LAM DJACK .AND CULT URTS t NOTES ON Wil OR F
Sloise Kerlin
0. introductory* 1, horf*3 thesis on language and culture. 2, ‘Expression and Content. 5. Instrument and Observation. 4. Lexicon and Morphology. 5. Conclusions.
0. Introductory.
The writings of Benjamin Lee Whorf (1) on language and culture have warranted much interest and discussion. Thera still seoiaa to he considerable disagreement, however, among both linguists and anthro- pologists, as to the validity and even the nature of his assertions* It is the intent of the present discussion to establish the sources of this confusion, which may be traced to an ambiguity inherent in khorf*a approach. No attempt is made to consider whether Whorf*a conclusions are justified by his data, since most of 'Is writings on this subject were informal and expository, and were designed to be provocative dis- cuss ion rather than blue systematic investigation of a problem 1. whorf*s the is on language and culture. 1.1. Whorf was primarily interested in the relationship be^veen language and thought, and his assumptions on the relationship bet'feen 3 ".ch behavior and culture ore implicit. In each of Is important essays he is concerned with demonstrating that ho?/ we talk about some- thing affects the way in which we think of it. in his work as an insurance investigator, he noted a relationship between the talking and the behavior in a given situation; when the former was inappropriate, 30 was the latter, behavior which contributed to or failed to prevent the occurrence of a fire• People who spoke of a certain mat-rial as 'stone” did not regard It as combustible, and behavior around gasoline storage drums spoken of as "empty” was not properly cautious.
1.2. thorf wished to refute the view that thought can be formulated independent of language, or that language merely exprasses thought. Tills common sense or "natural logic" vie?; ho ascribed to a lack of knowledge of the different structuring of experl 0X1Cn provided by languages differing from our ovm. Only by means; 01 shah knowledge can the observer gain sufficient perspective to ree’^f® that hi3 own per- captions of the world about him tire affected b the formulations imposed by the language appe- "laws of thought" are deduced. This Is becar the speaker with a set of distinctions and g' events which are compulsory, and which are r ture which varies from language to language
ill s'-language j o ther% 1 ae, to be universale, and he language provides * eatlons concerning into a total struo- > other factors,
2which may be regarded as constants, enter into experience, but given the same event, reports vary with the language in which they are pre- sonted. This is the "principle of (linguistic) relativity" which Vvhorf would like to establish: "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,” (2) 1.3. This principle would be of great importance to the scientist, in his role of observer. Vvhorf therefore recommends a study of differ- ing language structures from the point of view of their differing organizations of experience, so that the scientist may become aware of and try out new conceptualizations of his material which may enable hi i to make more fruitful observations. In some passages hhorf seems to suggest that the scientist is entirely limited by his linguistic framework, in others that it merely handicaps him, 1.4. '„hat is the relationship of this principle to the concept of culture? Whorf was not interested In the question of the determinants of cultural growth and change as such, but he does indicate that the role of language here is not primary. Rather, it interacts with other unspecified elements to produce the total cultural situation# Given a certain cultural movement, the language In which It is talked about by the participants in the culture can foster or inhibit it; but the change itself derives impetus from other sources. In the ofton-cited comparison between Hopi and Standard Average European, Whorf notes the many commercial and technological traits which are consonant with the verb and noun system of SAE, and ways in which the Hopi language and culture are in accord with one another. In a section on "Historical Implications" he asks which came first: "But in this partnership the nature of the language is the factor that limits free plasticity and rigid!fiea channels of development in the more autocratic way. This is because language is a system, not just an assemblage of norms. 'Large systemic outlines can change to something new only very slowly, while many other cultur- al Innovations are made with comparative quickness. Language thus represents the mass mind; It is affected by Inventions and innova- tiono, bi.it affected little and slowly, whereas Jco inventors and Innovators it legislates with the decree immediate." (3) It will be nelpful to Include here also bhorf*s two conclusions as a result of his Informal comparison of SAS and Hopi: 1.) "Concepts of »time* and »matter* are not given in subatantl- ally the same form by experience to all men but depend upon the nature of the language or languages through the use of which they have been developed. They do not depend so much upon any one system (e. g., tense, or nouns) within the grammar as upon the ways of analyzing and reporting experience which have become
