Identity and social distance: towards understanding Simmel's ‘The Stranger’

Identity and social distance: towards understanding Simmel’s ‘The Stranger’*

LEO DRIEDGER / University of Manitoba

JACOB PETERS I University of Winnipeg

On fait repandre dans cette etude le concept de Simmel de ‘Der Fremde’ (c’est-a-dire, I”Etranger’), m2me jusqu’aux communautes du type ‘etranger’ que developpent les minorites dans leurs enclaves ethniques. Lorsqu’il introduit I’idee generale de ‘distance sociale,’ Simmel pose, parmi d’autres rapports, I’existence d’une association positive entre, d’une part, I’identification avec la solidarite du groupe primaire et, d’autre part, I’augmentation de la distance sociale d’avec les groupes secondaires. On demontre dans cette etude que, plus l’identite du groupe primaire ethnique s’accroit, plus s’accroit egalement la distance sociale d’avec les groupes secondaires. Pourtant il en existe des variations. Quand on essayait d’expliquer ce phenomene de distance sociale on trouvait que I’identite ethnique en etait, elle, une mesure plus significative qu’etait celle du rang socio- economique. Lorsque I’on comparait, les uns aux autres, les individus que I’on pourrait appeler ‘high identifiers’ (c’est-a-dire, ceux ayant tendance a s’identifier etroitement avec I’ identit6 ethnique du groupe) dans chacun des groupes ethniques, on constatait que I’origine ethnique elle-m2me existait comme facteur pour expliquer la distance sociale. C’etaient surtout les FranGais qui, eux, ne se conformaient pas au modele du type ‘identification agrandie’ (menant une) ‘distance sociale agrandie.’ Ces Ctudiants fransais demontraient tres peu de distance sociale ; et l’on trouvait bien peu d’evidence pour faire reveler des prejuges chez eux, - quoiqu’ils preferassent se marier avec les Europtiens plutbt qu’avec les non-Europeens. Plusieurs parties de cette etude viennent a l‘appui, dans leur fond, de I’af!irmation de Simmel : la solidarite du groupe primaire se rapporte, en fait, a la distance sociale d’avec les groupes secondaires. D’ailleurs celui-ci reste un phenomene complexe qui exige : une conceptualisation bien plus precise ; des moyens de mesure plus sensibles ; et enfin, une classification en plus, et de I”identit6’ et de la ‘distance,’ ces deux variables que l’on considere dans la prisence etude.

This paper extends Simmel’s concept ‘Der Fremde’ (the stranger) to stranger-type communities which minorities develop in ethnic enclaves. When Simmel introduced the concept of social distance, he posed, among other relationships, the existence of a positive association between identification with ingroup solidarity and an increase in social distance from outgroups. This study of Canadian university students demonstrates that as ethnic ingroup identity increases social distance from outgroups also increases. There are some variations, however. Ethnic identity was more important in explaining social distance than was socioeconomic status. When high identifiers within each ethnic group were compared it was found that ethnic origin was a factor. The French especially did not conform to the high-identity-high-distance pattern. The students demonstrated relatively little social distance, and there was little evidence that students were prejudiced, although they preferred to marry Europeans rather than non-Europeans. Some parts of the study support Simmel’s contention that ingroup solidarity is associated with social distance from outgroups, but it is a * This research was made possible by grants from the Canada Council (69 1445) and the University of Manitoba Research Grants Committee, whose support is gratefully acknowledged. Steven Brickey provided useful statistical advice and Charles Axelrod made helpful comments.

Revised paper presented to the 1974 Annual Meetings of the Canadian Sociology and Anthropology Association in Toronto. August 23-26.

Rev. canad. Soc. &Anth./Canad. Rev. SOC. & Anth. 14(2) 1977

Identity and social distance / 159

complex phenomenon which requires more precise conceptualization, more sensitive means of measurement, and further sorting out of nearness and farness (identity and distance) factors to which this study addresses itself.

I N T R O D U C T I O N

Donald Levine et al. (1976:839), when discussing Simmel, suggest that he struck a rich vein of sociological inquiry when he proposed the concept of ‘Der Fremde’ - the stranger. The concept spawned new inquiries into such areas as marginal man, the sojourner, the newly ar- rived, and social distance, which early Ameri- can sociologists, many of whom studied under Simmel, developed. The interest of this paper is to extend Simmel’s writings on the stranger, to stranger communities, which minorities develop in ethnic enclaves. Fallers suggests that ‘stranger’ communities develop ‘socially segre- gated and hostilely-regarded communities of kinship units, knit together and defended by associational ties’ (Fallers, 1967:8, 12-13). This suggests the need to study the relationship between the two factors of identification within an ingroup, and the ingroup’s potential hostile perception of outgroups as a result of its ingroup identification.

When Simmel (1950:g) introduced the concept of social distance, he posed, among other relationships, the existence of a positive association between identification within ingroup solidarity and an increase in social distance from outgroups. Although the problem sounds in- triguing, we soon found that it was difficult to select factors of ethnic identity independent of outgroup social distance factors. It seemed that the statement was a tautology which simply correlated opposite sides of the same coin. The task in this paper is to test the independence of some ingroup ethnic identity factors with those factors related to social distance from outgroups. This involved the selection of ingroup ethnic identity and outgroup social distance factors which are independent. It also includes testing whether or not those who score high on independent ingroup identity factors do, indeed, score high on factors which indicate social distance from others.

Often the generalizations of early theorists, like Simmel, do not isolate the specific components of the independent and dependent variables (ethnic identity and social distance in this case). Furthermore, some of the factors within

the major components often overlap. A review of the literature demonstrated that factors such as endogamy and choice of friends were used in both the Bogardus social distance scale (as nearness factors), and the Driedger (1975b) ethnic identity index. Obviously a correlation of these two indexes (which contain two similar variables, endogamy and choice of friends), has built-in variables which are not mutually exclu- sive. Therefore, the correlations will be higher because of the tautology.

