Go to Semiotic Review of Books Home Page
Go to SRB Highlights
Go to SRB Archives


SRB Archives

This review appeared in Volume 3 (2) of The Semiotic Review of Books.

Xenophobic Discourse

by Bohdan Szuchewycz

Speech and Reasoning in Everyday Life. By Uli Windisch. Translated from French by Ian Patterson. New York: Cambridge University Press, 1990, 224 p. ISBN 0-521-35438-2

The translation and publication of Uli Windisch's Speech and Reasoning in Everyday Life is a welcome addition to the growing body of literature on racist and xenophobic discourse. Originally published in French in 1985, the work represents an independent, though clearly relevant approach to a topic which is a current focus of investigations in critical linguistics (Blommaert & Verschueren 1991; Smitherman Donaldson & van Dijk 1988; Sykes 1985; van Dijk 1984, 1987, 1991, 1992). Although the volume does not completely satisfy some of its stated goals, it does provide a valuable case study of swiss xenophobic discourse and identifies specific features and strategies which constitute that discourse.

The monograph consists of six chapters organized into three parts. Part 1 (Chapters 1 and 2) provides the reader with a general introduction consisting of eleven basic postulates and ten more general comments which summarize and clarify issues discussed in the author's earlier publications on racist discourse (Windisch, Jaeggi and Rham 1978, 1982) and articulate his underlying assumptions concerning the nature of social life, language, and thought, and their interrelationship. The theoretical and methodological foundations of the research are also presented. Windisch's perspective on language is pragmatic (i.e., language as a form of social practice) and centers upon identifying differences in the discursive styles of different social collectivities. From this sociological perspective language in use becomes a key to understanding both social and cognitive practices. Windisch thus utilizes a corpus of written transcripts of in-depth interviews with both the movements. Interviewees were selected from a pool of individuals who had written letters to various daily newspapers expressing their personal views concerning the presence of foreign workers in Switzerland and their response to the political initiative of the Action Nation (AN) to have such workers and their families expelled from the country.

Utilizing specific examples drawn from the interview data. Part 2 presents an analysis of the discourses in terms of the author's variables of cognitive centration (Chapter 3), social causality (Chapter 4) and the perception of time (Chapter 5). As this section constitutes approximately three ` quarters of the text, it will be discussed in detail below.

Borrowing Piaget's notions of ego centration and decentration, Windisch develops an approach to sociocentrism, and through an analysis of sociocentric discourse identifies the sociocognitive mechanisms which are characteristic of sociocentric thought. These include a tendency toward essentialism, normativism, false identification, and one dimensionality (40). Stressing that sociocentrism can never be viewed as a simple retention of the egocentrism of childhood, he defines sociocentrism as "a social actor subject's tendency to favour his own point of view with his own way of knowing. A sociocentric subject will tend to attribute his own categories of knowledge to other subjects' interpretation of social (and cognitive) practices (i.e., there is a predominance of assimilation over accommodation)" (34). This involves the reduction of new features to familiar, unchanging, and frequently stereotypical mental categories. Within such a sociocognitive structure "the social actor subject is more 'acted' (by his own categories of knowledge) than actor. The subject has little autonomy, and the cognitive and linguistic work is not extensive. He cannot distance himself from his own way of knowing reality" (35). Furthermore, affectivity and emotion become important aspects of this primarily subjective frame of reference.

Sociocognitive decentration, on the other hand, is characterized by a social actor subject who "acts as much as he is acted upon", who utilizes "more elaborated linguistic, cognitive and social practices" which in turn leads to "autonomy, (which) enables a degree of distance to be established vis-a-vis the multiple forms of determinism and heteronomy ... and encourages the desire to understand" (37). It is evidenced in discourse in a speaker's desire to relativize and shift perspective, and in strategies of withdrawal and detachment.

The second variable, the social discourses of causality, refers to the various ways by which everyday social reality is explained. Here the reader is presented with a classification of five logico-discursive forms and three paradigms of causal explanation. The logico-discursive forms derived through the analysis of the interview data are presented as ideal models or strategies of explanation which reveal the subjects' differing patterns of social thought in use. These include: 1) segmented/associative causality (in which cause and consequence are implicitly linked by simple sequential adjacency), 2) circular causality (in which causal relationships are supported by a contrario demonstration), 3) contingent causality (in which "temporal proximity" provides the basis for the attribution of causality), 4) causal supersaturation (in which a wide range of phenomena are attributed to a singular or very limited number of causes), and finally, 5) multiple causality (in which a single phenomenon is attributed equally to multiple causes).

