. 3 For a first and precursory outline, see Basson and O'Connor, 'Language and philosophy: some suggestions for an empirical approach' (1947); for a revaluation of the autochthonous African contribution to culture, Davidson, Black mother (1961; Italian translation, 1966) and Fanon, Lesdamnésde la terre (1961; Italian translation 1962). Expressly dedicated to ethnolinguistics is the journal Anthropological linguistics; but one should also see at least the linguistics journals Word, Language, and International journal of American linguistics, and the journals of anthropology like The American anthropologist and The Southwestern journal of anthropology. 4
Mohrmann, Latin vulgaire, latin des chrétiens, latin médiéval (1955), passim.
DIFFERENT THOUGHTS AND DIFFERENT LANGUAGES
59
Aristotle. To express in a Semitic language a metaphysic originally founded on the Greek language — beginning with the fact that in classical Hebrew there is no single abstract verb corresponding to s Ivou and its derivatives, and that common noun clauses are expressed without copula — notable difficulties of syntax and terminology had to be overcome. But such difficulties WERE overcome, and it is quite legitimate to study the development of Aristotelian thought in Hebrew, Arabic, and Latin. In the essay on categories of thought and of the language which we have already mentioned, Benveniste states that the Aristotelian categories are conceptual projections of linguistic conditions objectively present in the Greek language; and he proceeds with an exquisitely 'Whorfian' comparison between the presence in Greek of a very broad notion of being, which made possible the development of Hellenic metaphysics, and the absence of such a notion in a completely different natural language, taken as a pole of comparison: the Ewe language spoken in Togo. In Ewe there are five different verbs for translating 'to be'; but il ne s'agit pas d'un partage d'une même aire sémantique en cinq portions, mais d'une distribution qui entraîne un aménagement différent, et jusque dans les notions voisines .... À l'intérieur de la morphologie ou de la syntaxe ewe, rien ne rapproche ces cinq verbes entre eux. C'est par rapport à nos propres usages linguistiques que nous leur découvrons quelque chose de commun. Mais là est justement l'avantage de cette comparaison 'égocentriste'; elle nous éclaire sur nous-mêmes; elle nous montre dans cette variété d'emplois de 'être' en grec un fait propre aux langues indo-européennes, nullement une situation universelle ni une condition nécessaire.5
5
Benveniste, 'Catégories de pensée et catégories de langue' (1954), p. 428 (1966: pp. 72-73): 'It is not a question of a subdivision of the same semantic area in five portions, but rather of a distribution which involves a different arrangement, and that even within adjacent notions . . . Internally to Ewe morphology and syntax, nothing brings these five verbs together. It is in relation to linguistic usages of our own that we find something in common in them. But this is just the advantage of such an 'egocentric' comparison: it reveals us to ourselves; in this variety of the use of 'being' in Greek, it shows us a fact proper to Indo-European languages, and not at all a universal situation or a necessary condition'.
60
DIFFERENT THOUGHTS AND DIFFERENT LANGUAGES
It seems one should reply to this that if the Ewe-speaking West African populations developed sufficiently to provoke an interest like that which arose in Judaic, Arab, and medieval European circles for Aristotelian metaphysics, and did it in an autonomous way, that is without resorting to European languages for substitutes, then the syntactical and terminological difficulties of Ewe could also be overcome. After all, those populations believe or used to believe in Mawu, a creator-being thought of as unitary, and distinct from the trowo, which are clan, personal, and family genii. In these beliefs, thinking about and verbalizing their objects, these people use categories like those of unity and plurality, existence and permanence, which would offer the instruments both necessary and sufficient to put together just that generalised notion of being which Benveniste does not find at the level of linguistic structures. On the other hand, if there is a group of languages which could have made one think of a type of philosophy based on the existence of immutable entities, not subject to the variation of experience, this is the Chinese group, anchored as it is to the fixity of the characters instead of being prone like Western languages to the variation of alphabets and of their phonetic interpretation. But the correlation does not stand. A philosophy of immutable entities exists in Chinese, just as it does in Latin, French, or German. If one really wishes to point out a background philosophy of the Chinese tradition, in the sense (and with the gross lack of specificity) in which one says that rationalism and idealism 'are' German and empiricism 'is' English, what one finds is, if anything, a philosophy of common sense. But the reasons for this are not linguistic: indeed, during the last decades in China we have seen dialectics take the place of common sense or become integrated with it while the Chinese languages have remained substantially the same. At any rate, the important thing to be noted here is that all types of philosophy have been developed in China in an autochthonous manner and perfectly expressed both in writing and, one must suppose, in oral discussions. 6 6
Forke, Die Gedankenwelt des chinesischen Kulturkreises (1927); Levenson, Confucian China and its modern fate, 3 volumes (1963-1965).
DIFFERENT THOUGHTS AND DIFFERENT LANGUAGES
61
It must be recognized that both in the case of dissimilar philosophies, conceptions, etc. in the same language, and in that of similar philosophies in different languages, there have been in many instances special elaborations or even forced and unnatural modes of expression. Attempts have been made to infer from this that, ORIGINALLY, certain conceptions and world-views, forms of thought, philosophies, etc. were better suited to some languages than to others; whereas later, with the mingling of cultures or even simply with the formation of a culture, the relationship between a language and thought became more and more tenuous. It then became possible to think and to express in a given language thoughts which did not originate in it. We shall come back to this point.
13 SELF-EXTENSIVE POWER O F LANGUAGE
One of the fundamental hypotheses of linguistics is that in every natural language, with the possible exclusion of the most rudimental ones (situated at a much lower level than those of populations usually called 'primitive'), anything can be expressed and communicated. 1 Given that (1) a language, considered objectively in a particular moment of its development as a closed system, may not yet have been used to express certain thoughts which later on will be expressed in it (this is the case of Hebrew, Arabic, and Latin before Aristotelian metaphysics were expressed in them; it is also the case of English before Whorf expressed the structure of Hopi in it); and that (2) it is after all possible to translate, even if we do have to combat certain difficulties (chapter 11): it must be argued that every language possesses a generative and self-extensive power, or better that it can be subjected to such power. The position of the language in the process of generation and self-extension can be clarified as follows. Into the general notion of language [langage] flows not only the notion of the language in particular [langue], but also that of COMMON SPEECH, that is — the techniques, intersubjective, superpersonal, collective, communitary, in short, those that are in various ways common to all the speakers — with which we express ourselves and communicate when we use a language. If in opposition to a language, the social nature of which is recognised by all, there were only INDIVIDUAL speech, then the fact that we all learn to use that language in the same way would be incomprehensible; and indeed it would remain a mystery how all 1 Sapir, 'The grammarian and his language' (1924); in Selected writings, see pp. 153-154.
SELF-EXTENSIVE POWER OF LANGUAGE
63
the various individual 'speeches' originally agreed to produce something in common, something collective-right-from-the-start. (The fact may be that the notion of individual speech reflects, in a mystified form, the specific work which is always to be found in the production of linguistic use-values. But a distinction among the various types of linguistic work falls outside the scope of the present discussion.) Language [le langage] is the dialectical sum of a language [une langue] and common speech. As such, like every other human activity and contrary to what Saussure seems to have maintained, it is in principle perfectly knowable. The generative and selfextensive power of a language is therefore not a property inherent in it, but comes from the fact that we use it when we speak, and in this use of it, we bend it to our purposes. Language-in-general is everything that is used for linguistic production. In this production, as in any other, we can distinguish (1) a constant capital, which is the language [langue (made up of linguistic materials, instruments, and 'money'), (2) the linguistic working processes constituting common speech, and (3) a variable capital constituted by the linguistic workers, that is the speakers. The generative and selfextensive power usually attributed to the language as such [langue] is thus a characteristic of language-in-general, more or less as the growth of constant capital is in reality a characteristic of production viewed as a whole. But in linguistic production, as in material production, it can happen, and indeed it usually does, that the constant capital takes on a sort of apparently autonomous, monstrous life of its own, subordinating to itself those expenders of linguistic labour power, without whom it could never have formed nor could it continue to exist. Using the complex situation which is only outlined here as a background 2 , it seems that certain other limitations inherent in the thesis of linguistic relativity can be singled out, as well as the reasons for which the thesis continues to exercise a kind of sub2 Here I must refer to my volumes of 1961 and 1968 and to my essay 'Significato, ideologia e realismo artistico' (1967). [Forthcoming are Semiotica e ideologia and Linguistics and economics].
64
SELF-EXTENSIVE POWER OF LANGUAGE
terranean fascination of its own. We shall come back to the latter in chapter 15, but let us look now at the former. Disregarding the generative) and self-extensive power of language, the supporters of linguistic relativity have concentrated their attention on languages as objective structures, as autonomous systems which are completely or very predominantly immobile, and have been realized once and for all. At a more radical level they have sundered the language [langue] from common speech. They too saw speech as something individual; their merit has been that of discerning the ways in which it is subject to the machinery of the language [langue] rather than prevailing over it. A fortiori the idea of language [langage] as the sum or synthesis of the language [langue] and common speech could not even arise. In this way the relative permanence of the constant capital, that is of THE language, within the process of linguistic production was extrapolated and took on a role which was made artificially fundamental. A simple way of saying this in that the notion of a language replaced the notion of language-in-general [langage]. A part was put in the place of the whole, and what is more, a static part in the place of a dynamic whole; the materials, the instruments, and the money were invested with functions and properties which by rights belong only to the work which is done with them. But the obstacles in translating from one language to another, and the even radical structural differences to be found when certain groups of languages are compared, are all things which pertain to languages [langues] and definitely not to speech, and therefore not to language [langage], to linguistic production, either. By putting the language [langue] in the place of language [langage], the NEW WORK, which can be done with that very same constant capital by speech, was in effect neglected; and, the more so, the modifications, which can then be made on the constant capital so as to better adapt it to the new work to be done, were ignored. On the other production: to eat fruit is not even when we
hand, the USE OF PRODUCTS is not, sic et simpliciter, use automobiles is not to produce automobiles, to to look after orchards. The difference is apparent drink a glass of water without having ourselves
SELF-EXTENSIVE POWER OF LANGUAGE
65
filled the glass. He who uses products is not producing them. 3 The sense in which by speaking we reproduce linguistic products is an attenuated sense; the sense in which such products are consumed is similarly attenuated. The supporters of linguistic relativity have always idealistically believed that every speaker recapitulated in his own mind the production of linguistic products, whereas he is in fact simply USING them. In this way they have postulated mental processes which accompany the use of the constant linguistic capital and indeed which reproduce its objective structure in the mind of the speaker. Such processes would necessarily repeat, within each individual, the grand and toilsome social processes by means of which language [langage] has developed over hundreds of thousands of years, accumulating and becoming organized in the superindividual form of a language [langue] i.e. of a constant linguistic capital. Now, nobody wants to deny unrestrictedly the existence of processes which may be called mental. What we do deny is that they exist as SOMETHING WHICH ACCOMPANIES the use of a language: THIS use is ITSELF the mental process. Therefore not even an influence of the use on thought obtains. Whorf's Hopi friend, like any other speaker, uses, and consumes while partially reproducing, products which are already present in his language. It is by studying production, that the specialist Whorf discerned in it so many distinct operations. But there is no production within the speaker (the production took place a very long time ago, and is a question not of individual but of social processes); and the influence which Whorf thought he had perceived does not exist. To recapitulate, the thesis of linguistic relativity, on the one hand, stands on a splitting of the language [langue] from the rest of language [langage], which leads it to ignore the self-extensive power of language itself; while on the other hand it mentalizes the use of the language, that is, it confuses, in bourgeois fashion, the use of the already existing linguistic capital with its production, and attributes, to the system of products, properties which by rights belong only to labour. 3
These statements do not deny the fact that, at a more complex level, consumption and production are identical. For the threefold sense in which this happens see Marx, Zur Kritik der politischen Ökonomie, 'Einleitung von 1857' 2.a, pp. 622-626 (Italian translation, pp. 176-182).