The task was to isolate theoretically independent components which could be tested. A closer examination of the Bogardus social distance scale shows that the items are all concerned with the attitudes towards associations with others. On the other hand, a survey of ethnic identity scales showed that items were selected from a variety of components including association with others, identification with a specific ingroup culture, identification with ingroup institutions, social psychological identity, and ideology. Since the associational identity component overlapped with the Bogardus social distance scale, we eliminated associational factors, and chose to include in the measure of ethnic identity only identification with ethnic culture and institutions. More specifically, the problem became whether or not those who identified strongly with an ethnic culture and ethnic institutions would be less willing to interact socially with others. Do ethnic subcultures and ethnic institutions provide boundaries which make such individuals less willing to marry outside the group, less willing to choose outgroup friends, outgroup neighbours, or even outgroup working partners? Furthermore, in extreme cases, does ethnic cultural and institutional identification predispose some individuals to reject others to the extent that they wish to ban them from their country? Specifically the problem is to test whether strong ethnic institutional and cultural identification predisposes individuals to as- sociate less with others outside their group.

Simmel’s ‘The Stranger’ Simmel’s concept of ‘the stranger’’ is a rich one

I We are indebted to Donald N. Levine, Ellwood B. Carter, and Eleanor Miller Gorman for their discussion of ‘Simmel’s Influence on American Sociology, I , ’ American Journal of Sociology 8r:q. Their treatment of Simmel’s ‘Der Fremde’ and social distance encouraged us to reread Simmel in the original German, which was most worthwhile.

160 / L. Driedger and J. Peters

because it raises major questions, for example, the question as to how a person can retain within his own social psychological world a ‘ground’ of identification, as Lewin (1948) would put it, and at the same time relate se- curely to others. We would expect that when the stranger enters the strange environment of others, he is secure only if he is grounded in a reference group, or if he is socially and psychologically motivated by the norms of such a group. This raises many related questions, such as: what must be the nature of such a reference group; how often and how much must an individual contact the reference group to sustain his separate identity; what must be the quality of such sustenance; and, whether an individual can make only occasional forays into the strange world or whether he is able to work there consistently with occasional refresher periods in his own reference group. Park and other assimilationists dwelt on the seduction of the majority industrial world with little reference to Simmel’s other dimension of ‘the stranger’ - countervailing forces of identification. I t is unfortunate that for so long we have ignored Simmel’s ‘nearness’ or reference group dimension, because Simmel always saw these two polarities as constantly interacting.

It is important to see identity and distance in the total context of Simmel’s view of social phenomena, as Levine et a1 suggest:

Simmel’s formulation that the stranger repre- sents a combination of nearness and distance, far from being a logical flaw, aptly illustrates his more general assumption that all social phenomena reflect a combination of opposed tendencies. This has the interesting methodological implication that many phenomena would be more accurately measured if scales tapping two opposing dimensions were employed (Levine et al, 19762339).

‘The various ways in which Simmel employs the symbolism of simultaneous closeness and remoteness in “The Stranger” in fact provide a basic insight about strangerhood which is of great use in developing a more rigorous and systematic sociology of the stranger’ (Levine et al., 1976). Just as Simmel stimulated American sociologists to find ways of measuring distance, so we have been stimulated to isolate identity as an opposite measure of Simmel’s discussion on ‘the stranger’ in order to gain some insight into what the relationships may be.

The two opposite dimensions we have selected for examination are ethnic ingroup identification and social distance from outgroups. The identity polarity suggests elements which agroup holds in common with an ingroup as measured by the Driedger (1975b) ethnic behavioural cultural identity (ECBI) index. Dis- tance will focus on subjective social distance such as attitudes towards others (belonging to ethnic groups), and is measured by the Bogar- dus social distance scale. We are quite aware that the identity index measures only the cultural identification dimension, and the distance scale measures only the subjective interactive (Kadushin, 1962) dimension of distance. Both poiarities are heterogeneous concepts; we are tapping only one dimension of each concept to make them manageable.

Ethnic identity Lewin (1948) proposed that individuals need a firm clear sense of identification with the heritage and culture of their ingroup in order to find a secure basis for a sense of well-being. Iden- tification with a group implies commitment to an ingroup, which in turn means willingness to enter into intimate interaction with the members of the group. This in Simmel’s terms would indicate social nearness.

It is often assumed that ethnic identity can best be maintained when the ingroup develops a social system of its own with control over its institutions, so that the interaction patterns of the group will take place largely within the system. Such patterns will lead to the creation and maintenance of boundaries and control over systematic linkage. Breton (1964), suggests that religious, educational, and welfare institutions are crucial, while Joy (1972) adds the impor- tance of political and economic institutions. Vallee (1969) and Driedger and Church (1974) confirmed these claims by Breton and Joy by summarizing the need for organization of group structures and institutions which influence socialization and ethnic community decision- making. These studies assumed that residence within ethnic communities will result in much time spent within the ethnic institutions and cultural enclave, with less time for association with others.

The elements of ethnicity that are important for ethnic identity need to be defined, especially if we accept that there may be a multiplicity of factors (Nahirny and Fishman, 1965; Lazer- witz, 1953; Driedger, I975b). Many researchers


who have expounded on ethnic identity have been concerned with cultural identity factors (Driedger, 1975b; Geismar, 1954; Isajiw, 1974; Lazerwitz, 1953; Levinson, 1962; Rinder, 1959; Segalman, 1967). Segalman’s extensive review of the identity literature shows that most of these scholars have developed a multidimensional approach to ethnic identity.

Religious attendance, participation in parochial education, membership in ethnic organizations, use of an ethnic language, endogamy, and choice of friends are cultural identity factors often cited in the literature (Driedger, 1975b; Geismar, 1954; Lazerwitz, 1953; Lieberson, 1970; Nahirny and Fishman, 1965; Rinder, 1959; Royal Commission, 1970; Segalman, 1967). For the purpose of this study, four of the six cultural and institutional factors used in the ethnic identity inventory developed by Driedger (1975b) will be employed. Endogamy and choice of friends cannot be used since they are associational factors which will be used to measure nearness in the social distance scale.

We expect that students who identify strongly with their ethnic ingroup will be drawn into an ethnic milieu where they will wish to participate in their ingroup churches, schools, and organizations, use their ethnic language, choose a majority of their best friends from their group, and marry into their own group. Their ethnic institutions will occupy their time and shape their values to the point where interaction with others will be minimized. Identification with the ingroup will leave less time for primary associations with others, which in turn will result in attitudes less favourable towards outsiders. On the other hand, as less time is spent within the ethnic community, and less identification takes place within ingroup institutions and individuals, the students will look to outsiders more for friendships and potential mates in community schools and organizations.