3fixed in the language as Integrated »fashions of speaking1, and which cut across the typical grammatical classifications, so that such a fashion may include lexical, morphological, syntactic, and otherwise systemlcally diverse means coordinated In a certain frame of consistency*.,. 2*) "There are connections hut not correlations or diagnostic correspondences between cultural norms and linguistic patterns. Although it would he impossible to infer the existence of Grier Chiefs from the lack of tenses in Hopl, or vice versa, there is a relation between a language and the rest of the culture of the society which uses It. There are oases where the »fashions of speaking’ are closely integrated with the whole general culture, whether or not this be universally true, and there are oonnec- tions within this integration, between the kind of linguistic analyses and various behavioral reactions and also the shapes taken by various cultural developments. Thus the importance of Grier Chiefs does have a connection, not with tenselesaness Itself, but with a system of thought in which categories differ- ent from our tenses are natural. These connections are to be found not so much by focussing attention on the typical rubrics of linguistic, ethnographic, or sociological description as by ex mining the culture and the language (always and only when the two have been together historically for a considerable time) as a whole in which concatenations that run across these departmental lines may be expected to exist, and if they do exist, eventually to be discoverable by study." (4) 1.5. In summary than, bhorf»s postulates on language and culture may be listed as follows! 1.) Language mediates observation, and therefore affects the thought and behavior of the speaker. 2.) Languages vary in structure, and therefore, vary in their suitability for observing and communicating upon different events. This is of concern to th© scientist. 5.) In a given speech community, language interacts with other elements in the culture shared by the communicants, rnd limits or encourages certain kinds of cultural change, 4.) Therefore a relationship between the language and the cul- ture of a group may be expected, given a long enough period of association between the two, 2. xpreasion and Content. 2.1. r.ouls Ijolmslev, in a formally projected analysis of language which begins with the undivided whole, asserts that the first step in such an analysis must be the registration of two planes of language, that of the expression and that of the content. (5) This division is made by virtue of the sign-function, which links the two .lanes in a
4way willch la arbitrary yot fixed for a given language. In order to consider tills view, we will need to d scuss tiie way In which Ejelmslev uses the terras involved. According to Hjelmslev, the sign should not be described aa a sign for something, implying a designaturn which is extra-linguistic, but as a relationship between tv?o linguistic struc- tures, content and expression. Content, expression, and the sign- function thus I'm a "triad, in which each may be defined only in terms of the other two, and in which each presupposes the other two. If we take the sign boy for example, it may b© described as a fixed but arbi- trary relationship between the expression elements—the phonemes, or whatever distinctive units we may register, and the content elements— what has been termed the ’’meaning11-- male child'1, or whatever dlatlnc- tive units we may register. 2.2. The sign-function is the basis for linguistic analysis, since by virtue of it signs are delimited from one another, and the distinctive units of the language below the level of the sign are established. This is carried out by means of replacement testa, which taJ.ce account of the changes in either plane which are related to changes in the other. Thus a phonemic distinction would be a change In the units of the expression plane which is related to a change in the content plane, by means of particular 3ign-functian3. An analysis of the expression units of boy Into /b/ /o/ /y/ would reflect a survey of changes in the expression!; such as the substitution of /1/ for the initial unitjwhlch would result In a different sign (toy). There are substitutable units In the sign content also; thus we register a linguistically significant distinction between "male" and "female” in this content, since we may substitute "female” for "male" and again observe that we are dealing with a new sign, girl. 2*3. These units of language which result from the analysis of the sign-expression and the sign-content Hjelmslev terras figurae. There are thus exprs s i on-f igur ae and content-figure©, each class established by the way its members enter into sign-functions. Thore is thus a re- cognition of form in both planes of a language, and the content plane is so m to be a structure analogous to that of the expression plane. Hjelmslev points out that a language should not be described as a sya- tern of signs, but as a system of figurs© entering into signs* The number of the figurae of a language is strictly limited, but the number of signs and sign complexes exceedingly large. This is what makes for a system which is both manageable and extremely productive. 2.4. An attempted analysis of content figurae in the case of an iso- latch sign is comparable to an attempt at a phonemic analysis on the basis of a single utterance. The Important point is that we are here provided with the formal basis for an analysis of the content plane of a language, described aa a complete system in which units are defined in terms of their relationships to one another. In the case of a sign such a3 a Latin nominal suffix, the value of a separate analysis of expression figurae (the phonemes) and content figurae (number, gender,