These various discursive forms are not however independent of content or of the speaker's investment in particular material or ideological issues. Content is thus characterized in terms of three major paradigms of causation: deviancy, materialism, and indeterminacy. Subjects utilizing the deviancy paradigm attribute causation to the voluntary behaviour of different individuals or social groups (i.e. voluntarism). This behaviour is then judged to be normal/acceptable or deviant/unacceptable with respect to the norms of the subject's reference group. Thus the "problem of foreign workers" may be attributed to their own behaviour and/or to the actions of government officials and their policies regarding those workers and their families. In contrast, the materialist paradigm identifies material factors as giving rise to particular social facts. Here causation is attributed to a complex of social, economic and political factors such as unemployment in the home country, the capitalist desire for workers (frequently lower-waged workers) in Switzerland, the lack of support given by employers to the workers, etc. Its primary characteristic is that causation is attributed through an explanation of the workings of the social system. Finally, the indeterminacy paradigm is characterized in opposition to the other two. It differs from the deviancy paradigm in that it demonstrates an absence of social categorization; it differs from the materialist paradigm in that no account is taken of the relationship between individuals and the structure and functioning of the larger social system. Non-determination thus relies on the "naturalization of the social" (118), wherein social actors have little or no real influence over a social reality governed by natural social laws, and where the existing social hierarchy is regarded as a priori legitimate.

The final chapter of Part 2 discusses the perception of time and its representation in the interview data. A distinction between the discourse of mythic-cyclical time vs. linear time provides the basis for distinguishing the interview texts. Subjects who made reference to a lost idealized past and viewed changed negatively in terms of the devaluation and deterioration of tradition were characterized as representing a mythic-cyclical conception of time. Solutions to current problems are seen in the preservation and restoration of the patterns and values of earlier times. In contrast, those subjects who viewed change as inevitable, irreversible, progressive, and positive represent a linear model of time.

A major difference exists with respect to the status of foreigners within these two frameworks. Windisch argues that within the mythic-cyclical view, foreigners become the focus and serve as scapegoats for the wide range of dissatisfactions experienced by those in Swiss society who long for a return to "traditional" ways and values. A majority of these subjects expressed support for the AN initiative. Subjects who exhibited a linear perception of time, where the past is not mythicised, viewed the presence of foreign workers as but one of many social and material determinants, both national and international, contributing to the economic, social and political situation in Switzerland. All of these subjects opposed the AN initiative.

In Part 3, the final chapter, Windisch integrates the variables of centration, causality, and perception and discusses their articulation. In contrast to the approach taken in Part 2 where brief examples were drawn from many different interviews and individuals, this section attempts to examine the interaction and interrelationship of these factors in the discourse of specific speakers. Some effort is thus made to present a picture of the different rhetorical or discourse strategies employed within a complete text. The goal of the study was to discover "the way in which different social groups behave at the same time in connection with centration, causality and time ...(their) overall lingusitic and cognitive functioning , and its social variations" (his emphasis (183). This is accomplished through the construction of two ideal types or fundamental sociocognitive configurations based on the articulation of centration, causality, and the perception of time. Those subjects who held a favourable view of the proposed initiative to expel foreign workers typically produced a discourse which exhibited strong social centration, relied on causal supersaturation and the deviancy paradigm as the strategy of explanation, and made reference to an idealized and culturally and politically isolated Switzerland in the nonspecific (i.e. mythic) past. Within this configuration there was also a strong identification of fixed social norms as arising out of the past, the naturalization of the social, and voluntarism as the locus of causal explanation. For those who were generally opposed to the initiative sociocognitive decentration was the norm, with the materialist paradigm generalized and applied widely regardless of subject. A progressive linear perception of time with a positive attitude towards social change was also characteristic of the discursive style of this group.

Windisch consistently stresses the empirical basis of the research and the necessity of providing empirical controls over his theoretical and conceptual propositions. It is upon this aspect that I wish to concentrate my comments for, in spite of this emphasis, the reader is provided with little justification and only minimal information about the data used and the circumstances of its collection. Furthermore, the information which is provided is awkward to retrieve as it is scattered throughout the text in several chapters and in footnotes.