Social distance Levine et al. (1976), say that ‘Simmel’s utiliza- tion of the metaphor distance was by no means restricted to his pages on “the stranger”; it constitutes a pervasive and distinctive feature of his sociology as a whole.’ They summarize Simmel’s meanings as: ( I ) ecological attach- ment and mobility; (2) emotional involvement and detachment; and (3) the extent to which persons share similar qualities and sentiments. Simmel himself thought that distance could be expressed in many ways. Recent work on social

distance has attempted to sort out the many meanings of the concept and to devise ways of measuring the concept. Westie (1959) included the dimensions of residential distance, position distance, interpersonal physical distance, and interpersonal interaction distance. Banton (1960) distinguished four forms of social distance (attitudinal, positional, qualitative, and ecological). Kadushin (1962) speaks of norma- tive distance, interactive distance, cultural or valuational distance, and personal distance. Laumann (1965) has sought to measure subjective social distance and objective social distance. Thus, it is apparent that the metaphor ‘distance’ can be widely used and must be adapted to specific study situations.

Bogardus (1959) expanded Park’s personal dimension of social distance and chose to use ‘the degree of sympathetic understanding that functions between person and person, between person and group, and between group and group as his measure of social distance. Sym- pathy refers to feeling reactions of a favorably responsive type, and understanding involves that knowledge of a person which also leads to favorably responsive behavior . . .’ Bogardus (1925) selected one component of Simmel’s no- tion of social distance and substituted an operational continuum for Park’s (1923) informal ob- servations.

The Bogardus scale has been widely used, and Krech, Crutchfield, and Ballachey (1962: 154) note that ‘with appropriate modifications, this type of scale can be adapted to measure attitudes toward any category of persons.’ Only a few of the research contexts in which it has been used include attitudes to politics (Best and Sohner, 1956), urban studies (Turbeville, 1950; Mitchell, 1955), personal identity (Derbyshire and Brody, 1964), social status (Ellis, 1956), social class (Landis, Datwyler and Born, 1966; Triandis and Triandis, 1962), religion (Bogar- dus, 1967; Bealer, Willits, and Bender, 1963; Photiadis and Biggar, 1967), stratification (Wes- tie, 1959; Martin, 1963; Laumann, 1965). We wish to use the Bogardus scale to measure the associational component, the social distance of personal relationships.

The Bogardus social distance scale has been used in scores of studies in countries around the world. However, very little use has been made of the scale for social distance research in Canada. The Bogardus scale will be employed as a means of assessing the attitudes of Man- itoba university students towards ten ethnic groups other than their own.


Social distance and ingroup differentiations Two ingroup factors which could influence the relationship of ingroup identity with outgroup distance are socioeconomic status and ethnic origin. We need to see whether ethnic cultural identification is a stronger factor in explaining the relationship between our dependent and independent factors, or whether other factors are more influential.

A first factor which needs to be controlled is socioeconomic status. Proponents of assimilation would argue that as individuals become upwardly mobile in the urbanization, indus- trialization process they come into contact with more of the outgroup, which will result in loss of identity. Education, income, and high status occupations will wean the individual away from the ingroup community. In this study we argue that socioeconomic status will be of minimal consequence. There will be high ethnic identifiers of low and of high status and these will demonstrate higher social distance than low identifiers of all status levels.

Porter (1965) assumes that eventually socioeconomic status will be the most powerful attraction for ethnic groups. Since the British are now the most powerful economic group, minorities will soon perceive their ethnic pluralism as a hindrance to attaining status and abandon their ethnicity as they seek to compete economically. The attraction of status cannot be denied, but the fact that the Jews, who are, on the average, of the highest socioeconomic status in Winnipeg, also indicate high residential segregation, cultural identity, and institutional completeness, would seem to challenge the view that a rise in status causes a decline in pluralism. A study of German and Mennonite student identity in Winnipeg (Driedger and Pet- ers, 1973), when controlled for socioeconomic status, showed that status was relatively insig- nificant compared to ethnic identity. The trend seemed to be for high status students to score somewhat higher on identity in both their attitudes and behaviour. If in this study we find that there is high ingroup identification among both the high and low status groups, then the question is whether these high identifiers of either social status also show higher social distance from other outgroups. We think that will be the case.

Although studies in Winnipeg (Driedger, 1975b) clearly show that some ethnic groups (French and Jews) have a larger number of ethnic identifiers than others (Scandinavians), it would seem logical from Lewin's proposition

that high ethnic identifiers of any ethnic group should have a secure basis for a sense of well- being based on their identification with their ingroup culture and heritage. It would follow that high ethnic identifiers of any ethnic group, no matter what their particular culture, should also indicate high social distance from others. It is not the particular culture (German, Jewish, Polish) that is the differentiating factor, but the extent of identification with an ingroup ethos that is crucial.

Proponents of assimilation would argue that as individuals become upwardly mobile they come into contact with more of the outgroup. which results in loss of ingroup identity. The Jews however, are an interesting case in point in that they are mostly urban, have engaged mainly in business enterprise, and are of high socioeconomic status involving them in many contacts with members ofthe outgroup. Studies in Winnipeg show however, that the Jews ( I ) have maintained higher residential segregation than most groups (Driedger and Church, 1974; Nicholson and Yeates, 1969); (2) they have maintained very strong religious, parochial educational, and voluntary institutions (Breton, 1964; Driedger and Peters, 1973); and (3) their cultural identity is among the highest (Driedger, 1975b). Findings for the French in Winnipeg also indicate that their segregation patterns, institutional completeness, and cultural identity are high, although their socioeconomic status is not as high as that of the Jews. We would expect therefore, that these two ethnic groups will have a greater number of members who score high on ethnic cultural identity than the Scan- dinavians who, according to the studies men- tioned, indicated relatively low identity patterns. Since every group however, contains both high and low identifiers, we need to explore whether a particular type of ethnicity will also be associated with social distance, or whether ethnic cultural identity of any sort (ir- respective of ethnic group origin) simply results in higher social distance from outgroups. We propose that high cultural identity within any ingroup will result in greater social distance, because the individual feels more comfortable within the ingroup, spends much more time within the ingroup, and therefore does not as- sociate as much with others, which makes him more prone to distance from others.