14 SOCIAL TOTALITY AND EXTRA-LINGUISTIC CULTURAL FACTORS
The supporters of linguistic relativity do not seem to have taken sufficient account of CAUSAL DEPENDENCIES OTHER THAN THE LINGUISTIC ONE. Even in the cases where the linguistic differences under examination may be striking, it remains to be seen if and to what extent contributions to the differences in thought may have been made by (1) geographical, ecological, and biological factors, and (2) extra-linguistic cultural factors. The factors of the first group are obvious: two different environments necessitate different reactions and require different modalities of expression. But the EXTRALINGUISTIC CULTURAL FACTORS are of decisive importance. By this I mean all those factors which are not part of the language as a system and of its use; that is, material production, customs, and institutions in the widest sense. It is important to note that 'extralinguistic' here does not in any way mean 'making no use of symbolic processes', or 'which are without meaning', or 'not susceptible to linguistic or any other type of representation'. On the contrary, from another point of view (which is moreover complementary and parallel to that of those who study languages as such) one can very well say that production, customs, and institutions also constitute special 'languages' and that man expresses himself and communicates by means of them as well.1 1 Here one should see KINESICS, or the study of gestures and movements; PROXEMICS, or the study of spatial relationships; and, in general, all the new researches on non-verbal communication and on the codes it uses. A wellinformed introduction to semiotics is Eco's recent volume La struttura assente (1968); the collective volume Approaches to semiotics (1964) is still basic.
SOCIAL TOTALITY AND EXTRA-LINGUISTIC CULTURAL FACTORS 67
Now we are not saying that a man like Whorf would deliberately neglect the extra-linguistic factors; or on the other hand that he contented himself with correlating them to the linguistic factors, taking for granted, for example, a correspondence between a certain linguistic structure and a certain social institution. Indeed we have a number of precise statements of his on this. 2 But his position on this question remained ambiguous and incomplete (we should remember that he died prematurely, while his thought was still fully in the course of development). A good way of illustrating this point is to cite an objection raised by Claude Lévi-Strauss.3 According to Lévi-Strauss, when discovering correlations between a language and a culture Whorf showed that he was 'much less demanding' about the second term of the correlation; that is, while the linguistic data of which he made use were highly elaborated, the cultural data were inferred at an empirical level, without elaboration. The correlations are thus established between 'objects which belong to two very distant levels', and indeed 'which are not of the same nature'. To this I think we ought to reply that Whorf was working inside the language and deducing from it (whether rightly or wrongly) certain differences in mentality; only secondarily did he advance correlations. And when he did, it was not his intention to correlate the language and the culture, but, if anything, the language and thought (in one of the senses indicated in chapter 10, under a). What seems to have been the object of Whorf's more mature essays was not ethnological variation, but rather the breaking up or differentiation of the a priori from language to language. From this angle, Lévi-Strauss's argument boomerangs — Lévi-Strauss concentrated his attention on what was of greatest professional interest to him, comparing objects which belong to distant levels, and thus attributing to Whorf an ingenuity and oversimplification which no scrupulous reader would acknowledge. Nevertheless, Lévi-Strauss's objection does hint at a fundamental line of criticism. Not only is it unclear in what sense the relation2 Whorf, 'The relation of habitual thought and behavior to language' (1939); in Language, thought, and reality, see pp. 138-139 and note 1 on p. 139. 3 Lévi-Strauss, Anthropologic strutturale (1958), pp. 83-84 and p. 97.
6 8 SOCIAL TOTALITY AND EXTRA-LINGUISTIC CULTURAL FACTORS
ship between the language and thought is to be understood (whether causal, conditioning, concomitant, or whatever); not only is the language [langue] detached from speech and put in the place of language [langage]; one cannot even understand how it is permissible to separate language-and-thought from everything else. The fundamental logic of linguistic relativity can perhaps be summed up in in the following five points or steps: (1) Splitting of the language [langue] from language [langage], that is, of the constant linguistic capital from the rest of linguistic production. This has been discussed in the preceding section. (2) Extrapolation of the language (already split off in the above manner) from the totality of which it is part together with all the rest of the life of the community, that is from other activities and institutions. It should be noted that up to this point we could conceivably be dealing with operations which by themselves are neutral and legitimate. Every research must after all delimit its field. Language lends itself particularly well to such delimitation. Whoever declares that he is concerned, for example, with Rumanian or Navajo, will be immediately understood; his work will be placed within the area of fairly precise and defined traditions, while the techniques he will use and the type of conclusions he is liable to arrive at can be largely foreseen. (3) The establishment of a relationship, if not always directly causal, at least to some extent conditioning or of concomitance, between the language thus extrapolated and the rest of that totality. It should be noted that Whorf (we continue to consider his case paradigmatic) did not confine himself to searching empirically for this or that system of relationships, clearly determining or delimiting it, which is in principle always possible (examples: relationships among the language of Egyptian priests, their ceremonies, and their social position; relationships between the formation of the Eleatic notion of being and the rise of money as the universal equivalent, as in Thomson; 4 relationships between Eskimo terminology 4
Thompson, Studies in ancient Greek society, 2 vols (1949and 1955): II, xiv, 6.
SOCIAL TOTALITY A N D EXTRA-LINGUISTIC CULTURAL FACTORS
69
for snow or Arabic terminology for the horse and their respective ecological environments; relationships between German syntax and classical idealistic philosophy; relationships among the spoken language, the written language, and the national literature in twentieth century Italian; and so forth). Whorf looked for a system of relationships which was global and necessitating — and this brought him to take the next step as well: (4) Re-totalization of the two separated items, that is, of the language as such, and of all the rest of the original totality. This third operation is no longer neutral and legitimate, because it gives rise to a separatistic fallacy. In fact, if one 'retotalizes' in this situation, he has already excluded all OTHER possible relationships; that is, he has excluded in advance all relationships which do not belong to that single, global system of relationships which has been set up between the two separated parts. As a result of these steps, one was faced with two sub-totalities, separated from each other, yet confronting each other, each closed in itself, but one of them invested with a causative or determinative power over the other. The type of system of relationships to be discovered was by now established: it could only be the system of the relationships obtaining between the sub-totality constituted by the language and the sub-totality constituted by 'everything else'. The direction could still be chosen: the language could influence everything else, or everything else could influence the language — were it not for the fact that at this point a further step was taken: (5) From the sub-totality designated as 'all the rest of culture', thought was extracted (variously defined as the fundamental categories taken as a whole, the content of the thoughts which are actually thought, world-views, philosophies, and the like); and it was made into the effect or 'determinee' of the other sub-totality. This choice was understandable in the simple terms of the linguistic interests of the researchers; but these interests seemed to receive a real confirmation in the linguistic immediacy of thought. To the extent to which they abandoned themselves to the consequences of the fourth and fifth steps, Whorf and the others were advancing a PSEUDO-LINGUISTIC hypothesis, that is a hypothesis
70
SOCIAL TOTALITY AND EXTRA-LINGUISTIC CULTURAL FACTORS
expressed in the language of linguistics but not verifiable with the means provided by that science. Naturally it is always possible to find causal, unidirectional or conditioning relationships, to say nothing of relationships of reciprocal-action, within the original totality of culture: beginning with the way in which work is distributed and ending with the way in which political power and religious cajolery are exercised. But they are relationships which must be studied one by one, as they appear to empirical and historical investigation; and not after having extrapolated one sector which, re-totalized, must take on the task of containing within itself the causes of the events which take place in the other sector. A re-totalization of this kind is nothing less than idealistic; the idea of the thing is substituted for the thing itself. If a foundation must be sought, then it must be sought in the direction of the minimum conditions which allow man to be what he is, and all these conditions must always be kept together. We are here referring to the modalities of the division of labour which originally allowed the tribe to survive. 5 Language is immediately present, but certainly not in the form of a constant linguistic capital, capable of being isolated from everything else, and made to determine nothing less than thought. If we want to study the way in which thought is determined in all its developments up to the point of including spontaneous and sophisticated world-views, we shall have to turn our attention to the sum total of economic, social, and cultural conditions. We shall find that what we describe as linguistic is, if anything, a part of their phenomenology.
5 Let us recall merely the chapter 'Anteil der Arbeit an der Menschwerdung des Affen' in Engels' Dialektik der Natur, pp. 444-455 in Werke.
15 LINGUISTIC RELATIVITY A N D ALIENATION
The thesis of linguistic relativity presents various other contrasting and complementary aspects. On the one hand we should recognize that it does indirectly put us on our guard against accepting any kind of metaphysical system or philosophical super-science which, being the expression of a single cultural tradition, eventually overwhelms all the others (or, if incapable of overwhelming, simply ridicules, misunderstands, or ignores them). In this sense the insistence on single natural languages as original and irreducible linguistic models of the universe is a positive feature. It is a methodological indication which is important for the relativization of our instruments of understanding and research, a working hypothesis for the rediscovery of structures which might otherwise escape our attention. On the other hand it must also be admitted that the more we strive to see in linguistic relativity a set of propositions susceptible to empirical verification, the more we make of it an obstacle to the expansion of knowledge. Here it seems to invite us to call a halt to research before the insurmountable barriers of national or tribal spirit implanted in the very instrument of understanding, the language as the common patrimony of its speakers. If languages were really and only isolated universes, then it would never be possible to trace them back to something common to them all. Already in this sense, a bourgeois ideology is expressed in the thesis of linguistic relativity. But here an additional mediation takes place: the ideology is projected into an object other than itself. According to this ideology there are interests and values to be
72
LINGUISTIC RELATIVITY AND ALIENATION
defended against the levelling progress of knowledge — but the values in question are not our own values; they are the values of others. This is a strange dialectic indeed. The basic attitude is idealistic, even if historicized and relativized enough to let us glimpse the presumption 'from below' of those who content themselves with analysing their own everyday language as the criterion of truth, and the presumption 'from above' of those who put forward a universal philosophical science. It is not without significance that the ideology on which linguistic relativity rests is projected and expressed precisely in certain generalizations drawn from the study of the American Indian languages. Both the invitation to better understanding them, to avoid overwhelming them metaphysically, and the defence of their Weltanschauungen from scientific progress manifest the guilt feelings of American whites for having savagely destroyed the Indians beyond any possible need, I do not say of expansion, but even of personal defence. If noticeable traces of the African languages of slaves imported into the United States still remained there, these languages too would certainly have been used to furnish examples in support of linguistic relativity. But no traces are there, for the excellent reason that the children of the slaves were taken away from their mothers so that they would not learn their own language and would have to accept only English as their mother tongue. So that the capital constituted by the slaves might be exploited without hindrance, it was necessary to alter their natures; and with a singular though hideous intuition this was done at the root, taking away from them even their mother tongue. By their insertion into English linguistic production their use in material production was enormously facilitated. Let us take up again what is perhaps the most convincing point of view in the neopositivistic criticism of the claims of linguistic relativity. It is a point of view in which the positivistic attitude has an air of promoting the objectivity which is sought after by historians. I am speaking of the argument that different thoughts and conceptions have been developed in the same language, and that the same thought and conception has found expression in
LINGUISTIC RELATIVITY AND ALIENATION
73
different languages. It is a fact: this is how things are. All the same, if we go beyond such a statement of fact, we can attempt to set in motion a considerably different line of discussion, using the same statement in a much wider field. To emphasize the relative neutrality of the language means to assert man's detachment from his mother tongue, and this eventually leads to the complex series of problems of LINGUISTIC ALIENATION. It is precisely by saying things like this — that IN SPITE OF the spatio-temporal structures and substantival-adjectival character of the French language, a man like Bergson was able to express in it his philosophy of intuition, of duration, and of creative evolution — that one denounces a disturbing lack of identification between the speaker and his language (besides stressing the self-generative power of language). It is here that the subterranean fascination of linguistic relativity resides. As we have seen, the self-generative power of language [langage] is certainly not a property inherent in languages [langues] as such. And yet it also concerns languages [langues] as constant capitals: the generative power of labour cannot but regard capital. For languages, this happens in the sense that they assume the capacity of lending themselves to any kind of communication and conceptual elaboration. This is one of the directions in which the difficult notion of linguistic money is to be sought. It happens that languages interpose between the speaker and the expression of his reactions to the world a sort of autonomous thickness. This is a way of affirming a sort of linguistic relativity the other way round, of indicating how the speaker can be taken from what ought to be his own world-view and brought over to other world-views which right from the beginning are no longer the view originally incorporated in his mother tongue. In other words, whereas the idealistic supporters of linguistic relativity assert that thought and world-view depend directly on the structures of the language as a system [langue], what is here being maintained is that the self-generative power of language [langage], as crystallized in the structures of a particular language [langue], confers upon the latter a power of distortion. The labour power of the speakers, that is the variable
74
LINGUISTIC RELATIVITY AND ALIENATION
linguistic capital, is then used to impart an unnatural movement to the whole of production. In this way the speakers become 'spoken'. But not only in the immediate sense, which is once again the one the idealistic supporters of linguistic relativity propose, according to which the speakers cannot speak except within the structures of their mother tongue and in this sense are always representing them so that their language speaks through them; but also in a sense, dialectically more complex, according to which the speakers do speak things not foreseen by the objective structure of the language they are in fact speaking and in which they express themselves. In this way a laceration is created within linguistic production. The total linguistic capital is handed down and accumulated without its having any further relationship with the human reality of the working process, that is, repressing in itself its own variable portion. The differences become blunted; any linguistic worker becomes part of a social mechanism — the great machine of the language which produces and reproduces anything and everything. Linguistic money has taken over. A linguistic surplus-value has been created which now has nothing to do with the interests of the workers, i.e. of the speakers. To the extent to which this characterization is acceptable, one can say that the ideology of linguistic relativity consists principally in a great illusion which is the bearer of both non-verbalized false consciousness and precise and complex ideological propositions — both of these accompanied by their respective false praxis. 1 It is the illusion of believing that the THEORETICAL exhibition of the structures of the various languages, that is, of their constitutive differences, is by itself in some way sufficient to remedy the state of linguistic alienation or at least to alleviate it. The exhibition of the concrete working operations which must or can be carried out with a language, the description of them in their component steps, but leaving aside the question of exploitation — this is supposed to hold the arcane power of preventing or alleviating exploitation itself. The root of this illusion lies in a fact which is systematically neglected 1 See on this ray essay 'Ideology as social planning' (1967), now in Language as labour and as trade (The Hague: Mouton, forthcoming).