54
6
oase) is soli-evident. It will be observed that the figurae in the one plane which enter into a given sign-function, and the flgurae of the other piano which enter into that same sign-function, may not be further related. This is a consequence of the arbitrary nature of -the sign-function. It is characteristic of the sign-function that it may not bo resolved into a like number of figurae in each plane; oven If the number' happens to be the same, no further relationships between members of the two figurae groups may be observed. Thus if at one stage of the analysis the linguist chose to analyse the expression of boy into tv/o units, say a syllable periphery and a syllable nucleus, he would be unable to relate either of these more significantly to “male” or '''child1'. If lie were able to do so he would by definition not have arrived at the minimal sign. This would bo the case in deal- ing with a complex sign structure such as boys, whore a relationship between one sub-group of content and expression figurae r(/b/./o/./y/),("male"/child"' and another relationship between /T/s/). ("more than one"Ji7 could be observed. 2.6. e thus arrive at an important point: the lack of isomorphism between the content end expression planes of a language. This is the justification of a separate analysis of each plane, each consisting of a hierarchy of units which may be described in terms of their rela- tionships to one another. In lljelmslev’s view, each plane Is a lingula- tic. forsi which is imposed upon events external to it, and so produces a linguistic substance. In the case of the expression plane, the form at one level Is the phonemic pattern of the language which in combin- atlon with certain articulatory or acoustic events produces the pho- names of the language. In the case of the content plane, the form is the pattern of content figurae of the language, which in combination with other :inds of external events (physical, psychological, soclo- logical, etc., depending upon the kind of study undertaken) produces the content substance of the language; e. g,, a paradigm of color signs which does or does not recognize a distinction between green and blue. The question of substance and Its existence need not concern us here; the problem at Issue Is that of a descriptive statement, 2.6. Then che total structure of a language is submitted to the expression-content dichotomy, we Can avoid the use of the concept of the morpheme as it has been employed, as a unit of both "meaning11 and "form”; the resulting ambiguity is that; which handicapped Wharf. Hjelmsløv*s use of the tern sign, as a relation rather than an entity, avoids this difficulty, Problems In the use of the concept of the morpheme us it now stands are receiving attention at the present time. The distinction between content and expression provides us with the formal basis for the description of two distinct sets of phenomena, units and relations of the two planes of language. And the distinction between these two sets of phenomena is precisely that which Is relevant to the questions which Interested Whorf.
6i
3* Instrument and Observation. 3*1, In the following discussion It is asserted that the relation- ship noted by vVIiorf between the language spoken by a given group and their cult re Must necessarily exist; and that this relationship is between the concent of the language and the culture, and is independ- ont of the expression. With the distinction between content and expression, Whorffs thesis can be seen to be trivially true, in the sense that for any given utterance, there is a necessary relationship between the message and the content of the code, and toy definition an arbitrary relationship between the message and the expression of the code. There is a necessary relationship between the content plane of a language and the culture of the group which uses that language in the sane way as there is a necessary relationship between the postu- letes and the observations of the scientist. 3*2, The necessity of such a connection may be observed if we return to Whorf’s starting point, the connection between talking and behavior observed in a particular situation. In the case of the ‘‘empty*5 g&so- line drums which actually contained an explosive gas, the pertinent fact is that a judgment 15empty” was made, and a message vempty" was communicated. This judgment and message were a part of the experience of those participating in the incident, and as previous behavior may be directly related to subsequent behavior. In a given situation, th® talking done by the peoplo involved registers and communicates their experience in that situation, direct or indirect, implicit or explicit. It embodies agreements established or in process. The structure of the content plane of the language spoken is a common heri- tag© of hypotheses and conclusions shout the universe, a system of distinctions and generalizations which provides for the recognition of certain relations between events, and has had a certain survival value. 