Social distance and outgroup differentiations One of the interesting questions Simmel's 'The Stranger' raises is whether or not it is possible


to maintain positive social distance from outgroups in order to maintain ingroup identity, or whether social distance will usually develop into negative distance which might be a form of prejudice or discrimination against outgroups (Mackie, 1974). Positive distance would develop as a result of preoccupation with the ingroup target, without necessarily negative consequences for the outgroup. Negative distance would be an indication of hostility or prejudice towards an outgroup target. This distinction seems important, because the one is aimed at positive ingroup community building and the other is directed at negative consequences for outgroups.

It will be interesting, therefore, to see whether university students, for example, wish to maintain more distance between themselves and those of non-European origin than of Euro- pean origin (Mackie, 1974; Hirabayashi, 1963). This introduces the racial factor. Numerous studies in the United States show that race is an important factor which influences social distance findings.

A second outgroup differentiation which students might make is that within European origin groups and within non-European origin groups. For example, are students more willing to marry northern Europeans, than southern or eastern Europeans? Or, are Manitoba students more willing to marry or be friends with blacks than Canadian Indians? If, for example, students indicate greater distance from Canadian Indians than people of British origin, is this because of prejudice or hostility? If no such outgroup differentiations are made, then, presumably outgroup origin is not a factor.

The purpose of the study is to test the following hypotheses: I / The greater the identification of an individual with an ingroup the greater the social distance from outgroups . 21 High identification with an ingroup and increased social distance from outgroups will re- main when controlled for socioeconomic status. 3/ High identification with an ingroup and increased social distance from outgroups will re- main within each ethnic group when controlled for ethnic origin.

M E T H O D O L O G Y

Fifteen hundred and sixty questionnaires were collected from seventy-six classes selected randonly at the University of Manitoba. The

sample represented 15 per cent of the under- graduate enrollment on the campus in 1971. Ninety-two per cent of the students in attendance at the seventy-six classes completed the questionnaire. In order to obtain a reliable measure of ethnic

identity, we adopted four of the items and factors used by Driedger (1975b) in his ethnic cultural behavioral identity (ECBI) index. Driedger used both attitudinal and behavioural factors and found that there was a high correlation between the two types. We used the following four behavioural factors: ethnic language use, religious attendance, parochial school attendance, and participation in ethnic organizations. The four ethnic behavioural factors were used to develop an ethnic cultural behavioural identity (ECBI) index. A score of one was assigned to each of the four factors for a potential minimum score of zero and a maximum score of four. The 820 student respondents were classified into five groupings ranging from lowest identity (score of zero) to highest identity (score of four).

The four identity factors were operational- ized in the following manner. Attendance at church or synagogue services twice a month or more by the respondent indicated activity in religion. Use of the ethnic language was deter- mined by the percentage who spoke their ethnic language to their parents at home. One or more memberships in ethnic group voluntary organizations indicated organizational activity. Any attendance by the respondent was given one point for each of the four factors where he scored, and these were summed to give each respondent a score on the index (0-4).

To measure social distance the questionnaire also included ten Bogardus-type social distance scales, one scale for each of ten ethnic outgroups (British, Scandinavian, Dutch, German, Ukrainian, French, Polish, Italian, Russian, and Jewish). Respondents were asked whether they would (I) be willing to marry, (2) be willing to be a very close friend, (3) be willing to have as a next door neighbor, (4) be willing to work beside at a job, (5 ) be willing to have as a speak- ing acquaintance, (6) be willing to have as a visitor to his nation, or (7) wish to debar from the nation persons of each of the ten ethnic groups. In the analysis we used both raw scores in the multivariate analysis and mean scores which were a summary of the ten scales for each respondent. Respondents were asked to give social distance responses for their own in- groups, but these responses were not included


in their mean scores of distance from outgroups.

The Blishen scale (1967:41-53) provided a measure of socioeconomic status. Father's oc- cupation was used to determine where the student was positioned on the scale. Students who scored 50 plus on the Blishen scale were considered to be of high socioeconomic status and those scoring less than 50 were considered to be of low socioeconomic status. Farmers were placed in the low category, although Blishen does not include them in his index.

Seven ethnic origin groups were represented in sufficient number (fifty or more) to compare ingroup differentiations, including British (157), French (86), German (16o), Jewish ( IIZ) , Polish (56), Scandinavian (61), and Ukrainian (188) students. The seven groups represented 820 students. Students from many other groups too small to include were used as a control for ethnic origin in hypothesis two.

Twenty ethnic groups were used to determine social distance outgroup differentiations. These included eleven European origin groups (British, American, Scandinavian, Dutch, Ger- man, French, Ukrainian, Polish, Russian, Ital- ian, and Jewish), and nine non-European origin groups (Mexican, Negro, Japanese, West In- dian, Filipino, Canadian Indian, East Indian, Chinese, and Eskimo). When we worked with ingroup differentiations we used only ten Euro- pean groups to control for the racial factor. However, in discussing outgroup differentiations, we added nine non-Caucasian groups so as to introduce greater outgroup differentiation potential.

F I N DI NGS

Ethnic identity and social distance by total sample The 820 respondents of the sample represented various degrees of ethnic cultural identity. De- grees of identity were measured by the ECBI Index (scores 0-4). As indicated in Table I some scored a low of zero, while others scored all the way to a high identity score of four.

In order to find whether those who indicated high identity favoured more social distance, we asked respondents to evaluate ten outgroups (British, Scandinavian, Dutch, German, Ukrai- nian, French, Polish, Italian, Russian, Jewish) by indicating whether they were willing to marry (score of I ) , all the way to whether they wished to debar such groups from the nation (score of 7). A Bogardus social distance scale

(scores 1-7) was used for each of the ten ethnic outgroups. The social distance mean scores (SD x) are given in Table I for each of the ten outgroups, by where the respondents were positioned on the ingroup ethnic identity index (scores 0-4). It was found that as ingroup ethnic identity increases, social distance from outgroups increases also, as predicted in hypothesis one.