On the first account, Windisch at one point stresses the need to employ data which are as "natural" and as "raw" as possible - that is, data produced in real everyday situations. Although he suggests that "situation where individuals are asked to produce data for the researcher need to be avoided" (6), the entire corpus upon which the analysis is based consists of texts produced in an interview context. Informal interviews have proven very valuable in similar discourse focused research (see van Dijk 1987 for a discussion of their value); however, in the present context some justification should have been made for the methodological assumption that interviews produce a style of discourse which is equivalent to everyday, natural speech with respect to the sociocognitive structures it reveals.

Neither is information provided on the actual interviews themselves; and this too is unexpected given the author's stress on the situation and pragmatic nature of social discourse. Interviews are a type of communicative event; like other social events, they operate with their own sociolinguistic constraints, and these must be taken into account when such data provides the basis for social scientific generalization (Briggs 1986). This awareness is especially important with respect to the role of the interviewer in generating the discourse which is eventually to be subjected to analysis. In many instances in the present work only the responses are included; the actual question which elicited the response is not reproduced. In instances where the interviewer's question is provided it is possible to note the direct influence of the interviewer on the subsequent line of reasoning taken by the subject. Note the following example with respect to voluntarism:

Q: Do you think the foreigners who come to Switzerland could find work in their own countries if they wanted to?

A: I don't know if they would want to. Judging by the people in our block, I'm certain that none of them would want to go back. (my emphasis) (131).

In other words it is quite possible that the researchers may themselves have contributed to the construction of the sociocognitive structures revealed in the analysis and attributed to the subjects.

Thirdly, only vague information is provided on the method of subject selection. In-depth interviews ("more than one") were conducted with "some 50 of the most representative letter writers" (15). The basis of this judgement of representativeness on the part of the researchers is not specified. One assumes that representativeness was determined by the textual features of the letters, as opposed for example to a sample of writers who were representative of differing social categories or groups. Given that one of the concerns of the research was to document "the way in which patterns and categories of thought - ways of knowing - can vary substantially within our own society according to social group" (his emphasis) (4) the absence of an indication of the subjects' gender, age, education, or class is surprising. It is briefly noted, for example, that individuals who have a mythic-cyclical representation of time tend to be older and to come from less privileged backgrounds (178). The author clearly recognizes such potential correlations at various points in the text; however, the absence of such fundamental sociological information makes it impossible to explore the potentially significant social distribution of the differing sociocognitive structures identified in the interview texts. Instead, social analysis is confined to examining the discourse of "the people who make up the base of the population" (7) (defined, in a footnote, as the working and middle classes). These discourses are then subcharacterized as "the texts of interviews with supporters of the xenophobic movements, and the texts of interviews with their opponents" (21) in order that the analysis might be comparative. Again the reader might wonder about the criteria underlying such a decision. Were the subjects asked if they were supporters or not, or was the assignment based upon discourse features of the interviews themselves?

Windisch notes "a tendency in political and social discourse analysis to look particularly at the discourse of people in authority, leaders or specialists, rather than at what ... the people who make up the base of the population have to say" (7). To rectify this imbalance he chooses to look exclusively at the discursive behaviour of the base. While an examination of these everyday discursive practices is extremely valuable, one must not neglect the important interrelation ships between these forms and the discourses of the state and its agents ('experts', 'specialists', and particularly the media). Indeed, van Dijk points out the strong relationship between elite and popular forms of racism and argues that "many of the more subtle, 'modern', 'everyday', or 'new' forms of cultural racism, or ethnicism ... are taken from elite discourse" (1992:88). The specific topic which formed the frame of the interviews concerned a specific political initiative, put forward by a specific political group, which voters were invited to endorse or reject. The AN initiative would certainly have received extensive media coverage and been the subject of much public (and private) political debate. It is thus naive to assume that the everyday discourse of the base of the population was independent and unaffected by the surrounding discourses. This leads to a further point. Although the Action Nationale initiative was central to the methodology (with respect to the data collected and analyzed and to the goals of the research) very little information concerning the initiative was provided. North American readers in particular who would most probably be unfamiliar with the internal politics of Switzerland would have benefited from a specification of the precise content of the AN initiative and its legal and social implications. Greater discussion of the background and the social debate surrounding the proposal would have been very helpful.