LINGUISTIC RELATIVITY AND ALIENATION
75
by those farmers of facts who are the neopositivistic critics of linguistic relativity: THOSE OPERATIONS MUST AFTER ALL HAVE BEEN CARRIED OUT, OTHERWISE WE WOULD NOT HAVE THEIR PRODUCTS
us NOW. The work of which the idealist speaks is mystified work; but it is still work. If there is a product, there must have been the work which produced it. In the case of the American Indians the situation is particularly clear and convincing; not for nothing has it been in the study of their languages that the most fascinating formulations of linguistic relativity were formed. The whites have deprived the Redskins of their world-views; now they are trying to 'compensate' them by recognizing that these views were 'naturally' part of the fabric of their languages and constituted richer, better articulated, and more perfect constant capitals than ours. At the same time, linguistic relativity can in this way even involuntarily lend itself to the smuggling in of a subtle racist streak. The American Indians ARE different BECAUSE their languages MAKE THEM different. How wicked of us to massacre peoples so DIFFERENT FROM OURSELVES. The fact that use is not the same thing as production also plays a part here: the American Indians use refined linguistic products, but after all they personally are not the same Indians who originally invented them. Let not the merits of the fathers be visited overmuch upon the sons! The linguistic alienation of which we have been speaking is an alienation ab ovo. Our line of discussion is therefore one which in the first place concerns a general characteristic of human history. It is from the beginning that the speaker is forced to carry out an undifferentiated linguistic work, one that produces and reproduces linguistic goods which are extraneous to the real operations for reconstructing from the inside the language which he does nevertheless continue to speak. Thus right from the start, a transformation of linguistic producers into LINGUISTIC CONSUMERS is set in motion. More and more the USE of language takes the place of its PRODUCTION. The notion of linguistic use supplants that of linguistic work. Thus the linguistic worker lets himself be dragged on by the 'spontaneous' motion of the great machine of the lanBEFORE
76
LINGUISTIC RELATIVITY AND ALIENATION
guage. In a sense which is fundamental because it is constitutive of what is human, HE NO LONGER THINKS ABOUT WHAT HE IS DOING WHEN HE SPEAKS. (Obviously, in other senses he does continue to think about it; he has the possibility of doing so: for example, when he tries to describe an object with great accuracy; or wants to make himself understood by a particular interlocutor; or uses language with a very specific aim in mind; or when he is moved by an extra-linguistic interest, the satisfaction of which demands a particular use of linguistic techniques; or when he is a linguist or an ethnolinguist or a poet by trade and therefore attempts to achieve a definite linguistic result. As we have seen, reconstructing the internal fabric of a language as minutely as possible is typical precisely of the scholars of linguistic relativity.) In this way the speaker loses his contact and intercourse with nature and with other men; that contact and that intercourse which had originally presided over the formation of his language, and which had been deposited and were represented in it as specific productive operations and in their results, understood as use-values. Since that happened, the linguistic product has been handed down only as already-produced; and the model of the product is re-produced only for the purpose of allowing the consumption of its exemplars in order to aliment the system of production. The needs which language should satisfy — basically, that of being able to really express oneself and that of communicating and of being really understood within the division of labour — have moved into the background. There is a phase in which the child, with his marvelous spontaneity, tries to take hold of the machine of the language as a satisfier of needs, to put himself in the position of producing from it, every time ex novo, articulations which are in consonance with his experience of the world and of his fellows, to be himself as a speaker. Thus he plays with words, tampers about with them and mixes them up in various directions; he asks 'why?' about everything; and he realizes full well the internal human significance of the words he has learned. It is a phase which may last only a few weeks, for it manifests itself in that extremely delicate moment when the child has already mastered linguistic techniques sufficiently to
LINGUISTIC RELATIVITY AND ALIENATION
77
make himself understood, but has not yet been absorbed into the system of linguistic production. Then he is immediately dragged off, made to conform, and repressed, with the result of transforming him into a linguistic reproducer-consumer. 'You have to say it this way'; 'it's that way just BECAUSE'; 'you don't say that'; 'this thing is that other thing'; and so on. Many arguments against linguistic relativity are correct in the sense that they refute the idealistic illusion held by its supporters. They describe with separatistic pedantry the state of things as they are, that is, some aspect or other of the universal substitution of undifferentiated work for specific work. They are, typically, arguments of positivistic (or neo-positivistic) inspiration like those of Black, Feuer, Lenneberg, and, in general, of the new generation of American linguists. To understand the whole situation, one must recognize the validity of such arguments at their own level, realize that they are correct as far as description is concerned. But at the same time one must comprehend the, as it were, hidden motives of those who uphold linguistic relativity. To be rendered valid, these motives must be demystified. Here too, the road is that of turning upside down an idealistic conclusion. Only in this way can an attempt be made to raise the whole discussion to a higher dialectical level. One can assume that with the coming into contact of various languages certain phenomena occurred which were considerably different from those which could have occurred had every language continued to be produced in all of its specific richness. Languages, instead of cross-fertilizing each other, have at best reciprocally levelled each other out. Naturally, those who had the worst of this were the speakers of languages belonging to 'primitive 'societies, unable to resist the onslaught of civilisations more advanced than they were in the technology of material production, although not necessarily in that of linguistic production. In many cases the total capital of others has been attacked in its variable component: the method has been that of PHYSICALLY DESTROYING the speakers. ('He was killed because he SAID things that did not please the bosses.' In this sentence a relationship is stated which still holds true in 1968 and also among whites.) The immense wealth of the American
78
LINGUISTIC RELATIVITY AND ALIENATION
Indian languages regarding man-nature relationships, which was evidence of a human situation less alienated than ours, was destroyed by extermination of the Indians. We have just spoken of the children of the slaves. Just imagine if the bearers of techniques, ideologies, and religions founded upon a substantive-predicative and spatio-temporal vision of the world could have found it tolerable that some 'miserable' Redskins should see the world in a completely different way. But indeed the value of their diversity was not even felt. Hence — with the reawakening in the twentieth century of interest in diversity — the feeling of guilt in those who have approached the American Indian world with such historically belated understanding and admiration; hence the ideological desire to restore to the Indians of America, at least in theory, something that had been stripped from them with every possible means in the praxis of destruction, or at least the most intimate and familiar nucleus of their stolen patrimony — that is, all they knew how to do and transmitted to their children as original and autonomous linguistic workers. The variable capital having been almost entirely destroyed (without residue in the majority of cases, since a good part of the American Indian languages are now extinct), the attempt was made to revalue the corresponding constant capital, to bring the extinct workers back to life by re-evoking the wealth of their working processes. Thus those linguistic workers, who previously had been treated only as things, are now being used as ghosts. The solution to the antinomies of linguistic relativity can certainly not consist in new empirical research on the relationships between things like the vocabulary of colours and the actual perception of colours; nor in such research on things that are still more difficult to experiment like the temporal scansion indicated by the language and the actual temporal scansion to be found in the individual behaviour or in the institutions of the culture in which that language is (or was) spoken. Just these are the pseudoscientific pretensions of the neopositivists, who would now like to substitute a projection of Western separatistic techniques for the reconstruction of what has been destroyed. Nor yet can the solution consist in the idealistic reconstructions, laden with fallacies in their turn, of a Whorf or a
LINGUISTIC RELATIVITY A N D ALIENATION
79
Weisgerber (even if those of the latter are limited to the vindication of the vision of one specific people, his own — Weisgerber avoids the additional mediation of projecting one's own ideology onto other races, and under this aspect his glottocentrism is, at least, self-contained). Underneath the antinomies of linguistic relativity lies the whole set of problems regarding alienation and language. These problems are particularly acute in the so-called 'civilizations of the white race' which for thousands of years have been swollen with dogmatism and arrogance. These civilizations are the bearers of 'separate' sciences, that is of sciences each of which is isolated from the rest of life, and whose technical language is an 'abnormal' development of an often very small portion of the common language of the people. They have indiscriminately destroyed or ravaged all that they have found in their path, without paying the slightest heed to the point of view of others; they are by now dangerous to themselves and to the rest of mankind; and one still cannot foresee HOW, pervaded as they largely are by the most omnivorous of systems, they will manage to revise their basic attitudes, in themselves extinguishing the evil of misconceived otherness.
NOTE
The materials gathered here are partly the result of two seminars in semantics devoted to the problems of linguistic relativity and held for graduate students in philosophy at the Universities of Michigan (Ann Arbor) and Texas (Austin) in the first and second semesters of the academic year 1962-63; so this is the place to finally thank those young scholars, generally serious and eager notwithstanding the deficiencies of their historical and ideological formation, for the contributions which they made to the discussion. This essay has to some extent the appearance of an anthology, to a greater extent that of a summary (at times really rather schematic), and should be revised and brought up to date with reference to the numerous publications which have appeared in the meantime (this has been done here with regard to works published up to the spring of 1968). A thorough treatment of the subject should consider many topics which have here been omitted: for instance, to mention a few, a comparison between the theses of Whorf and Sapir and those of Weisgerber and Trier, an examination of contemporary research on linguistic universals and deep structures (which were merely hinted at in chapter 9), Whorf's connected distinction (later taken up by other scholars) between phenotypes and cryptotypes, the relationships between the language and perception, and in the sector of exemplification, the material drawn from languages other than the American Indian ones, especially the complex and exemplary question of Chinese. The bibliographical references have been proportionately pruned; but they retain a breadth considered sufficient to give the reader some idea of the state of development of the problems discussed. (Summer
1968).