Although the mean distance from outgroups grows consistently as identity increases (using social distance mean scores in Table I), the question remains as to whether this differentiation is significant. To determine whether or not the scores of the low identifiers (score of o on the ECBI index) were lower on the social distance scale than the high identifiers (score of 4 on the ECBI index), we ran a correlation coefficient test (using distance scale scores and identity scale scores) which resulted in an r of .1530. This is statistically significant at the .o01

probability level. We obtained the r by collaps- ing the ten scores of each individual on his rat- ing of the ten ethnic groups and dividing by ten. However, since the r score was low, the question still remains as to whether there is a sub- stantive social distance difference between the high and low identifiers. The data in Table 11 seem to indicate that the differences on the extreme ends (0 and 4 on the ECBI index), are considerable, although the differences in the middle (1-3 on the ECBI index) are quite similar, which must be one reason for the low r .

Another finding is that the social distance range is not large, as shown in Table 11 (we collapsed the distance scale for presentation only). The real difference is between those who are willing to marry into an outgroup (Bogardus scale score of I), and those who only desire close friendship with outgroups (Bogardus scale score of 2). The number of respondents who scored in the 3, 4, and 5 Bogardus scale categories was relatively small; almost none were represented in scale categories 6 and 7. This would seem to be a second reason for a low r score. Since there were so few respondents (9.4 per cent) who scored 3-7 on the Bogardus scale we collapsed all who scored above 1.5 into the category two in our analyses in the remaining tables. When we ran a correlation coefficient test using the two distance categories (uncollapsed and collapsed), the results were very similar to the previous one.

As stated in hypothesis I , high ethnic identifiers tended to consistently indicate higher social distance than low identifiers, although the


TABLE I

COMPARISON OF DEGREE OF STUDENT SOCIAL DISTANCE (USING BOGARDUS SCALE MEANS) FROM TEN ETHNIC GROUPS BY INTENSITY OF ETHNIC IDENTITY (USING ECBI INDEX)

Ethnic identity (ECBI Index)

Low High Ethnic groups 0 I 2 3 4

British SOX = 1.26' 1 . 3 3 1.41 1.55 2.06 Scandinavian 1.36 1.62 1.53 1.74 2 .22 Dutch 1.41 1.52 1.63 1.78 2.29 German 1.50 1.77 1.90 1.77 2.40 Ukrainian 1.55 1.86 1.94 1.98 t French 1.61 1.66 1.78 2.11 2.92 Polish 1.61 1.80 1.96 2.09 2.81 Italian 1.73 2.00 2.01 2.18 2.69 Russian 1.72 1.92 2.08 2.28 3 . 0 0 Jewish 1.93 2.22 2.36 2.49 3.47

* Mean social distance score of students who scored-zero on the ECBI Index; they indicated low social distance from the British (A' = 1.26). t N less than 10; all other N s are 10 or greater.

TABLE 11

RELATIONSHIP BETWEEN INGROUP IDENTIFICATION (ECBI INDEX) AND OUTGROUP SOCIAL DISTANCE (BOGARDUS SCALE)

Ethnic identity (ECBI Index)

Social distance Low 0 I 2 3 4 High Totals ~

Low 66.0 49.8 50.0 44.3 27.8 54.1 (Scale score mean below 1.5) (169) (133) (71) (39) ( 5 ) (417) High 34.0 50.2 50.0 55.7 72.2 45.9 (Scale score mean above 1.5) (87) (134) (71) (49) (13) (345) Totals 100% 100 100 1 0 0 100 100

( N = 256) (267) (142) (88) (18) (771)

r = 0.1530;P < 0.001

differences were not as great as was expected if the hypothesis was to be confidently confirmed. The major finding was that university students do not place great distance between themselves and others. It would seem that high identifiers prefer greater distance, but we need a sample where many more respondents indicate greater distance in order to test the hypothesis adequately. A sample of adults might indicate more distance. It also seems that the Bogardus social distance scale does not measure the fine social distance distinctions necessary to dif- ferentiate a population of students in Manitoba.

Ethnic identity and social distance by socioeconomic status The proposal (hypothesis 2) that ethnic identity will be a stronger factor in determining social distance than socioeconomic status was con-

firmed. The association of ingroup identification with social distance from outgroups remained the same when controlled for socioeconomic status (partial r = ,1575 in Table 111) as the uncontrolled finding (r = .1530 in Table 11). The low SES status group results were almost identical to the uncontrolled group; the high SES group was almost the same with a few variations. The question remains whether socioeconomic status would be more influential if the social distance differentiations would be greater in another sample. Our findings in this sample suggest that SES would be less influential than ethnic identification.

Although hypothesis 2 does not call for a test of the association between high ingroup identification and increased social distance from outgroups within each of the ethnic origin groups by socioeconomic status, we attempted


two extremes. Their discrepancy scores were not quite as low as that of the British, and certainly not as great as that of the French. They neither strongly confirmed hypothesis 3, nor strongly denied it. The Eta score of .31 would suggest that the hypothesis was not confirmed. Ethnic origin was a significant intervening factor.

When we explored the influence of two ingroup differentiation factors, socioeconomic status and ethnic group origin, on the increase of social distance, we found that socioeconomic status was not as influential as ethnic identity, while the ethnic origin factor did play an important role in changing that pattern. The question remains as to whether or not there might be outgroup differentiations.

Social distance and outgroup differentiations The data in Table v show that university students in Manitoba are much more willing to marry northern Europeans (British, Scandina- vians, and Dutch) than they are willing to marry non-Europeans. Four times as many are willing to marry persons of British origin than persons of Eskimo origin. How can these distinctions be explained? Is it because they wish to perpetuate northern European stock, or is it because they wish to identify with majority group members, or is it because they are prejudiced against non- Europeans?

If there was prejudice against non-Euro- peans, then we would expect many students to prefer considerable distance between themselves and non-Europeans, but this is not the case. Very few respondents scored in the 3-7 range of the Bogardus scale. About one fourth (27.5 per cent in 3-7 scale range) preferred more distance from Eskimos than marriage or close friendship would allow, while a smaller percentage desired distance from any of the other nine- teen groups. It is true that more respondents desired a 3-5 range of distance from non- Europeans than Europeans, but the percentage was relatively small. We conclude that Man- itoba students indicated a desire for more distance from non-Europeans than Europeans, but there was little evidence for negative attitudes towards them. If there is prejudice, the Bogar- dus scale is not sufficiently refined to measure such prejudice. More refined scales are needed to measure prejudice.