One of the goals of the research was to account for the relative effectiveness of some political rhetoric as opposed to others. While we are presented with an analysis of sociocognitive structures as revealed in the discourse of subjects, and thus with some sense of the audiences of political discourse, in order to more fully address this question as to the effectiveness of particular discourses, some attention should have been paid to actual political texts and the discursive and sociocognitive structures which they reveal. The assumption is that effective political discourse is successful because it succeeds in utilizing a sociocognitive orientation which matches that of its audience. While this seems a reasonable hypothesis, it remains to be demonstrated that this is indeed the actual case, i.e. that these parallel structures do in fact exist. Secondly, an analysis of subjects' reactions to specific political discourses would help to establish more clearly the role of shared sociocognitive representations. That is, does the sharing of sociocognitive outlook necessarily guarantee that a particular instance of political rhetoric will be positively received? While this may indeed be an important aspect, it is quite possible that other social, cognitive and linguistic factors may be involved.

Finally, I think that one must exert some caution in assuming that linguistic expression is a direct reflection of cognitive and logical processes, an assumption which Windisch does not question when he states, for example, that "we want to look simultaneously at what the man in the street is thinking and how he says it ... How does he think and say what he thinks and says?" (his emphasis) (6). The author cites Labov as an example of a sociolinguist who demonstrated that members of different speech communities have different cognitive and linguistic competencies. However, in his now classic critique of the deficit hypothesis for educational failure, Labov (1972) cogently and persuasively demonstrated the dangers of drawing conclusions about cognitive skills from verbal behaviour, and particularly verbal behaviour produced in interview contexts. Indeed, his presentation of "the logic of non-standard English" did much to sensitize cognitive oriented researchers to the independence of the standard features of the code from a speaker's ability to construct logical arguments.

Like the work of Windisch, Van Dijk's (1987) study of communication and prejudice also focused on socially shared opinions and attitudes in terms of shared cognitive representations, stressing that "prejudice is not just an individual attitude of (bigoted) people, but a structurally founded form of social cognition" (ibid. 391). Van Dijk's research also argued that the cognitive strategies and structures of racist discourse are remarkably similar in a number of different countries in Western Europe and North America. Windisch's study of Swiss xenophobia is a valuable addition to this comparative enterprise in that his findings both confirm and supplement the observations of van Dijk. Indeed, it is the identification of centration, causality and the perception of time, and their fundamental role in xenophobic talk, that constitutes the greatest value of the present work.

Van Dijk and other researchers who adopt a critical perspective are clear on what they see as the ultimate social functions of such discourse; ethnic prejudice serves "in the maintenance of power and privileges of the dominant White majorities in these countries" (1987:392) and as such is "constitutive of the social and political dimensions of structural racism in society" (Smitherman-Donaldson & van Dijk 1988:13). Windisch unfortunately remains silent on this matter.


Blommaert, J., & Verschueren, J. (1991) "The pragmatics of minority politics in Belgium". Language in Society 20 (4):503-532.

Briggs, C. (1986) Learning How to Ask. Cambridge: Cambridge University Press.

Labov, W. (1972) "The logic of nonstandard English". In Language and Social Context. Giglioli, P.P. (ed.) Harmondsworth: Penguin: 179-215.

Smitherman-Donaldson, G., & van Dijk, T.A. (Eds.). (1988) Discourse and Discrimination. Detroit: Wayne State University Press.

Sykes, M. (1985) "Discrimination in discourse". In Handbook of Discourse Analysis Vol.4: Discourse Analysis in Society. van Dijk, T.A. (ed.). London: Academic Press: 83-102.

van Dijk, T.A. (1984) Prejudice in Discourse. Amsterdam: Benjamins.

---. (1987) Communicating Racism. Newbury Park: Sage.

---, T.A. (1988) News as Discourse. Hillsdale, NJ: Erlbaum.

---, T.A. (1991) Racism and the Press. London: Routledge.

---, T.A. (1992) "Discourse and the denial of racism". Discourse and Society 3 (1):87-118.

Windisch, U., Jaeggi, J., & Rham, G. de. (1978) Xenophobie? Logique de la pensee populaire. Lausanne: Editions L'Age d'Homme.

Windischi, U., Jaeggi, J., & Rham, G. de. 1982) Pensée sociale, langage en usgage et logiques autres. Lausanne: Editions l'Age d'Homme.

Bohdan Szuchewycz is an Assistant Professor in the Communications Studies Program at Brock University. His scholarly activity and research are focused in the area of the ethnography of communication, and on issues of language and power. Publications include articles on religious discourse in Ireland, and the literature of the New Right.

Go to Semiotic Review of Books Home Page
Go to SRB Highlights
Go to SRB Archives