BIBLIOGRAPHY
A. REFERENCES USED IN THE 1968 ESSAY Abbagnano, Nicola, Dizionario difilosofia (Torino: Utet, 1961). Actes du IVe Congrès de VAssociation internationale de littérature comparée (Fribourg, 1964), rédigés par François Jost (The Hague-Paris: Mouton, 1966), 2 vols. Section 8, 'Cosmopolitisme et langue nationale', is made up of the writings of Kushner and of Markiewicz. Aginsky, Burt W. and Ethel G., 'The importance of language universals', Word IV: 3 (December 1948), 168-172. Approaches to semiotics. Transactions of the Indiana University Conference on Paralinguistics and Kinesics, ed. by Th. A. Sebeok, A. S. Hayes, M. C. Bateson (The Hague: Mouton, 1964). Basilius, Harold, 'Neo-Humboldtian ethnolinguistic', Word VIII: 2 (August 1952), 95-105. Basson, A. H. and D. J. O'Connor, 'Language and philosophy: some suggestions for an empirical approach', Philosophy XXil: 81 (1947), 49-65; reprinted in Methodos V: 19 (1953), 203-221. Benveniste, Émile, 'Tendences récentes en linguistique générale', Journal de psychologie normale et pathologique LXXIV: 1 and 2 (janvier-juin 1954), 130-145; reprinted in Problèmes. — , 'Catégories de pensée et catégories de langue', Études philosophiques N.S. XIII: 4 (octobre-décembre 1954), 419-429; reprinted in Problèmes. — , Problèmes de linguistique générale (Paris: Gallimard, 1966) (NRF, Bibliothèque des sciences humaines). Bertalanffy, L. von, 'An essay on the relativity of categories', Philosophy of science XXII (1955), 243-263. Black, Max, 'Linguistic relativity: the views of Benjamin Lee Whorf', The philosophical review LXVIII: 2 (1959), 228-238. Bright, William, editor of Sociolinguistics, q.v. Boston studies in the philosophy of science, 1961-1962, ed. b y M . W. Wartofsky (Dordrecht: Reidel, 1963). Brown, Roger, 'Language and categories', Appendix (pp. 247-312) to A study of thinking, q.v. — , Words and things (Glencoe, III.: The Free Press, 1958). Brown, Roger W. and Eric M. Lenneberg, 'A study in language and cognition', Journal of abnormal and social psychology XLIX: 3 (July 1954), 454-462.
82
BIBLIOGRAPHY
Carroll, John B., 'Introduction' (pp. 1-34) to Language, thought, and reality by Whorf, q.v. — , Language and thought (Englewood Cliffs, N.J.: Prentice-Hall, 1964). Italian version by O. Battachi Codispoti, enlarged under the editorship of M. W. Battacchi with the title Psicologia del linguaggio (Milano: Martello, 1966). Casagrande, Joseph B., 'Language universals in anthropological perspective', in Universals of language (1953), 279-298. Chase, Stuart, Il potere delle parole, translation by G. Civiletti (Milano: Bompiani, 1966); see especially chapter X : 'Le parole e la visione del mondo' pp. 113-122 [the original Power of words was published in 1953]. Childe, Gordon, Il progresso nel mondo antico. L'evoluzione delle società umane dalla preistoria agli inizi dell'età classica, translation by A. Ruata (Torino: Einaudi, 1963). — , L'evoluzione delle società primitive, translation by F. Codino (Roma: Editori Riuniti, 1964). — , Società e conoscenza (Milano: Mondadori, 1962). Chomsky, N o a m , Syntactic structures (The Hague: Mouton and Co., 1957). — , Current issues in linguistic theory (The Hague: Mouton, 1964). — , Aspects of the theory of syntax (Cambridge, Mass.: The M.I.T. Press, 1965). — , Topics in the theory of generative grammar (The Hague-Paris: Mouton, 1966). — , Cartesian linguistics. A chapter in the history of rationalist thought (New York and London: Harper & Row, 1966). Cohen, Jonathan L., The diversity of meaning (London: Methuen, 1962). Cohen, Marcel, see Les langues du monde. Colorai, Eugenio, 'Critica filosofica e fisica teorica' (posthumous), Sigma (Roma) II: 4 and 5 (1948), 261-292. Davidson, Basii, Black mother. Africa: the years of trial (London: Gollancz, 1961). Italian translation by L. Felici, Madre nera. L'Africa nera e il commercio degli schiavi ( = P.B.E. 78) (Torino: Einaudi, 1966). Deese, James, The structure of associations in language and thought (Baltimore: John Hopkins Press, 1966). Devoto, Giacomo, Profilo di storia linguistica italiana (Firenze: La Nuova Italia, 1953). D e Waelhens, Alphonse, Existence et signification (Louvain: E. Nauwelaerts, 1958). Diamond, Stanley, 'Anaguta cosmography: the linguistic and behavioral implications', Anthropological linguistics II: 2 (February 1960), 31-38. Issue dedicated to the symposium 'Translation between language and culture', held in 1959). Directions in psycholinguistics, ed. by Sheldon Rosenberg (London: CollierMacmillan, 1965). Dorfles, Gillo, Nuovi riti, nuovi miti (Torino: Einaudi, 1965). Eco, Umberto, La struttura assente. Introduzione alla ricerca semiologica (Milano: Bompiani, 1968). ( Nuovi saggi italiani, 1) (English translation, The semiotic landscape [The Hague: Mouton, i.p.]). Eggan, Dorothy, 'Le rêve chez les indiens hopis', translated from English by M.-E. Voinot, in Le rêve et les sociétés humaines, 1967, q.v. (pp. 213-252).
BIBLIOGRAPHY
83
Elders, Léon, 'Les rapports de la langue et de la pensée japonaises', Revue philosophique de la France et de l'étranger CLVI: 3 (juillet-septembre 1966), 391-406. Engels, Friedrich, Dialektik der Natur, in Karl Marx u. Friedrich Engels, Werke (Berlin: Dietz, 1962), Band 20, S. 305-570. Italian translation by L. Lombardo-Radice (Roma: Edizioni Rinascita, 1955). New edition (Editori Riuniti, 1967). — , Der Ursprung der Familie, des Privateigentums und des Staates, in Marx u. Engels, Werke, Band 21, S. 25-173. Italian translation by D. Della Terza, L'origine della famiglia, della proprietà privata e dello stato, introduction and notes by Fausto Codino (Roma: Editori Riuniti, 1963). Ervin, Susan M.: see Landar, H. S. Essays in conceptual analysis, selected and edited by Antony Flew (London: Macmillan, 1956). Explorations in communication, An anthology, ed. by E. Carpenter and M. McLuhan (Boston: Beacon Press, 1960). Translated under the title La comunicazione di massa by R. Ambrosini (Firenze: La Nuova Italia, 1966.) Fanon, Frantz, Les damnés de la terre (Paris: Maspéro, 1961), preface by JeanPaul Sartre. Italian translation by C. Cignetti (Torino: Einaudi, 1962). Feuer, Lewis S., 'Sociological aspects of the relation between language and philosophy', Philosophy of science XX: 2 (1953), 85-100. Fishman, Joshua A.: 'A systematization of the Whorfian hypothesis', Behavioral science V: 4 (October 1960), 323-339. — , 'Who speaks what language to whom and when?', La linguistiquel: 2 (1965), 67-88. — , Review of Joyce O. Hertzler: A sociology of language, Word XLIII: 2, Part 1 (June 1967), 586-604. Flew, Antony, 'Introduction' to Logic and language, Second Series, q.v., pp. 1-10. — , 'Philosophy and language', in Essays in conceptual analysis, q.v., I, pp. 1-20. Forke, Alfred, Die Gedankenwelt des chinesischen Kulturkreises {Handbuch der Philosophie, Sonderausgabe) (München und Berlin: Oldenburg, 1927). Gastil, Raymond D., 'Relative linguistic determinism', Anthropological linguistics I: 9 (December 1959), 24-38. Gipper, Helmut, Bausteine zur Sprachinhaltsforschung (Düsseldorf: Schwann, 1963). (Sprache und Gemeinschaft, 1, series directed by Leo Weisgerber). Gleason, H. A. Jr., An introduction to descriptive linguistics (New York: Henry Holt and Co., 1955); revised edition (New York: Holt, Rinehart and Winston, 1965). Greenberg, Joseph H., Language universals (Janua Linguarum, Series Minor, 59), (The Hague-Paris: Mouton, 1966). Guchman ,M. M., 'È Sèpiri 'ètnografiòeskaja lingvistika', Voprosy jazykoznanija III: 1 (1954), 110-127. Hegel, G. W. F., Enzyklopädie der philosophischen Wissenschaften (1830), Neu herausgegeben von F. Nicolin u. O. Pöggeler (Hamburg: Felix Meiner, 1959). Trans, by B. Croce, Enciclopedia delle scienze filosofiche in compendio (Bari: Laterza, 1907).
84
BIBLIOGRAPHY
Hertzler, Joyce O., A sociology of language (New York: Random House, 1965). Hjelmslev, Louis, Prolegomena to a theory of language, transi, by F. J. Whitfield (Madison: The University of Wisconsin Press, 1963), revised English edition; the previous edition appeared in Baltimore in 1953; the original Danish text appeared in 1943; Italian translation and introduction by G. C. Lepschy: I fondamenti della teoria del linguaggio (Torino: Einaudi, 1968). — ,'Langue et parole', Cahiers Ferdinand de Saussure, II (1943), 29-44; in Essais linguistiques, q.v., pp. 69-81. — , 'Structural analysis of language', Studio linguistica I: 2 (1947), 69-78; in Essais linguistiques, pp. 27-35. — , 'La stratification du langage', WordX: 2 and 3 (August-December 1954), 163-188; in Essais linguistiques, pp. 36-68. — , 'Dans quelle mesure les significations des mots peuvent-èlles être considérées comme formant une structure?', Actes du huitième Congrès international des linguistes (Oslo, August 1957), (Oslo, 1958), pp. 636-654; in Essais linguistiques, under the title 'Pour une sémantique structurale', pp. 96-112. — , Essais linguistiques (= Travaux du Cercle linguistique de Copenhague, XII) (Copenhague: Nordisk Sprog- og Kulturforlag, 1959). Hoijer, Harry, 'Classificatory verb stems in the Apachean languages', International journal of American linguistics XI: 1 (1945), 13-23. — , 'Cultural implications of some Navaho linguistic categories', Language XXVII: 2 (April-June 1951), 111-120; reprinted in Language in culture and society, ed. by Hymes (1964), pp. 142-147. — , 'The relation of language to culture', pp. 554-573 in Anthropology today, ed. by A. L. Kroeber (Chicago: University of Chicago Press, 1953). — , editor of Language in culture, 1954, q.v. — , 'Semantic patterns of the Navaho language', in Sprache. Schlüssel zur Welt (1959), q.v., pp. 361-373. Horowitz, Arnold E.: see Landar, H. J. Humboldt, Wilhelm von: Gesammelte Schriften, herausgegeben von der Königlich Preussischen Akademie der Wissenschaften (Berlin: B. Behr's Verlag, 1903-1936). Seventeen volumes: the first fifteen edited by Albert Leitzmann (1903-1918); volumes sixteen and seventeen (Politische Briefe) edited by Wilhelm Richter (1935-1936). In the more handy Werke in fünf Bänden (Berlin: Rütten & Loening, Ausgabe für die D.D.R., herausgegeben von Andreas Flitner und Klaus Giel, 1960-1966), see Band III (1963) : Schriften zur Sprachphilosophie, among which we recall: 'Über den Dualis' (1827), 'Über die Verschiedenheiten des menschlichen Sprachbaues' (1827-1829), and 'Über die Verschiedenheit des menschlichen Sprachbaues und ihren Einfluss auf die geistige Entwicklung des Menschengeschlechts' (18301835). Husserl, Edmund, Logische Untersuchungen (Tübingen: Max Niemeyer, 1968: reprint of the 1913 edition). Hymes, Dell, editor of Language in culture and society, 1964, q.v. — , 'Two types of linguistic relativity (with examples from Amerindian Ethnography)', in Sociolinguistics, 1964, q.v., pp. 114-167.