The distinction that the respondents made between willingness to marry and willingness to be a close friend is noteworthy. These appear to

be distinctions of nearness, rather than farness, and this could well be related to variations of identification more than distance, prejudice, or discrimination (Driedger, 1975a). It may be that many high ethnic identifiers simply perceive of marriage into an outgroup as the beginning of assimilation, therefore, they do not wish to do so. Although only about one-fourth are willing to marry into non-European groups, about one-half are quite willing to form close friendships with non-Europeans. This nearness distinction would suggest an identification differentiation, not prejudice towards outgroups.

Considerable differentiations are made within the European origin category. Four out of five were willing to marry into the majority British group. Only half as many were willing to marry a Jewish person. Does this indicate prejudice? Glock and Stark (1966) have shown that there is considerable prejudice against Jews in the United States. It would appear that greater student distance from Jews is evidence of potential prejudice in this sample as well, but the Bogardus scale does not measure the multidimensional elements of prejudice so we cannot answer that question. The differentiations among the nine non-European outgroups are small. It is interesting to note that a few more students were willing to marry blacks than Es- kimos, but the difference was not great.

C O N C L U S I O N S At first we wondered whether the results of this study warranted publication. In the process of the study of ethnic identity and social distance we encountered so many problems of conceptualization and methodology that we wanted many times to begin again with a new sample, with new data, with better scales, and with more intensive theoretical conceptualization in order to report more convincing results. During our study, however, we became aware of new dimensions in the problem. Since many ofthese require more study, we think they are worth reporting. Indeed, we feel that some of our in- sights and findings should be pursued.

Simmel’s concept ‘The Stranger’ is a fruit- ful one which led many early American sociologists, who had studied under him and read his works, to pursue the dimensions of nearness and farness, identity and distance. Many of us have strayed far from Simmel’s original ideas and have been much too satisfied with second-hand tidbits without reading the theoretical context out of which his concept of

TABL

E V

TH

E D

EG

RE

E O

F SO

CIA

L D

IST

AN

CE

TH

AT

UNIVERSITY

STU

DE

NT

S W

OU

LD

PR

EFE

R T

O M

AIN

TA

IN B

ET

WE

EN

TH

EMSE

LVES

A

ND

TW

EN

TY

GR

OU

PS O

F E

UR

OPE

AN

AN

D N

ON

-EU

RO

PEA

N O

RIG

IN

Soci

al D

ista

nce

Scal

e (B

ogar

dus)

~

~ ~

2 3

4 5

6 7

I W

illin

gnes

s W

illin

gnes

s W

illin

gnes

s W

illin

gnes

s W

illin

gnes

s W

ould

Ethn

ic g

roup

s to

mar

ry

clos

e fri

end

neig

hbou

r w

ith o

n jo

b ac

quai

ntan

ce

visi

tor

only

na

tion

deba

r fro

m

Will

ingn

ess

to h

ave

as

to h

ave

as

to w

ork

to h

ave

as

to h

ue as

Euro

pean

orig

in

%

Brit

ish

79.7

14

.1

2.8

3.1

1.2

0.8

0.3

Am

eric

an

77.5

12

.9

2.8

2.1

2.8

0.6

1.2

Scan

dina

vian

66

.3

23.3

5.

7 2.

5 1.

6 0.

7 0

.0

Dut

ch

65.7

22

.3

7.0

2.3

1.6

0.8

0.3

Ger

man

63

.9

22.9

6.

7 2.

3 3.

1 0

.4

0.6

Fren

ch

62.1

23

.5

6.8

2.3

2.9

0.9

1.6

Ukr

aini

an

59.8

27

.4

6.4

2.7

2.5

0.9

0.4

Po

lish

54.5

28

.3

9.0

3.4

3.3

1.2

0.4

R

ussi

an

51.0

28

.6

8.0

4.2

4.6

2.

5 1.

5 It

alia

n 47

.8

32.6

7.

9 4

.0

4.9

1.8

0.9

Jew

ish

42.2

35

.6

10.0

3.

9 5.

4 1.

9 0.

9 N

on-E

urop

ean o

rigin

M

exic

an

28.4

44

.5

13.2

6.

0 6.

4 1.

2 0

.4

Japa

nese

26

.9

49.6

13

.6

3.5

4.4

1.7

0.3

Wes

t Ind

ian

26.6

47

.6

12.8

5.

8 5.

5 1.

4 0.

3 Fi

lipin

o 25

.6

46.7

14

.1

6.5

5.0

2.

0 0.

3 C

anad

ian

Indi

an

24.3

46

.9

11.6

7.

8 7.

2 0.

9 1.

3 Ea

st I

ndia

n 23

.4

48.0

14

.8

5.6

5.2

2.5

0.4

C

hine

se

22.9

50.9

13

.3

3.8

5.8

2.3

0.8

Eskimo

20.3

52

.4

13.8

5.

9 5.

8 2.

1 0

.3

Neg

ro

27.4

53

.4

12.2

4

.0

2.1

0.8

0.1


‘The Stranger’ grew. His approach to social phenomena as a combination of opposed tendencies is well captured in ‘The Stranger’ concept. The stranger presumably has a reference group which sustains him, yet he relates to outsiders at the same time. How and to what extent does he relate to this ingroup? What happens to the ingroup ties of ‘The Stranger’ when he relates to the members of outgroups? How in- tensely can he relate to both groups? Must ingroup loyalty suffer with increased outgroup contact, orcan loyalties to bothgroups be maintained at the same time? Can individuals as Barth (10) suggest cross boundaries of in- groups and outgroups without assimilation or prejudice?

In sorting out the nearness dimension as ingroup identification, one discovers that it is a multidimensional concept with many structural, social, cultural, and social psychological elements which need to be considered. In this study we used a fragment of this multidimensional concept. We found that Manitoba university students who scored low on ethnic identification also scored low on social distance; as ingroup identity increased social distance from outgroups also increased.