3.3. Studies of the psychology of perception reveal the importance of " set” or 11 orientation". Saplr compares the role of language in this respec; to a set of geometrical axes: "The world of linguistic forms, held within the framework of a given language, Is a complete system of reference, very much as a number system is a complete system of quantitative reference, or as a sot of geometrical axes of coordinates i3 a complete system of reference to all points of a given space. The mathe- matical analogy is by no means as fanciful as It appears to bo. To pass from one language to another is psychologically parallel to passing from one system of reference to another. The environ- Ing world which is referred to is the same for either language; the world of points is the same for either frame of reference." (6) The difficulties which the translator faces in trying to find equiva- lent units between non-equivalent systems are well known* The
7V
organization of experience provided by a language, as Wharf has rightly emphasized, is analogous to the theoretical framework with which the scientist makes an observation} Indeed, the scientific scheme la merely a special case. The language spoken toy a given group provides it with a classification of events about it which en~ atoles it to deal more or less successfully with those events. In the same way, the postulates of the scientist provide a more or less fruitful organization of the material he considers. This Is not to imply that there is no ordering of events external to the observer, tout merely that he perceives events only in terms of previous obser- vations, no matter how long or complex the chain of analogies and syntheses may toe. 5.4. But whereas the scientist is optimally conscious of the rela- tionship between his instruments end his observations, the speakers of a language are in no such position. But some kinds of linguistic change, such as the acquisition of a new vocabulary item, may be doa- critoed as o. reclassification of events toy those participating, by means of a new agreement, which brings one set of observations more into line with another. A language as the medium by which a culture is trans- Blitted aynchronioally and diachronioally must of necessity embody I iplicltly or explicitly the world view which is the heritage of the group. Otherwise, the language could not have served as a means of communication for the group developing a given culture, nor would it allow the group to function at the time of any given observation of the group. The consideration of an acculturation situation will enable us to further examine this point, as well as Whorf’s qualification that a language and a culture must be together for some time before such a connection can be expected. If a language and a culture are newly juxtaposed, say by conquest, then a new culture is brought in with the new language. Insofar as the new language Is adapted to use in previ- ously existing cultural situations, there would of course be an entirely new language-«-that is, a new content structure would be es- tablished which would correspond to the communication needs of the group. Such a dialect would be distinct on the ba3is of the content features, irrespective of differences on the expression plane. An extreme example might be one of the Pidgins developed where there has been culture contact across a wide cultural gap. Or we may alter Whorf’s Incident of the "empty1* gasoline drums as follows: the workers in a warehouse know that certain storage drums are empty of gasoline but may contain an explosive gas, and they behave accordingly. But to a newcomer the message "empty" means not "empty of gasoline, but possibly containing an explosive gas", but simply II empty’’'. He may then behave Inappropriately, what Is Involved here is not a limitation of the language Involved, but a failure to common- Icate. The new worker does not speak the dialect of the group; he Is not acquainted with the significant distinctions subsumed in the con- tent of the 3lgn empty in the present context.
83.5. Hjelmslevts distinction between content and expression, then, enables us to define precisely the aspect of language to which culture, is related, that is the content, and the necessity of this relation- ship. It also enables us to isolate the aspect of language which is Independent of culture, that is the expression. By definition the function between these planes is an arbitrary one. This distinction will al30 be helpful in considering the question of possible "prImi- tlvo" languages. The content plane of a language spoken b; a group possessing an unsophisticated science and cosmology would reflect such a level of development at a given time, while it would be adaptible to future change, just as any set of formulations of the scientist Is. Ho limitations imposed by the expression plane are foreseen, since studies of redundancy indicate that a language has many times more machinery in the expression plane than could conceivably be exhaus- tlvely utilized, 3.6, Sapir has an interesting statement of the independent dlstribu- tlon of features of the expression and content planes: " o may consider the subject matter of morphology as made up of certain logical or psychological categories of thought that re- ceive grammatical treatment and of formal methods of expressing these. The distinct character of these two groups of morphologi- cal phenomena may be illustrated