BIBLIOGRAPHY
85
Kaplan, Bernard, see Werner and Kaplan, Symbol formation. Kluckhohn, Clyde and Dorothea Leighton, The Navaho (New York: Doubleday Anchor Book, 1962; revised edition by L. H. Wales and R. Kluckhohn; first published in 1946). Kushner, Eva, 'Yvan [I wan] Göll: deux langues, une âme', Actes du IVe Congrès de l'Association internationale de littérature comparée, 1964, q.v., pp. 576587. La Barre, Weston, The human animal (Chicago and London: The University of Chicago Press, Phoenix Book, 1954). Landar, Herbert, Language and culture (New York: Oxford University Press, 1966). Landar, Herbert J., Susan M. Ervin, and Arnold E. Horowitz, 'Navaho color categories', Language XXXVI: 3, Part 1 (July-September 1960), 368-382. Language in culture, Proceedings of a Conference on the interrelations of language and other aspects of culture (Chicago, March 1963); edited by Harry Hoijer, The American Anthropological Association 56: 6, Part 2, Memoir No. 79, (December 1954). Language in culture and society, A reader in linguistics and anthropology, edited by Dell Hymes, foreword by A. L. Kroeber (New York, Evanston, and London: Harper and Row, 1964). Les langues du monde, by a group of linguists, directed by Antoine Meillet and Marcel Cohen (Paris: H. Champion, 1925, 2nd ed. 1952). Lee, Dorothy [Demetracopoulou], 'Conceptual implications of an Indian language', Philosophy of science V: 1 (1938), 89-102. — , 'Linguistic reflection of Wintu thought', International journal of American linguistics X (1944), 181-187; reprinted in Freedom and culture, q.v., pp. 121-130. — , 'The conception of the self among the Wintu Indians', Journal of abnormal and social psychology XVI: 3 (1950), 538-543; reprinted in Freedom and culture, pp. 131-140. — , Freedom and culture (Englewood Cliffs: Prentice-Hall, 1961), a collection of fifteen essays. Leighton, Dorothea, see Kluckhohn, Clyde, The Navaho. Lenneberg, Eric H., 'Cognition in ethnolinguistic', Language XXIX (1953), 463-471. — , 'The relationship of language to the formation of concepts', Summary of oral presentation, Nov. 1961, in Boston studies in the philosophy of science, (1963), q.v., 48-54. — , 'A biological perspective of language', pp. 65-88 in New directions in the study of language (1966), q.v. — , Biological foundations of language, with appendices by Noam Chomsky and Otto Marx (New York, London and Sydney: John Wiley & Sons, 1967). Lenneberg, Eric H. and Roger W. Brown, 'A study in language and cognition', see Brown (1954). Levenson, Joseph R., Confucian China and its modern fate (Berkeley and Los Angeles: University of California Press, 1963-1965). Vol. I: The problem of intellectual continuity (1963); Vol. II: The problem of monarchical decay (1964); Vol. Ill: The problem of historical significance (1965).
86
BIBLIOGRAPHY
Lévi-Strauss, Claude, Anthropologie structurale (Paris: Pion, 1958). Italian translation by P. Caruso (Milano: 11 Saggiatore, 1966). Logic and language, an anthology of essays by various authors, edited by A.G.N. Flew (Oxford: Blackwell, 1st series 1951, 2nd series 1953). Luther, Wilhelm, Weltansicht und Geistesleben. Versuch einer wissenschaftlichen Grundlegung der philosophischen Sprachanalyse an Beispielen aus der griechischen Geistesgeschichte von Homer bis Aristoteles (Göttingen: Vandenhoeck & Ruprecht, 1954). Macnamara, John, Bilingualism and primary education. A study of Irish experience (Edinburgh: University Press). Markiewicz, Zygmunt, 'La servitude sans grandeur d'un écrivain bilingue', Actes du IVe Congrès de VAssociation internationale de littérature comparée (1966), q.v., pp. 588-594. Martinet, André, 'Arbitraire linguistique et double articulation', Cahiers Ferdinand de Saussure XV (1957), 105-116. — , Eléments de linguistique générale (Paris: Armand Colin, 1961 ). Italian translation by G.C.Lepschy Elementi di linguistica generale (Bari: Laterza, 1966). Marx, Karl, Zur Kritik der politischen Ökonomie, 'Einleitung von 1857', in Karl Marx und Friedrich Engels, Werke, Band XIII (Berlin: Dietz, 1964). Italian translation by E. Cantimori Mezzomonti (Roma: Editori Riuniti, 1957). Mead, George Herbert, Mind, self, and society (Chicago: University of Chicago Press, 1934). Meillet, Antoine, see Les langues du monde. Mohrmann, Christine, Latin vulgaire, latin des chrétiens, latin médiéval (Paris: Librairie C. Klincksieck, 1955). Previously published in Revue des études latines XXIX (1952). Mounin, Georges, Les problèmes théoriques de la traduction (Paris: Gallimard, N R F , 1963). Morgan, William: see Young, Robert W., New directions in the study of language, ed. by Eric H. Lenneberg (Cambridge, Mass.: The M.I.T. Press, 1966). Nunnally, JumC., 'Individual differences in word usage', in Directions in psycholinguistics, q.v., pp. 203-234. O'Connor, D. J., 'Review of Language, thought, and reality. Selected writings of Benjamin Lee Whorf ( 1956)', in Mind LXVII: 267 (July 1958), 420-422. O'Connor, D. J. and A. H. Basson, 'Language and philosophy: some suggestions for an empirical approach', see Basson. Pinnow, Heinz-Jürgen, Die nordamerikanischen Indianersprachen. Ein Überblick über ihren Bau und ihre Besonderheiten (Wiesbaden: Harrassowitz, 1964). Problèmes de psycho-linguistique, Symposium de l'Association de psychologie scientifique de langue française (Paris: P.U.F., 1963). Quine, Willard Van Orman, Word and object (Cambridge, Mass.: The M.I.T. Press, 1960) (second printing 1964). Reimen, Jean René, 'Esquisse d'une situation plurilingue, le Luxembourg', La linguistique I: 2 (1965), 89-102. Le rêve et les sociétés humaines, sous la direction de Roger Caillois et G.E. von Grunebaum (Paris: Gallimard, 1957) (NRF, Bibliothèque de sciences humaines).
BIBLIOGRAPHY
87
Rosiello, Luigi, Struttura, uso e funzioni della lingua (Firenze: Vallecchi, 1965). Rossi-Landi, Ferruccio, 'De la communication d'une langue au point de vue épistémologique et au point de vue opératif', Congrès international de philosophie des sciences (Paris, octobre 1949); Actes (Paris: Hermann, 1951), I: Colloque d'épistémologie, pp. 177-182. — , 'Del non-traducibile', Congresso di studi metodologici (Torino, 1952), Atti (Torino: Ramella, 1954), pp. 112-117. — , 'Della presente traduzione', appendix to 'Del tipo di lavoro svolto da Gilbert Ryle', pp. XLII-LIV of the introduction to Gilbert Ryle: Lo spirito come comportamento (1955), q.v. — , Significato, comunicazione e parlare comune (Padova: Marsilio, 1961). — , 'Significato, ideologia e realismo artistico', Nuova Corrente 44 (1967), 300-342. — , Il linguaggio come lavoro e come mercato (Milano: Bompiani, 1968); a collection of the essays 'Il linguaggio come lavoro e come mercato' (1965), 'Sul linguaggio verbale e non-verbale' (1966), 'Per un uso marxiano di Wittgenstein ' (1966), 'Lavorando all'omologia del produrre' (1967), 'Per uno schema omologico della produzione' (1967), 'Ideologia come prozione sociale' (1967), also cited separately; for reference data see the volume. Rothacker, Erich, 'Ontologische Voraussetzungen des Begriffs Muttersprache', in Sprache. Schliissel zur Welt (1959), q.v., pp. 39-46. Russell, Bertrand, A critical exposition of the philosophy of Leibniz (Cambridge, 1900); new edition (1937), sixth printing (London: George Allen & Unwin, 1964). — , The analysis of mind (London, 1921); eighth printing (London: George Allen & Unwin, 1961). Ryle, Gilbert, 'Categories', Proceedings of the Aristotelian Society XXXVIII (1937-1938), 189-206; reprinted in Logic and language II (1953), q.v., pp. 65-81. — , 'Ordinary language', The philosophical review LXII (1953), 167-186. — , Lo spirito come comportamento, Italian edition of The concept of mind (1949) (Torino: Einaudi, 1955). — , 'The theory of meaning', in British philosophy in the mid-century, ed. C. A. Mace (London: Allen & Unwin, 1957). Sapir, Edward, Language. An introduction to the study of speech (New York: Harcourt, Brace & World, 1921; A Harvest Book, 1949). — , 'The grammarian and his language', American Mercury I (1924), 149-155; now in Selected writings, pp. 150-159. — , 'The status of linguistics as a science', Language V (1929), 207-214; now in Culture, language and personality, pp. 65-77. — , Selected writings of Edward Sapir in language, culture and personality, ed. by D . G . Mandelbaum (Berkeley and Los Angeles: University of California Press, 1949, fourth printing, 1963). — , Culture, language and personality (Berkely and Los Angeles: University of California Press, 1952); selected essays ed. by D. G. Mandelbaum. Sebeok, Thomas A., editor of Style in language (1960), q.v. Seiffert, Leslie, 'Neo-Humboldtian semantics in perspective: 'Sprache und Gemeinschaft', Journal of linguistics IV: 1 (April 1968), 93-108.
88
BIBLIOGRAPHY
Smythies, J. R., Analysis of perception (London: Routledge and Kegan Paul, 1956). Sociolinguistics, Proceedings of the UCLA Sociolinguistic Conference, 1964, ed. by William Bright (Janua Linguarum, Series Maior, 20) (The Hague: Mouton, 1966). Sprache. Schlüssel zur Welt. Festschrift für Leo Weisgerber, herausgegeben von Helmut Gipper (Düsseldorf: Schwann, 1959). Stenzel, Julius, Philosophie der Sprache (München-Berlin, 1933); reprinted (München: R. Oldenburg Verlag, 1964). Stern, H. H., Foreign languages in primary education, The teaching of foreign or second languages to younger children (London: Oxford University Press, 1967). A study of thinking, by Jerome S. Bruner, Jacqueline J. Goodnow, and George A. Austin, with an appendix on 'language and categories' by Roger Brown (New York: John Wiley and Sons, 1956) (sixth printing 1962). Style in language, ed. by Thomas A. Sebeok (New York: The M.I.T. Press and John Wiley, 1960). Tabouret-Keller, A., 'L'acquisition du langage parlé chez un petit enfant en milieu bilingue', in Problèmes de psycho-linguistique, q.v., pp. 205-219. Terracini, Benvenuto, Conflitti di lingue e di cultura (Venezia: Neri Pozza, 1957) (La biblioteca di cultura, 19). Thomson, George, Studies in ancient Greek society (London: Lawrence and Wishart). I. The prehistoric Aegean (1949, new edition 1954, reprinted 1961); II. The first philosophers (1955, new edition 1961). Trager, George L., 'The systematization of the Whorf hypothesis', Anthropological linguistics I: 1 (1959), 31-35. Trier, Jost, Der deutsche Wortschatz im Sinnbezirk des Verstandes. Die Geschichte eines sprachlichen Feldes, I.: Von den Anfängen bis zum Beginn des 13. Jh. (Heidelberg: Winter, 1931). Tschirch, Fritz, Weltbild, Denkform und Sprachgestalt. Grundauffassungen und Fragestellungen in der heutigen Sprachwissenschaft (Erkenntnis und Glaube, Heft 13). (Berlin: Vichem Verlag, 1954). Universals of language, ed. by J. H. Greenberg (Cambridge, Mass.: The M.I.T. Press, 1963); report of a Conference held at Dobbs Ferry, New York, April 1961. Voegelin, C. F. and F. M., 'Hopi domains: a lexical approach to the problem of selection', International journal of American linguistics (Indiana University Publications in Anthropology and Linguistics, Memoir 14, 1957). Indianersprachen, q.v., in — , Review of Pinnow, Die Nordamerikanischen Language XLIII: 2, Part 1 (June 1967), 573-583. Vygotsky, Lev S., Thought and language, ed. and trans, by E. Haufmann and G. Vakar (Cambridge, Mass.: The M.I.T. Press, 1962). Italian translation by various translators, ed. by A. Massucco Costa (Firenze: UniversitariaG. Barbèra, 1966). Waismann, Friedrich, 'Verifiability', Proceedings of the Aristotelian Society, Supplementary VolumeXIX (1945), pp. 119-150; now in Logic and language, I, q.v., pp. 117-144. - , 'Analytic-synthetic V', Analysis XIII (October 1952), 1-14, the series 'Analytic-synthetic' includes six essays published in Analysis from 1949 to 1952.