We also wanted to see to what extent ingroup differentiation such as socioeconomic status might tend to diminish social distance from others, even among those of high ingroup identity. We found that about as many low status students of high ingroup identification favoured social distance as the total sample. This was somewhat evident among the high status students, but the association remained as predicted. High identification with an ingroup was associated with increased social distance from outgroups when controlled for socioeconomic status.

The association between high ingroup identification and high distance from outgroups also remains for the British and Scandinavians, when controlled for ethnic origin, but not for the French. British and Scandinavian students indicated very low ethnic identification with their ingroup and very low social distance from outgroups, with very little discrepancy between the two; while French students indicated high ingroup identification and low social distance with the largest identity-distance discrepancy. The other four ethnic groups (German, Polish, Ukrainian, Jewish) fell in between. Ethnic origin was an important intervening factor to con- sider.

Those who are responsible for planning a Canadian ethnic mosaic may be alarmed that ethnic identity is associated with social distance. It certainly would be difficult to build a nation where groups perpetuate negative distance and wish to debar others or wish to have others only as visitors. The social distance differentiations, however, tend to cluster at the social nearness end of the scale. Ethnic identifiers seem relatively willing to have individuals from a variety of groups as close friends, or as next door neighbors, or as partners at work, but high ethnic identifiers especially are not willing to marry outsiders. We would propose that willingness to interact extensively is sufficient to build national identity, even though many may be more selective in their marriage choices. Selective identification within religious and political groups is relatively accept- able, so to add ethnic groups should not be an insurmountable problem in the development of Canadian identity. There was little evidence that students were prejudiced, or wished to dis- criminate against outgroups although this needs to be measured more precisely.

We conclude that there is a positive relationship between high ingroup identity and high outgroup social distance as Simmel implied. Other factors, such as socioeconomic status and ethnic origin, may slightly modify the general finding, but the influence of ethnic identity remains the strongest. In addition to the findings in this study, we

found many interesting theoretical and methodological problems which need more intensive study in the future. The concept of social nearness, or ingroup identification, is multidimensional and it requires the selection of numerous ecological, structural, cultural, social, and psychological elements. We used only a cultural ethnic dimension, and in the process of selecting elements such as endogamy and choice offriends found that problems of overlap between identity and distance scales resulted. It would seem that farness is meaningful only in relation to a definition of nearness, but how can independent and dependent dimensions be designed?

Our sample of university students indicated very little farness on the Bogardus scale. Sam- ples of adults in Winnipeg might indicate greater distance so that the full range of the scale could be used, but this is doubtful. On the other hand, the Bogardus scale was designed to measure racial differences, but it does not seem to dis-


criminate between the finer gradations of distance which are most likely present in a university student sample. A more perceptive distance scale is needed to measure more subtle Canadian distance phenomena.

More and more we are convinced that distinctions must be made between positive social distance grounded in the desire for the maintenance of ingroup identification and a multicul- tural society, and negative social distance based on prejudice and discrimination. Although some scholars have used the Bogardus social distance scores as evidence of prejudice, we do not think it sorts out positive and negative dimensions of distance sufficiently. Certainly there must be some prejudice among university students in Manitoba, but all that the Bogardus scale tells us is that some wish to marry into non-European groups, and others wish only to be their close friends. More measures and scales of ethnic prejudice in Canada must be developed in order to probe this subtle phenomenon.

R E F E R E N C E S Ames, Richard G., and Arline F. Sakuma 1969 ‘Criteria for evaluating others: a re-

examination of the Bogardus social distance scale.’ Sociology and Social Re- search 54:5-24

Banton, Michael 1960 ‘Social distance: a new appreciation.’

Barth, Fredrick 1969 Ethnic Groups and Boundaries. Boston:

Bealer, Robert C., Fern K. Willits, and Gerald W. Bender 1963 ‘Religious exogamy: a study of social dis-

Sociological Review 8: 1 6 e 3

Little, Brown, and Co

tance.’ Sociology and Social Research 4859-79

Best, Wallace H., and Charles Sohner 1956 ‘Social distance methodology in the mea-

surement ofpolitical attitudes.’ Sociology and Social Research 40: 266-70

Blishen, Bernard 1967 ‘A socio-economic index for occupations

in Canada.’ Canadian Review of Sociol- ogy and Anthropology 4:41-53

Bogardus , Emory S . 1925 ‘Measuring social distance.’ Journal of

1959 Social Distance. Los Angeles: Antioch

1967 A Forty Year Racial Distance Study. Los

Applied Sociology 9:299-308

Press

Angeles: University of California Press Breton, Raymond 1964 ‘Institutional completeness of ethnic

communities and personal relations to immigrants.’ American Journal of Sociol- ogy 70: 193-205

Brown, Robert L. 1973 ‘Social distance perception as a function

of Mexican-American and other ethnic identity.’ Sociology and Social Research 57r273-87

Derbyshire, R.L., and E.B. Brody 1964 ‘Personal identity and social distance in

Negro college students.’ Sociology and Social Research 48:301-I I

Driedger, Leo 1976 ‘Ethnic self-identity: a comparison of in-

group evaluations.’ Sociometry

1975a ‘Attitudes of students toward immi- 39: 131-41

grants of European and Non-European origin.’ Paper presented to the Confer- ence on Multiculturalism and Third World Immigrants held at the University ofAlberta

1975b ‘In search of cultural identity factors: a comparison of ethnic students.’ Canadian Review of Sociology and Anthropology 12: 150-62

Driedger, Leo, and Jacob Peters 1973 ‘Ethnic identity: a comparison of Menno-

nite and other German students.’ Menno- nite Quarterly Review 47: 225-44

Driedger, Leo, and Glenn Church 1974 ‘Residential segregation and institutional

completeness: a comparison of ethnic minorities.’ Canadian Review of Sociol- ogy and Anthropology I I Z30-52

Ellis, Robert A. 1956 ‘Social status and social distance.’

Sociology and Social Psychology 4It240-6

Fallers, Lloyd A. 1967 Immigrants and Associations (ed.). The

Geismar, Ludwig 1954 ‘A scale for the measurement of ethnic

Hague: Mouton

identification.’ Jewish Social Studies 16: 33-60

Glock, Charles, and Rodney Stark 1966 Christian Beliefs and Anti-Semitism.

