Volume 16 Number 1
October 1976

Rodney A. Clifton

LANGUAGE, as a means of communication, is so important for the socialization of individuals that major linguists, such as Edward Sapir (see Reference 22) and Benjamin Lee Whorf (see Reference 27) have argued that the native language spoken by an individual shapes the way in which he thinks about reality. The reality an individual observes is organized and becomes meaningful entirely in terms of the specific language he speaks.

This argument does not say that an individual can only perceive those aspects of reality which are allowed by his native language. A wide range of reality may be perceived through the various sense perceptors, but the interpretation of that reality is made in terms of a specific language. It is commonly known, for example, that Eskimo has a multitude of words and expressions for the concept "snow?’ while English has only a single word. A person who speaks only English may observe different types of snow, but his language does not allow him to classify snow in a manner in which the Eskimo language allows.

Research by some anthropologists and linguists has provided considerable ethnographic evidence which supports this argument (see References 4, 9, 15, 16, 17, and 29). Specifically, evidence exists (see References 10, 11, and 18) that a group’s language determines various aspects of its cognitive development. These studies would suggest that children from non-European language groups have different types of cognitive development than children from European language groups.

The British socialinguist, Basil Bernstein (see Reference 2) has extended the argument that language shapes the manner in which individuals interpret reality. His thesis is that there are great differences in both the "organization of words and structural connections" (semantic structures) between the middle and lower socio-economic classes in the British society. Bernstein goes on to argue that lower-class students do not succeed in school precisely because they do not use the same organization of words or semantic structures as middle-class teachers and students.

Even though Bernstein has distinguished between the structure of a language (semantics) and the organization of words, most of his research has been directed towards examining the organization of words into what he calls the elaborate and restricted language codes. His research shows that students from lower socioe-conomic groups use a restricted code whereas students from upper socio-economic groups use an elaborate code (References 5 and 6).

If Bernstein’s analysis is valid for the British society, what must be the fate of many Indian and Eskimo children who are enrolled in Canadian schools without a conception of either English or French? Surely, the lack of success of these children may also be explained in terms of the differences between their native language and the language of instruction.

The present study is designed to examine the semantic structures of a sample of Cree-Indian and non-Indian junior high school students. Since there is some evidence that the sexes also have certain differences (see Reference 1), the semantic structures of both sexes will be examined separately.

The Instrument. Osgood and his colleagues have developed a theoretical orientation and a measuring technique, called the semantic differential, which has had a considerable degree of success in defining the semantic structures of individuals from a number of different cultural and linguistic groups (see References 13, 19, 20, 21, 24, and 25). This research suggests that the semantic structures of a variety of languages are similar. In a number of studies three major dimensions of the semantic structure, evaluative dimension, potency dimension, and activity dimension have been found.

However, it would be immature to claim that these three dimensions are universal to all language and cultural groups simply because many groups, including many North American Indian tribes, have not been examined. A close inspection of the evidence reported by George Suci (see Reference 23) reveals that the semantic structures of samples of Navajo, Zuni, and Hopi subjects are quite different from the semantic structures of Euro-American subjects.

In this study, 13 bipolar adjective scales (good-bad, beautiful-ugly, large-small, kind-cruel, fast-slow, pleasant-unpleasant, strong-weak, nice-awful, active-passive, happy-sad, heavy-light, hot-cold, and fair-unfair) were factor analyzed into three factors which would, theoretically, represent the three major dimensions of semantic structure. The purpose of the factor analysis is to organize the bipolar adjective scales into a minimal set. Each concept will tend to have similar ratings on the scales which evoke similar mediational processes. If two cultural or linguistic groups locate concepts in similar positions of the semantic space then they have similar semantic structures.

Twelve concepts (school, books, English, teacher, discipline, examination, reading, homework, studying, learning, me, and Cree) were judged by the subjects. Even though most of these concepts are related to education, the semantic structure is not dependent directly upon the concepts evaluated. Previous studies have shown that words, such as these, elicit stable results (see References 19 and 24).