BIBLIOGRAPHY
89
Wartburg, Walther von, Einführung in Problematik und Methodik der Sprachwissenschaft, Zweite, unter Mitwirkung von Stephan Ullmann verbesserte und erweiterte Auflage (Tübingen: Max Niemeyer, 1962). French translation by A. Maillard and S. Ullmann: Problimes et méthodes de la linguistique (Paris: P.U.F., 1946). Weinreich, Uriel, Languages in contact (The Hague: Mouton, 1953), foreword by A. Martinet. Weisgerber, Leo, Die Entdeckung der Muttersprache im europäischen Denken (Schriften der Gesellschaft für Deutsche Sprache, Heft 1). (Lüneburg: Heliand Verlag, 1948). — , Das Tor zur Muttersprache (Düsseldorf: Schwann, 1950). — , Das Menschheitsgesetz der Sprache (Heidelberg: Quelle & Meyer Verlag, 1950). — , Von den Kräften der deutschen Sprache (Düsseldorf: Schwann). Band I. Grundzüge der inhaltbezogenen Grammatik (1962), (3rd. edition revised); Band II. Die sprachliche Gestaltung der Welt (1962), (3rd. edition revised); Band III. Die Muttersprache im Aufbau unserer Kultur (1957); BandIV. Die geschichtliche Kraft der deutschen Sprache (1959). — , Grundformen sprachlicher Weltgestaltung (Köln und Opladen: Westdeutscher Verlag, 1963). — , Die vier Stufen in der Erforschung der Sprache (Sprache und Gemeinschaft, Grundlegung Band II). (Düsseldorf: Schwann, 1963). — , Zur Grundlegung der ganzheitlichen Sprachauffassung (Düsseldorf: Schwann, 1964). Aufsätze 1925-1933, herausgegeben von Helmut Gipper. Werner, Heinz and Bernard Kaplan, Symbol formation. An organismic-developmental approach to language and the expression of thought (New York: John Wiley, 1963). (Bibliography of 284 titles). Whorf, Benjamin Lee, 'Aztec linguistics', report read in 1928 at the Twentythird international congress of Americanists, New York: Abstracts of papers, n. 116; cit. by John B. Carroll in the 'Introduction' to Language, thought, and reality, pp. 25-26. — , 'The punctual and segmentative aspects of verbs in Hopi' (1935), in Language, thought, and reality, pp. 51-56. — , 'An American Indian model of the universe' (1936), in Language, thought, and reality, pp. 57-64. — , 'Some verbal categories of Hopi' (1938), in Language, thought, and reality> pp. 112-124. — , 'The relation of habitual thought and behavior to language' (1939), in Language, thought, and reality, pp. 134-169. — , 'Gestalt technique of stem composition in Shawnee' (1940), in Language, thought, and reality, pp. 160-172. — , 'Science and linguistics' (1940),in Language, thought, and reality, pp. 207-219. — , 'Linguistics as an exact science' (1940), in Language, thought, and reality, pp. 220-232. — , 'Language and logic' (1941), in Language, thought, and reality, pp. 233-245. — , 'Language, mind, and reality', in Language, thought, and reality, pp. 246-270. — , Language, thought, and reality, selected writings, edited and with an introduction by John B. Carroll; foreword by Stuart Chase (New York: John
90
BIBLIOGRAPHY
Wiley, and The Massachusetts Institute of Technology, 1956). Word classes, special issue of Lingua XVII: 1 and 2 (November 1966), (Reprinted in Amsterdam: North-Holland Publishing Co., 1967). Yinger, J. Milton, Toward a field theory of behavior. Personality and social structure (New York: McGraw-Hill, 1965). Young, Robert W. and William Morgan, The Navaho language (United States Indian Service, Education Division, 1943).
B. A D D I T I O N A L R E F E R E N C E S ( I list here some writings which appeared after the summer of 1968, together with others of which I took notice after completing my essay, or which I had left out of the 1968 bibliography. They are all relevant for a multilateral approach to the problems of linguistic relativity. - Spring, 1971.) Albrecht, Erhard, Sprache und Erkenntnis. Logisch-linguistische Analysen (Berlin: VEB Deutscher Verlag der Wissenschaften, 1967). Beljaev, Boris Vasil'evic, Saggi di psicologia dell' insegnamento delle lingue straniere, transl. by A. Carpitella (Firenze: La Nuova Italia, 1968). Benjamin, Walter, Illuminationen, Ausgewählte Schriften (Frankfurt/Main: Suhrkamp, 1961). — , Angelus Novus, Saggi e frammenti, translation and introduction by Renato Solmi (Torino: Einaudi, 1962). — , Angelus Novus, Ausgewählte Schriften 2 (Frankfurt/Main: Suhrkamp, 1966). Bernstejn, N. A., Sulla costruzione dei movimenti (Mosca, 1947), [quoted by Ivanov, q.v.]. Boas, Franz, Race, language and culture (New York: The Free Press, 1940); see the section on Language, pp. 199-239. — , 'Introduction' to Handbook of American Indian Languages (and: Indian linguistic families of America North of Mexico, by J. W. Powell), edited by Preston Holder (Lincoln: University of Nebraska Press, 1966). Brooks, Nelson, Language, and language learning (Theory and practice) (New York: Harcourt, Brace & World, 1960). Italian translation by Vittorio Giglio, L'apprendimento delle lingue straniere (Teoria e pratica). (Firenze: la Nuova Italia, 1968). Brown, Roger Langham, Wilhelm von Humboldt's conception of linguistic relativity (Janua Linguarum, Series Minor, 65) (The Hague-Paris: Mouton, 1967). Brutian, G. A. 'O gipoteze Sepira-Vorfa', Voprosy filosofii 1 (1969), 56-66. Calame-Griaule, Geneviève, Ethnologie et langage. La parole chez les Dogon (Paris: Gallimard, 1966) ( N R F , Bibliothèque des sciences humaines). Carroll, John B., The study of language. A survey of linguistics and related disciplines in America (Cambridge, Mass.: Harvard University Press, 1953). Cassirer, Ernst, Language and myth (New York: Harper, 1946); translation by Susanne K. Langer of Sprache und Mythos (Berlin, 1925).
BIBLIOGRAPHY
91
— , Philosophie der symbolischen Formen, Band I: Die Sprache; Band II: Das mythische Denken; Band III: Phänomenologie der Erkenntnis (Darmstadt: Wissenschaftliche Buchgesellschaft, 1964). Italian translation by Eraldo Arnaud: Filosofia delle forme simboliche, Vol. I: Il linguaggio (Firenze: La Nuova Italia, 1961); Vol. II: Il pensiero mitico (idem, 1964); Vol. III, Tomo primo: Fenomenologia della conoscenza (idem, 1966); Tomo secondo (idem, 1966). Catford, J. C., A linguistic theory of translation. An essay in applied linguistics (London: Oxford University Press, 1965). Christensen, Niels Egmont, On the nature of meanings. A philosophical analysis (Copenhagen: Munksgaard, 1961, repr. 1965). Church, Joseph, Language and the discovery of reality. A developmental psychology of cognition, with a foreword by Robert B. MacLeod (New York: Random House, 1961). Cellier, Léon, Fabre d'Olivet. Contribution à l'étude des aspects religieux du romantisme (Paris: Librairie Nizet, 1953). Cohen, Marcel, Pour une sociologie du langage (Paris: Albin Michel, 1956). De Laguna, Grace Andrus, Speech. Its function and development (Bloomington: Indiana University Press, 1963 [1927]). Eaton, Trevor, The semantics of literature (De proprietatibus litterarum, Series Minor, 1) (The Hague-Paris: Mouton, 1966). Engels, Friedrich, Anti-Diihring, in Marx und Engels, Werke (Berlin, Dietz, 1962), Band 20, S. 1-303. Ervin, Susan M., 'Language and thought', pp. 81-91 of Horizons of anthropology, ed. by Sol Tax (London: G. Allen & Unwin, 1965) (Aldine, 1964). Ferguson, Charles A., 'Diglossia', Word XV: 2 (August 1959), 325-340; reprinted in Language in culture and society, edited by Dell Hymes (New York: Harper and Row, 1964), pp. 429-439. Francescato, Giuseppe, Il linguaggio infantile. Strutturazione e apprendimento (Torino: Einaudi, 1970). Gabka, Kurt, Theorie zur Darstellung eines Wortschatzes. Mit einer Kritik der Wortfeldtheorie (Halle [Saale]: [VEB] Max Niemeyer Verlag, 1967). Gair, James W., 'Sinhalese diglossia', Anthropological linguistics X: 8 (November 1968), 1-15. Glenn, Edmund S., 'Semantic difficulties in international communication', ETC. XI: 3 (Spring 1954), 163-180; now in The use and misuse of language, ed. by S. I. Hayakawa (Greenwich, Conn.: Fawcett Publications, A Premier Book, 1962), pp. 47-69. Hall, Edward T., The silent language (Greenwich, Conn.: Fawcett Publications, A Premier Book, 1959). Hall, Robert K.,An essay on language (Philadelphia: Chilton Books, 1968). Hallig, Rudolf and Waither von Wartburg, Begriffssystem als Grundlage für die Lexikographie. Versuch eines Ordnungsschemas, 2., neu bearbeitete und erweiterte Auflage; also in French: Système raisonné des concepts pour servir de base à la lexicographie. Essai d'un schéma de classement, 2e édition recomposée et augmentée (Berlin: Akademie-Verlag, 1963 [1952]). Hamburg, Carl H., Symbol and reality, Studies in the philosophy of Ernst Cassirer (The Hague: Martinus Nijhoff, 1956).