by pointing out that neighboring languages may influence, or at any rate resemble each other in the one set without necessary corresponding Influence or reseat- blanoe In the other. Thus, the device of reduplication is wide- spread in American Indian languages, yet the concepts expressed by this method vary widely. Here we deal with a widespread formal device as such. Conversely, the notion of inferential activity, this is, of action, knowledge of which is based on in- ference rather than personal authority, is also found widely expressed in American languages, but by means of several distinct formal processes. Here we deal with a widespread grammatically utilized category of thought as such,” (7) The expression plane of a language is formally independent of the culture; this does not exclude the historical association of certain cultural traits and certain expression features. The determinants of the features of both planes of a language v/ould bo those subsumed under comparative and language contact studies; i. e., developmental and. dlffusional. 4. Lexicon and Morphology. 4.1, It may bo objected that the present discussion has dealt with lexicon rather than morphology, and that the connection Whorf sought to establish with culture concerned the latter. Most of the time Whorf speaks informally of the ’’grammar" of a language as the means by which experience is ordered, but he was aware that such a division
9of language with respect to this question was not valid, as la shown In the passage concerning "fashions of speaking" quoted above (1*4*1.). •This statement indicates that tøhorf would have rejected a description of different content structures as different vocabularies, differing only In a linear fashion with respect to the presence or absence of particular items. Whørf was concerned with the structured nature of the content plane of a given language; it is precisely the operation of this plane which interested Whorf. 4*2. However, it is in terms of vocabulary that the accomodation be- tween language and culture has been commonly recognized* It is usually said that a relationship between language and culture may be observed in the lexicon, but not in the morphology. A lexicon from any speech community is expected to reflect significantly the interests and empha- sea of the group* here are adduced the proliferation of Eskimo terms for snow, and Southwestern terminology for environmental features. Such a statement assumes that the items listed in a lexicon all stand in the same relationship to one another, distinguishable by their identity, much in the same way as they occur as entries. But minimal signs are related to one another in complex ways which reflect the content struc- ture of the language much as the taxonomist’s chart reflects the distinctions and generalizations lie ha3 chosen to employ. Therefore a lexicon would ideally consist of a content analysis of a language of the sort Hjelmslev has recommended. 5. Conclusions. 5.1, That Whorf’s "fashions of speaking" may be identified and submitted to formal analysis as the content structure of a given language. 5.2, That the connection postulated by UShorf between language and oul- ture must necessarily exist, since the content structure of a language bears the same relationship to the experience of those using it as that between the postulates and observations of the scientist. 5*3. That Hjelmslev’s distinction between expression and content enables us to describe the nature of such a connection, which is independent of the particular expression features employed. This conceptual scheme is based on the sign, as a function, rather than the morpheme, as an entity.
o-
10Bibliography 2 o xv of rhorf’a .-najor essays have been reprinted by the Foreign Service Institute, Department of State, Washington,_D.. .0*, under tlx© title Four Articles on Motallngnfstlcs> Apt --I, j.950* Included here are:
2. 3. 4. 5. 6.
i!: cience and Linguistics'1, Teen* peview. Vol. xlxt, no* 0 (1940) uLln^u.1311 o3 bs an xacfc Sc l one6” B Toon» ‘hvjvlow § ‘/ol* xlI ii, no * " 2 (1940) "Lan uages and Logic4', Ieoh, Review. Vol. xliii, no. o (1941) Relation of Habitual 'Thought and Behavior to Language Language. Culture, and Personality. Henasha, Wise*, 1941. page numbers given here are those of the FSI reprinting. In addi- tion, as of Interest here, may be mentioned: 14An American Indian Model of the Universe” »XJAL, Vol. ~’Vl, no, ^
(1950) pp. 67-72 n hie Punctual find Segmentative Aspects of Verbs in liopi” , Lg.. Vol. xli:127-151 (1956), This contains one of the strong- ggr statements concerning the scientist and his language, "Some Verbal Categories of ,lopiw, Lg*, Vol. xiv:275-236 (1938) ”Science and Linguistics”, p. 5 (FSI)
»■-.■he Relation of abitual Thought and Behavior to Language”, p. 43,(.31)
»"-he halation of Habitual Thought and Behavior to Language”, pp. 44-45 (FSI)
Louis Hjelms lev, Omkring sprogteoriens grundlag.&olso, Copenhagen, 1943. The writer has made use of a. draft translation by P. J. ibitfield, distributed for classwork at the Linguistic Institute, Indiana University, Summer 19o2, and is solely responsible xor the attempted summary and interpretation.
award Sapir, "The Grammarian and his Language”, reprinted In lected writings of Edward Sapir. David Mandelbaum, od., Cnlv, <3aT* Press, 1949, p. 153
of
Edward Sapir, ’‘Language and Environment”, as above, pp# 97-98
7