The Sample. The subjects for this study were drawn from the 407 students enrolled in the only junior high school in a small city in central Alberta. The total population of Indian students, 28 males and 25 females, and a 34% sample of the non-Indian students, 60 males and 60 females, were selected. Of the Indian student population, 10 males and 10 females were enrolled in grade seven, 9 males and 10 females were enrolled in grade eight, and 9 males and 5 females were enrolled in grade nine. For the sample of non-Indian students, 20 males and 20 females enrolled in each of grades seven, eight, and nine, were selected by a stratified random procedure. Only a portion of the non-Indian population was selected for the analysis so that the sub-sample would be relatively similar in size. The students ranged in age from 12 to 16 years with no significant differences between the sexes or the ethnic groups within each grade level.

The Indian students were all Crees from a reservation in central Alberta. These students came from homes in which Cree was the primary language of oral communication. The non-Indian students were from a small city and the surrounding farming area and were mainly from English-speaking homes. A very small percentage of the non-Indian students came from homes in which another language was also spoken.

The results of the factor analysis are presented in Table 1, and the coefficients of factorial similarities are presented in Table 2.

From the factor loadings reported in Table 1, the first factor is readily interpretable as evaluative. For both males and females from both ethnic groups, seven scales (good-bad, beautiful-ugly, kind-cruel, pleasant-unpleasant, nice-awful, happy-sad, and fair-unfair) have high loadings on Factor I (Fi) and low loadings on Factors II (Fii) and III (Fiii). It is evident from this table that Factors II and III are not as readily interpretable. That is, scales such as large-small, have high loadings on Factor Il for one group of students (Indian females) and low loadings for another group (Indian males). The potency and activity factors seem to be infused into a general "dynamism" factor, and the infusion is only partly resolved by rotating the factor structures for maximum congruence (table not reported).

Other similarities are also evident from this table. For example, the communalities (h²) are very similar for each group of students on each scale. Thus, for the scale good-bad, the communalities range from .55 to .63, which means that for this scale the three factors explain between 55% and 63% of the variance. It is also evident that the sum of square loadings and the percent of total variance explained are also very similar for all four groups of students. Between 32% and 38% of the variance of all 13 scales is explained by Factor I, between 12% and 16% of the variance is explained by Factor IT, and between 9% and 12% is explained by Factor III.

The coefficients of factorial similarities, Table 2, provide strong evidence that the factor structures of all four groups have a very high degree of similarity. As expected from Table 1, the first factor (evaluative) has the highest degree of similarity. All of the coefficients of factorial similarities for this factor are above .99. Even though the other two factors have lower coefficients, Factor II has coefficients ranging from .94 to .99 and Factor III has coefficients ranging from .88 to .96, a great degree of similarity is evident.

The overall results of the analysis clearly demonstrate the cross-cultural, as well as cross-sexual, similarities of the semantic structures of a group of Indian and non-Indian junior high school students. The most salient and consistent dimension of the semantic structure was the evaluative dimension. The remaining two dimensions, activity and potency, were more ambiguous. However, the coefficients of similarities indicated that there was a very high degree of similarity on all three dimensions for all four groups of students. The findings of this study do not support an extension of Bernstein’s theory which would hold that the semantic structures of the Indian and non-Indian students would differ.

These findings are significant in terms of the future of social research and the education of Indian students. In general terms the study suggests that the assumption that different sectors of society understand a language in a similar manner has, at least, some degree of validity. Beyond this, it is suggested that even though Indian children come to school speaking a language which is very different from English, and continue to speak this language in their home, by the time they reach junior high school they use semantic structures which are very similar to those used by their non-Indian classmates. As a result, teachers and administrators must not simply blame the lack of success of Indian students upon their inability to use the English language. Indians may not use the same range of English words as their non-Indian classmates, but they do use the same semantic structures.

The findings suggest that Indian students learn English much better than they learn to play the social role of "student." This interpretation is made in terms of the findings of the previous research which showed that Indian students have a difficult time adjusting to the norms and values of the school. This is particularly interesting when one realizes that norms and values are, in most cases, dependent upon a language for their dissemination. Future research should be directed towards examining the inter-relationships between language and values in the socialization of Indian students.

1. Barron, Nancy. "Sex-Typed Language: The Production of Grammatical Cases." Acta Sociologia, 14:1-2, 24-42 (1971).

2. Bernstein, Basil. "Social Class and, Linguistic Development: A Theory of Social Learning." In Education, Economy, and Society, Eds. A. H. Halsey, J. Floud, and C. A. Anderson. Glencoe, IL: The Free Press (1961).