92
BIBLIOGRAPHY
Harris, Marvin, The nature of cultural things (New York: Random House, 1964). Haugen, Einar, 'Problems of bilingualism', Lingua II (1949-50), 271-290. Horizons of anthropology, ed. by Sol Tax (London: Allen & Unwin, 1965). Humphrey, George, Thinking. An introduction to its experimental psychology (London: Methuen & Co., 1951) (New York: John Wiley & Sons, Science Editions, 1963). Husserl, Edmund, Logical investigations, translated by N. J. Findlay from the second German edition of the Logische Untersuchungen (1913, 1921) (London: Routledge & Kegan Paul, 1970). Ivanov, Vjaceslav Vsevolodovic, Logileskaja struktura nauinogo znanija (Moskva, 1965). Quoted in Ivanov's essay 'Il ruolo della semiotica nello studio cibernetico dell'uomo e del collettivo', Italian translation by Lucetta Negarville in Sigma 17 (March 1968), 66-80; see also 'Ruolo della semiotica nell'indagine cibernetica dell'uomo e della collettività' in I sistemi di segni e lo strutturalismo sovietico (Milano: Bompiani, 1966), pp. 41-54. Jaspers, Karl, Psychologie der Weltanschauungen (Berlin: Julius Springer Verlag, 1925). Italian translation by Vincenzo Loriga: Psicologia delle visioni del mondo (Roma: Astrolabio, 1950). Kainz, Friedrich, Psychologie der Sprache, 1. Band: Grundlagen der allgemeinen Sprachpsychologie, Dritte unveränderte Auflage (Stuttgart: Ferdinand Enke Verlag, 1962); 2. Band: Vergleichend-genetische Sprachpsychologie (1960); 3. Band: Physiologische Psychologie der Sprachvorgänge (1965); 4. Band: Spezielle Sprachpsychologie (1967); 5. Band, 1. Teil: Psychologie der Einzelsprachen, I. (1965); 5. Band, 2. Teil: Psychologie der Einzelsprachen, II. (1969). Klaus, Georg, Die Macht des Wortes. Ein erkenntnistheoretisch-pragmatisches Traktat (Berlin: VEB Deutscher Verlag der Wissenschaften, 1965). Kusnetsov, I. V., 'The structure of the scientific theory and the structure of the object', Soviet studies in philosophy VII: 2 (Fall 1968), 15-26 (original in Voprosy filosofii 5, (1968) 72-83. Language and culture (Symposium on), Proceedings of the 1962 Annual Spring Meeting of the American Ethnological Society (Seattle: University of Washington, 1963); edited by Viola E. Garfield. Language, thought, and culture, writings by Roger W. Brown, Irving M. Copi, D o n E. Dulaney, William K. Frankena, Paul Henle, Charles L. Stevenson; ed. by Paul Henle (Ann Arbor: The University of Michigan Press, 1958). Lefebvre, Henri, Critique de la vie quotidienne (Paris: L'Arche, 1958 et 1962); Vol. I (1958) Vol. II (1962). — , Le langage et la société (Paris: Gallimard, N R F , 1966). Leisi, Ernst, Der Wortinhalt. Seine Struktur im Deutschen und Englischen (Heidelberg: Quelle & Meyer, 1953). — , Das heutige Englisch. Wesenszüge und Probleme (Heidelberg: Carl Winter, Universitätsverlag, (1955). Leopardi, Giacomo, Zibaldone di pensieri, ed. by F. Flora, 2 volumes (Milano: Mondadori, 1961). Llewelyn, J. E., 'On not speaking the same language', Australasian journal of philosophy I: X L (1962), 35-48; II: X L (1962), 127-145.
BIBLIOGRAPHY
93
Lohmann, Johannes, Philosophie und Sprachwissenschaft (Erfahrung und Denken, 15). (Berlin: Duncker & Humblot, 1 9 6 5 ) . Lyons, John, Structural semantics. An analysis of part of the vocabulary of Plato (Oxford: Blackwell, 1963). Martinet, André, 'Réflexions sur les universaux du langage', Presidential address presented at the Brussels meeting of the SLE, 1st April 1968, Folia Linguistica, Acta Societatis Linguisticae Europaeae I: 3 and 4, 125-134. Marty, Anton, Untersuchungen zur Grundlegung einer allgemeinen Grammatik und Sprachphilosophie (Halle: Max Niemeyer, 1908). — , Nachgelassene Schriften: Psyche und Sprachstruktur, edited by Otto Funke (Bern: A . Francke, 1940). — , Nachgelassene Schriften: Über Wert und Methode einer allgemeinen beschreibenden Bedeutungslehre, edited by Otto Funke (Bern: A . Francke, 1950). Mesthene, Emmanuel G., How language makes us know. Some views about the nature of intelligibility, with a foreword by John Herman Randall, Jr. (The Hague: Martinus NijhofF, 1964). Miller, George A . , Language and communication ( N e w York: McGraw-Hill, 1951) (1963 Paperback). Miller, Robert L., The linguistic relativity principle and Humboldtian ethnolinguistics. A history and appraisal (Janua Linguarum, Series Minor, 67) (The Hague-Paris: Mouton, 1968). Morris, Charles, Signification and significance. A study of the relations of signs and values (Cambridge, Mass.: M.I.T. Press, 1964). Nida, Eugene A . , 'Linguistics and ethnology in translation-problems', Word I: 2 (1945), 194-208. Pagliaro, Antonino, 'Senza tempo', pp. 65-70 in II segno vivente, Saggi sulla lingua e altri simboli (Napoli: Edizioni scientifiche italiane, 1952). — , La parola e l'immagine (Napoli: Edizioni scientifiche italiane, 1957). Pike, Kenneth L., Language in relation to a unified theory of the structure of human behavior (Janua Linguarum, Series Maior, 24) (The Hague-Paris: Mouton, 1967). Plans and the structure of behavior, by George A. Miller, Eugene Galanter, Karl H. Pribram ( N e w York: Holt, Rinehart and Winston, 1960). Ponzio, Augusto, 'Ideologia della anormalità linguistica', Ideologie 15 (1971), 50-73. Postman, Neil and Charles Weingartner, La linguistica. Una rivoluzione nell'insegnamento, translation and integrating notes by Giorgio R. Cardona (/problemi della didattica, 37) (Roma: Armando Armando, 1968). Originai Linguistics. A revolution in teaching (New York: Delacorte Press, 1966). Powell, J. W., Indian linguistic families of American north of Mexico (and: 'Introduction' to Handbook of American Indian Language by Franz Boas), edited by Preston Holder (Lincoln: University of Nebraska Press, 1966). Riverso, Emanuele, Introduzione alla filosofia e all'analisi del linguaggio (Napoli: Instituto Editoriale del Mezzogiorno, 1960). Rossi-Landi, Ferruccio, Semiotica e ideologia (Milano: Bompiani, forthcoming). —, Language as labour and as trade (The Hague: Mouton, forthcoming). — , 'Linguistics and economics', in Current trends in linguistics, Vol. 12 (The
94
BIBLIOGRAPHY
Hague: Mouton, forthcoming). Ruesch, Jürgen and Weldon Kees, Nonverbal communication (Berkeley and Los Angeles: University of California Press, 1956). Sartre, Jean-Paul, Les mots (Paris: Gallimard, 1964). Schaff, Adam, Introduction to semantics, Translation from Polish by Olgierd Wojtasiewicz (Oxford: Pergamon Press, 1962). — , Jçzyk a posnanie (Warzawa: Pañstwowe Wydawnictwo Naukowe, 1964). — , Sprache und Erkenntnis (Wien: Europa-Verlag, 1964). — , Essays über die Philosophie der Sprache (Frankfurt: Europa-Verlag, 1968). — , Langage et connaissance (Paris: Anthropos, 1969). Schlanger, Judith E., 'La langue hébraïque, problème de linguistique spéculative', Revue internationale de philosophie XXI: 4 (1967), 486-507. Searle, John R., Speech acts. An essay in the philosophy of language (Cambridge: University Press, 1969). Shands, Harley C., Thinking and psychotherapy. An inquiry into the processes of communication (Cambridge, Mass.: Harvard University Press, 1960). Sinclair-De Zwart, Hermina, Acquisition de langage et développement de la pensée. Sous-systèmes linguistiques et opérations concrètes (Paris: Dunot). Slama-Cazacu, Tatiana, Introducere in psiholingvisticä (Bucureçti: Editura Çtiin^ificâ, 1968). — , Comunicarea in procesul muncii (Bucurejti: Editura Çtiin^ificâ, 1964). Snell, Bruno, Die Entdeckung des Geistes. Studien zur Entstehung des europâischén Denkens bei den Griechen (Hamburg: Ciaassen Verlag, (1948). Italian translation by Vera degli Alberti and Anna Solmi Marietti: La cultura greca e le origini del pensiero europeo (Torino: Einaudi, 'P.B.E. 36', 1963). English translation by T. G. Rosenmeyer: The discovery of the mind. The Greek origins of European thought (New York: Harper & Brothers, The Academy Library, 1960) [Harvard University Press, 1953]. — , Der Aufbau der Sprache (Hamburg: Ciaassen Verlag, 2nd revised and enlarged edition, 1952). Italian translation by Lea Ritter Santini: La struttura del linguaggio (Bologna: Il Mulino, 1966). Sommerfeit, Alf, La langue et la société. Caractères sociaux d'une langue de type archaïque (Oslo: H. Aschehoug & Co., 1938). Sperber, Hans, Einführung in die Bedeutungslehre (Bonn: Dümmlers Verlag, 1965). Sprout, Harold and Margaret, The ecological perspective on human affairs. With special reference to international politics (Princeton, N.J.: Princeton University Press, 1965). Swadesh, Morris, 'Linguistics as an instrument of prehistory', Southwestern Journal of Anthropology XV: 1 (spring 1959), 20-35. Tovar, Antonio, Catàlogo de las lenguas de América del Sur. Enumeración, con indicaciones tipológicas, bibliografía y mapas (Buenos Aires: Editorial Sudamericana, 1961). Triandis, Harry C., 'Cultural influences upon cognitive processes', Advances in experimental social psychology, ed. by Leonard Berkowitz, Vol. I (1964), pp. 1-48. Tullio-Altan, Carlo, 'Considerazioni sull'ipotesi Sapir-Whorf', Sociologia III: 3 (September 1969), 77-115; English translation at pp. 116-146.
BIBLIOGRAPHY
95
Ullmann, Stephen, The principles of semantics. A linguistic approach to meaning (Glasgow: Jackson; Oxford: Blackwell, 1951). Universals in linguistic theory, ed. by Emmon Bach and Robert T. Harms (New York: Holt, Rinehart & Winston, 1968). Urban, Wilburg Marshall, Language and reality. The philosophy of language and the principles of symbolism (London: Allen & Unwin, 1939). Waismann, Friedrich, The principles of linguistic philosophy, edited by R. Harré (London: MacMillan, New York: St. Martin's Press, 1965). Weingartner, Charles, see Postman, Neil: La linguistica. Una rivoluzione nell'insegnamento (Roma: Armando Armando, 1968). Whatmough, Joshua, Language. A modern synthesis (New York: A Mentor Book [The New American Library], 1957). Whorf, Benjamin Lee, Linguaggio, pensiero e realtà. Raccolti di scritti a cura di John B. Carroll, translation by Francesco Ciafaloni, introduction by Alberto Mioni (Torino: Boringhieri, 1970).
SUBJECT INDEX
African languages of the slaves: 72. American English: 57-58. Amerindian languages (in general): 10-11, 53, 72, 77-78, 80. analytical philosophy: 39-42. a priori and language: 38-39, 67. Arabic: 11, 58, 59, 62, 69. Aranda: 8. archaic particularism: 11. Aztec: 11. bilingualism: 5-6 and n. bourgeois ideology: 65, 71-72, 78-79. children: 5-6, 72, 76-77, 78. Chinese: 7, 8, 32, 47, 54, 60, 80. codifying procedures: 33, 53-55. common (collective) speech: 4, 42, 62-63; — see under language-in-general and speakers as variable capital conditioning power of language on thought (or other social factors): 41, 46, 49-50, 68-69; — see also thought and language; linguistic relativity, formulations of. constant linguistic capital as a language: see under language-aslangue. culture and language: 67-68. Danish: 57. decoding: see codifying procedures, deep structures and linguistic relativity: 31, 38, 43, 55, 60, 80. determinism of language on thought:
see conditioning power of language on thought differences among languages: 7-8, 11, 29-30, 41, 55, 64; — and similarity of problems: 58-61, 72-73. Egyptian: 68. English: 6, 7, 11, 19, 32, 35, 36, 40, 48, 54, 60, 62, 72. Eskimo: 13, 68-69. Ewe: 59-60. experience: — as Erlebnis: 32; — as non-repeatability: 5, 6-8; — as repeatability: 5, 6-7; — of one's mother tongue: see mother tongue; — of the world: 30-31, 4 2 ^ 3 , 49, 76. extra-linguistic cultural factors: 66-70. French: 11, 57, 60, 73. German: 6, 11, 32, 47, 57, 60, 69. glottocentrism: 41, 79. Greek: 44, 45, 59, 68. Hebrew: 54, 58, 59, 62. Hopi: 10-12, 13, 14-17, 18-21, 37, 62, 65; — model of the universe: 18-21; — particle and field in: 14-17; — terminological wealth: 16; — verb system: 14-16. individual or group differences in speaking irrelevant here: 7, 32, 33, 56-57.