New York: Harper and Row Hirabayashi, Gordon K. 1963 ‘Social distance and modernizing Metis.’

Pp. 355-74 in B.Y. Card, G.K.


Hirabayashi, andC.L. French (eds.),The Metis in Alberta Society. Edmonton: University of Alberta

Hunt, Chester L., and Luis Lacar 1973 ‘Social distance and American policy in

the Phillipines.’ Sociology and Social Re- search 57:495-509

Isajiw, Wsevolod 1974 ‘Definitions of ethnicity.’ Ethnicity

I:III-24 Joy, Richard J. 1972 Languages in Conflict. Toronto: McClel-

Kadushin, Charles 1962 ‘Social distance between client and pro-

land and Stewart

fessional.’ American Journal of Sociology 67: 5 17-3 I

Kinlock, Graham C. 1973 ‘Race, socio-economic status, and social

distance in Hawaii.’ Sociology and Social Research 57: 156-67

Kitano, Harry H.L. I 969 Japanese Americans. Englewood Cliffs,

Krech, David, Richard S. Crutchfield, and Egerton L. Ballachey 1962 Individual in Society: A Textbook of So-

N.J.: Prentice-Hall, Inc

cial Psychology. New York: McGraw- Hill

Landis, Judson R., Darryl Datwyler, and Dean S. Born 1966 ‘Race and Social Class as Determinants of

Social Distance.’ Sociology and Social Research 51:78-86

Laumann, E.O. 1965 ‘Subjective social distance and urban oc-

cupational stratification.’ American Journal of Sociology 71 :26-36

Lazerwitz, Bernard 1953 ‘Some factors in Jewish identification.’

Levine, Donald N., Ellwood B. Carter, and Eleanor Miller Gorman 1976 ‘Simmel’s influence on American sociol-

Jewish Social Studies 15:24

ogy.’ American Journal of Sociology 8 I : 81 3-45

Levinson, B.M. 1962 ‘Yeshiva College sub-cultural scale: an

experimental attempt to devise a scale of the internalization of Jewish traditional values.’ Journal of Genetic Psychology 10 1 : 37599

Lewin, Kurt 1948 Resolving Social Conflicts. New York:

Harper

Lieberson, Kurt 1970 Language and Ethnic Relations in Cana-

Mackie, Marlene 1974 ‘Ethnic stereotypes and prejudice: Al-

berta Indians, Hutterites and Ukrain- ians.’ Canadian Ethnic Studies 6:39-52

da. New York: Wiley

Martin, J.W. 1963 ‘Social distance and social stratification.’

Sociology and Social Research 47: 179-86 Mitchell, Roy 1955 ‘An ethnic distance study in Buffalo.’

Nahirny, Vladimir, and Joshua A. Fishman 1965 ‘American immigrant groups: ethnic iden-

tification and the problem ofgenerations.’ Sociological Review 13:311-26

Nicholson, T.G., and Maurice H. Yeates 1969 ‘The ecological and spatial structure of

Sociology and Social Research 41:35-40

the socio-economic characteristics of Winnipeg, 1961.’ Canadian Review of Sociology and Anthropology 6: 162-78

Park, Robert E. 1923 ‘The concept of social distance.’ Journal

Photiadis, John D., and Jeanne Biggar 1967 ‘Religiosity, education, and ethnic dis-

of Applied Sociology 9: 339-44

tance.’ Journal of Applied Sociology 67:666-72

Porter, John 1965 The Vertical Mosaic. Toronto: Univer-

Rinder, Irwin D. 1959 ‘Polarities in Jewish identification: the

personality of ideological extremities.’ Pp. 493-502 in Sklare (ed.), The Jews: Social Patterns of an American Group. New York: Free Press

sity of Toronto Press

Royal Commission on Bilingualism and Bicul- turalism 1970 The Cultural Contribution of the Other

Ethnic Groups. Volume I V . Ottawa: Queen’s Printer

Schmitt, Madeline H. 1972 ‘Near and far: a re-formulation of the so-

cial distance concept.’ Sociology and So- cial Research 57:85+7

Segalman, Ralph 1967 ‘Jewishidentity scales: A Report.’ Jewish

Simmel, Georg 1950 The Sociology of Georg Simmel. Kurt

Wolff (ed.). Glencoe, Illinois: Free Press 1955 Conflict. Kurt Wolff (trans.), Glencoe, I1-

linois: Free Press

Social Studies 29: 92-1 I I


Sinha, A.K.P., and O.P. Upadhyaya 1962 ‘Eleven ethnic groups on a social distance

scale.’ Journal of Social Psychology 5T49-54

Sklare, Marshall, Mark Vosk, and Mark Zym- browski 1959 ‘Forms and expressions of Jewish iden-

tification.’ Jewish Social Studies 17: 203-1 7

Smith, M.W. 1970 ‘Measuring ethnocentrism in Hilo,

Hawaii: a social distance scale.’ Sociol- ogy and Social Research 54: 220-36

Sumner, William Graham 1906 Folkways. New York: Dover Triandis, Harry C., and L.M. Triandis 1962 ‘A cross-cultural study of social dis-

tance.’ Psychological Monographs: Gen- eral and Applied 76: 1-2 I

Turbeville, Gus 1950 ‘A social distance of Duluth, Minnesota.’

Sociology and Social Research 34:415-23

Uyeki, Eugene S. 1960 ‘Correlates of ethnic identification.’

Vallee, Frank G. 1969 ‘Regionalism and ethnicity: the French

case. Pp. 19-25 in B.Y. Card (ed.), Perspectives on Regions and Re- gionalism. Edmonton; Western Associa- tion of Sociology and Anthropology.

Westie, Frank R. 1959 ‘Social distance scales: a tool for the

American Journal of Sociology 65468-74

study ofstratification.’ Sociology and So- cial Research 43:251-8

Zaida, S.M.H. 1967 ‘A study of social distance as perceived

by students of Karachi University.’ Jour- nal of Social Psychology 71: 197-207

Documents

Identity and social distance: towards understanding Simmel's ‘The Stranger’