3. Bernstein, Basil. "Elaborated and Restricted Codes: Their Social Origins and Some Consequences." American Anthropologist, 66:55-69 (1964).

4. Conklin, Harold C. "Linguistic Play in Its Cultural Context." Language, 35:631-636 (1959).

5. Deutsch, Martin. "The Role of Social Class in Language Development and Cognition." American Journal of Orthopsychiatry, 35:78-88 (1965).

6. Entwisle, Doris R. "Semantic Systems of Children: Some Assessments of Social Class and Ethnic Differences." In Language and Poverty: Perspectives on a Theme, Ed. Frederick Williams. Chicago: Markham (1970).

7. Evans, Glen T. "Congruence Transformations: Procedures for Comparing the Results of Factor Analyses Involving the Same Set of Variables." Multilithed. Toronto: Ontario Institute for Studies in Education (1970).

8. Evans, Glen T.Transformation of Factor Matrices to Achieve Congruence." British Journal of Mathematical and Statistical Psychology, 24 (May):22-48 (1971).

9. Frake, Charles O. "The Diagnosis of Disease Among the Subanun of Mindanao." American Anthropologist, 63:113-132 (1961).

10. Greenfield, Patricia M. and Jerome S. Bruner. "Learning and Language." Psychology Today (July): 40-43, 74-79 (1971).

11. Hess, Robert D. and Virginia C. Shipman. "Early Experience and the Socialization of Cognitive Modes in Children." Child Development, 36:869-886 (1965).

12. Kaiser, H. F. "The Varimax Criterion for Analytic Rotation in Factor Analysis." Psychometrika, 23:187-200 (1958).

13. Kumata, Hideya and W. Schramm. "A Pilot Study of Cross-Cultural Methodology." Public Opinion Quarterly, 20:229-238 (1956).

14. Lee, Dorothy D. "The Linguistic Aspect of Wintu Acculturation." American Anthropologist, 45:435-440 (1943).

15 Lee, Dorothy D."Notes on the Conception of the Self Among the Wintu Indians." Journal of Abnormal and Social Psychology, 45:538-543 (1950a).

16. Lee, Dorothy D. "Lineal and Nonlineal Codifications of Reality." Psychosomatic Medicine, 12:89-97 (1950b).

17. Lounsbury, Floyd G. "A Semantic Analysis of Pawnee Kinship Usage." Language, 32:158-194 (1956).

18. Oestereicher, Emil. "Toward a Sociology of Cognitive Structures." Social Research, 39:134-154 (1970).

19. Osgood, Charles E. "Studies in the Generality of Affective Meaning Systems." American Psychologist, 17:10-28 (1962).

20. Osgood, Charles E. "Semantic Differential Technique in the Comparative Study of Cultures." American Anthropologist, 66:171-200 (1964).

21. Osgood, Charles E., George J. Suci, and Percy H. Tannenbaum. The Measurement of Meaning. Urbana: University of Illinois (1957).

22. Sapir, Edward. "The Status of Linguistics as a Science." Language, 5:207-214 (1929).

23. Suci, George J. "A Comparison of Semantic Structures in American Southwest Culture Groups." Journal of Abnormal and Social Psychology, 61:25-30 (1960).

24. Tanaka, Yasumasa, Tadasu Oyama and Charles E. Osgood. "A Cross-Culture and Cross-Concept Study of the Generality of Semantic Spaces." Journal of Verbal Learning and Verbal Behavior, 2:392-405 (1963).

25. Triandis, Harry C. and Charles E. Osgood. "A Comparative Factorial Analysis of Semantic Structures of Monolingual Greek and American College Students." Journal of Abnormal and Social Psychology, 57:187-196 (1958).

26. Tucker, L. R. A Method for Synthesis of Factor Analysis Studies. Personal Research Section Report No. 984, Washington, D.C.: Department of the Army (1951).

27. Whorf, Benjamin Lee. "Science and Linguistics." The Technology Review, 42:229-231, 247-248 (1939).

28. Whorf, Benjamin Lee. "Language and Logic." The Technology Review, 43:250-252, 266-272 (1940).

29. Witherspoon, Gary J. "Navajo Categories of Objects at Rest." American Anthropologist, 73:110-127 (1971).

30. Wrigley, C. F. and J. O. Neuhaus. "The Matching of Two Sets of Factors." American Psychologist, 10:418-419 (1955).