SUBJECT INDEX individual speech: 49, 62-63, 64. influence of language on thought and/or world-view: see linguistic relativity (thesis of): formulations of; thought and language, interdisciplinary treatment: 1, 2-3. intermediate world: 35, 47-49 and n. Italian: 7, 32,35,37-38,47,48,57, 69. Japanese: 45. langage: see under language-in-general. language-as-langue, as constant capital: 4, 63-65, 70, 73, 75; — and culture: 67-69; — as an isolated universe: 36-38,48, 71; — as historical product: 12, 72-73; — as social machinery: 32, 56-57, 64, 74, 75-76; — as social reality: 6-7. language-in-general (langage) as total capital (wealth, patrimony): 4, 42, 63, 73, 77-78; — and common speech: 4,49,63-65; — and perception: see perception and language; — and the language (langage and langue): 4 and n, 40, 49, 51, 62-65, 68, 73; — and thought: see thought and language; — as the dialectical sum of a language and common speech: 4,49, 63-65; — see also: language as langue; linguistic production; self extensive or generative power of language; and the entries beginning with linguistic. langue: see language-as-Zon^ue. Latin: 35, 57, 58-59, 60, 62. learning to speak: 5-6, 33, 62-63, 76-77.
97
linguistic alienation: 73-79. linguistic capital: see under languagein-general and language-as-langue. linguistic commodities or goods: 42, 75. linguistic consumers: 74, 75-77. linguistic instruments: 63, 64, 73. linguistic materials: 63, 64. linguistic "money": 63, 64, 73, 74. linguistic production: 63-65, 68, 72-73, 74-75, 76-77; — and material production: 72, 77. linguistic products: 65. linguistic relativity (thesis of): — and deep structures: see deep structures and l.r.; — and linguistics: see linguistics and l.r.; — and philosophy: see philosophy and l.r.; — and psychology: see psychology and l.r.; — as an illusion: 74; — as pseudo-linguistic hypothesis: 69-70; — as separation of langue from langage: 64-65; — excessive interpretation of: 35-45; — formation of: 1,10-12; — formulations of: 1,7, 8, 9, 11-12, 28-34, 43,49, 56-57, 71; — fundamental logic of: 68-70; — Italian approximations to: In, 3n, 13n, 40n; — its fascination: 2, 12, 51, 64, 73; — neo-idealistic character of: 2-3, 64, 65, 70, 71-73, 78-79; — neo-positivistic approach to:2-3, 52, 72-73, 74-75, 77; — over-hasty formulation, acceptation, or denial of: 8-9, 12, 43-45, 46n; — thesis or hypothesis: 1, 71.
98
SUBJECT INDEX
linguistic surplus-value: 74. linguistic use-values: 63, 76. linguistic work: 63, 75, 76; — see techniques of expression and communication; and under language-in-general. linguistic working process: see techniques of expression and communication, linguistics and linguistic relativity: 43-45. linguistic universals vs. linguistic relativity: 43,47, 80. material production: 72, 77. Maya: 11. mental processes (invented by supporters of linguistic relativity): 65. mental universe of a language: 28-30, 37-38, 54-55. metaphysics: 71. mind: 31; — being the working by signs: 8 and n. — individual and social: 8 and n; mother tongue: 4-8,33, 37,44,48, 50, 72, 73, 74; — analysis of one's: 39; — experience of one's: 4-8, 12, 49; — detachment from: 72, 73. Navajo (Navaho): 10-11, 13, 25-27, 35-37, 53, 54, 68; — "chemical" character of: 25-27; — terminological wealth of: 25. neo-positivistic approach: 2-3, 72, 78. Nitinat: 10. non-linguistic factors operating on thought: 66. non-linguistic (non-verbal) human behaviour: 8, 41, 66-70. non-verbal communication: 36. parole: see common (collective) speech; individual speech; and under
Ianguage-in-general. perception and language: 43, 52, 80, philosophy and linguistic relativity: 38-42 and n, 44, 45, 58-59, 72, 73; — and see: problems (philosophical and others) in different languages; differences among languages and similarity of problems. philosophy and linguistics: 40 and n. physical objects and language: 47-49; — see also: perception and language. Portuguese: 11. problems (philosophical and others) in different languages: 8, 56-59, 61, 72-73. production and use: 64-65, 75-76. psychology and linguistic relativity: 42-43 and n, 52. reality vs. language: 46-49. Rumenian: 68. Russian: 7, 11, 32. SAE languages: 11, 16-17, 19, 35-37. signantia and signata: 31, 44, 47-48. self-extensive or generative power of language: 62-65, 73. separatists fallacy: 69. Shawnee: 13. solipsism, linguistic: 54. Spanish: 37-38. speakers as variable linguistic capital: 4, 63, 73-74, 77-78. — "spoken" by their language: 7, 73-74; — see also under language-in-general. speech: see common (collective) speech; individual speech; and under language-in-general. Sprachgefühl: 33. techniques of expression and communication: 4, 7, 9, 62-63, 74, 76, 78.
SUBJECT INDEX thought: 40, 50; — and language: 9, 18-19, 28-31, 32, 33, 36, 38, 40-41, 50-51, 59-61, 67-70; — concomitance of: 50-51; — and see techniques of expression and communication. total linguistic capital: see under language-in-general. totality and sub-totalities in dealing with language: 68-70. translating: 9, 26, 27, 37, 42, 53-55, 64.
99
use and production: see production and use. use and usage, linguistic: 40. variable linguistic capital as the speakers: see under language-ingeneral. Weltanschauung: see world-view. Wintu: 13, 22-24; — holism in: 23-24; — human activities in: 22-23. world-view: 21, 23-24, 33, 35, 52, 56-59, 70, 72, 73, 75, 78. Zwischenwelt: see intermediate world.
INDEX OF NAMES
This index includes names appearing in the text and notes, but not those in the Bibli igraphy. Abbagnano, Nicola: 5 n. Aginsky, Burt W. and Ethel G.: 44 n. Aristotle: 45, 59. Austin, John L.: 40 n. Basilius, Harold: 46 n. Basson, A. H.: 58 n. Benveniste, Émile: 44 and n, 59 and n, 60. Bergson, Henri: 57, 73. Bertalanffy, L. Von: 56 n. Black, Max: 46 n, 77. Boas, Franz: 12. Brown, Roger W.: 2 n, 52 n, 56 n. Carroll, John B.: 11 n. Casagrande, Joseph B.: 2 n. Chase, Stuart: 3 n. Chomsky, Noam: 43, 44 n. Cohen, L. Jonathan: 56 n. Cohen, Marcel: 10 n. Colorni, Eugenio: 32. Cremonini, Cesare: 57. Croce, Benedetto: 3 n. Davidson, Basil: 58 n. Deese, James: 56 n. Devoto, Giacomo: 5 n. De Waelhens, Alphonse: 45 n. Diamond, Stanley: 56 n. Dilthey, Wilhelm: 32. Dorfles, Gillo: 3 n.
Eco, Umberto: 66 n. Eggan, Dorothy: 10 n. Elders, Léon: 45. Engels, Friedrich: 70 n. Ervin, Susan M.: 52 n. Fanon, Frantz: 58 n. Feuer, Lewis S.: 46 n, 57, 58 n, 77. Findlay, John N.: 39 n. Fishman, Joshua A.: 5 n, 46 n, 52 n. Flew, Antony G. N.: 39 and n. Forke, Alfred: 60 n. Galilei, Galileo: 57. Gastil, Raymond D.: 56 n. Gipper, Helmut: 44. Gleason, Henry Allan, Jr.: 56 n. Greenberg, Joseph H.: 43 and n. Guchman, M. M.: 28 n. Hallowell, A. I.: 43 n. Hegel, Georg W. F.: 11 and n, 41, 57. Heidegger, Martin: 21. Hertzler, Joyce O.: 46 n. Hjelmslev, Louis: 44. Hoijer, Harry: 12, 26 n, 27 n. Homer: 45. Horowitz, Arnold E.: 52 n. Humboldt, Wilhelm von: 1,11,35,48. Husserl, Edmund: 32, 38 and n, 39. Hymes, Dell: 46 n.
INDEX OF NAMES Jenkins, James J.: 2 n. Kant, Immanuel: 38. Kaplan, Bernard: 43 n. Kierkegaard, Sören: 57. Kluckhohn, Clyde: 11 n, 25 and n, 26, 27 n, 36, 53. Kushner, Eva: 5 n. La Barre, Weston: 56 n. Landar, Herbert J.: 52 n , 56 n. Lee, Dorothy: 3 n, 12, 13, 22 and n, 23 and n , 24 n. Leighton, Dorothy: 11 n, 25 and n, 26, 27 n, 36, 53. Lenneberg, Eric H.: 8 n, 46 n, 52 n, 77. Levenson, Joseph R.: 60 n. Lévi-Strauss, Claude: 67 and n. Luther, Wilhelm: 45. MacNamara, John: 5 n. Markiewicz, Zygmunt: 5 n. Martinet, André: 5 n, 44 and n. Marx, Karl: 65 n. Mead, George Herbert: 8 n, 33. Meillet, Antoine: 10 n. Mohrmann, Christine: 58 and n. Morgan, William: 27 n. Mounin, Georges: 44 and n. Nunally, Jum C.: 2 n. O'Connor, D. J.: 56 n, 58 n. Pagliaro, Antonio: 3 n. Pinnow, Heinz-Jürgen: l l n . Powell, J. W.: 12. Quine, Willard Van Orman: 56 n. Reimen, Jean René: 5 n.
101
Rosiello, Luigi: 3 n. Rothacker, Erich: 48 and n. Russell, Bertrand: 45. Ryle, Gilbert: 40 n, 41 n, 42 n. Sapir, Edward: 1, 3 n, 10, 28 and n, 29, 36, 62 n, 80. Saussure, Ferdinand de: 44, 63. Sayce, A. H.: 45. Seiffert, Leslie: 46 n. Shands, Harley C.: 43 n. Smythies, J. R.: 42, 43 and n. Sommerfeit, Alf: 1, 10 n, 11 n. Stenzel, Julius: 44. Stern, H . H.: 5 n. Tabouret-Keller, Andrée: 5 n. Terracini, Benvenuto: 3 n. Thomson, George: 68 and n. Trager, George L.: 46 n. Trier, Jost: 1, 3 n, 45 n, 80. Tschirch, Fritz: 45 n. Voegelin, C. F. and F. M.: 11 n, 21 n. Vygotsky, Lev S.: 56 n. Waismann, Friedrich: 39 and n. Wartburg, Walther von: 3 n, 45 n. Weinreich, Uriel: 5 n. Weisgerber, Leo: 1, 3 n, 6, 35, 44, 45 and n, 48, 79, 80. Werner, Heinz: 43 n. Whorf, Benjamin Lee: 1, 3 n, 10, 11 and n, 12 and n, 13 and n, 14, 15, 16, 17, 18, 19, 21 and n, 28 and n, 29, 30, 31, 32, 35, 38, 39, 42, 44, 51, 53, 54, 55 and n, 59, 62, 65, 67 and n, 68, 69, 78, 80. Wittgenstein, Ludwig: 42 n. Yinger, J. Milton: 43 n. Young, Robert W.: 27 n.