Kristina Boréus


The aims of this paper are (a) to show how the 'depth of categorization' may be studied in a corpus when the categories in question are categories of people; (b) to report the results of such a study within the framework of a larger research project; and (c) to discuss the importance of the question of depth of categorization.


Racism, and presumably the exclusion of other groups which challenge a certain view of normality, such as lesbians and gays, has according to various indicators, increased in Sweden during the last decades.

The general study, which frames the analysis of categorization, is about how such exclusion and discrimination have displayed themselves in the Swedish public sphere at three points in time. The study uses a model in which such exclusion is studied as three interrelated components: categorization of people as belonging to groups that are 'different'; disdain for such groups; and argumentation to the effect that people who are seen to belong to such groups should be treated in special, negative ways (for instance being forced to undergo sterilization or be deported).

The study is based on an analysis of election propaganda, parliamentary debates, debate in the larger daily newspapers, as well as news reporting. Samples from topical issues in public debate are studied during three different points in time: 1932/'33, 1970/'71, and 1994/'95.

Categorization based on different, real or imagined, facts about people, are considered. Among these are the following grounds for categorization:

- that people commit crimes, thus deviating from the norm of being law-abiding;

- that people are seen as belonging to some ethnic or national group, apart from what is seen to be the Swedish national group;

- that people have what is considered to be physical shortcomings, such as handicaps or chronic diseases, thus deviating from the idea of having a sound body;

- that people are seen to be lacking in mental or intellectual capacities, for instance being classified as mentally retarded or disturbed, thereby deviating from the standard of normality;

- that people have certain phenotypic traits, such as one's color of skin, that are different from what are considered to be the Swedish phenotypic traits;

- that people have sexual inclinations that deviate from the heterosexual norm.

People being categorized according to the above mentioned facts, have, during the last century, in Europe and elsewhere, continuously or at times, been the victims of more or less cruel exclusion and persecution.

Different types of discourse analytical methods are used in the research project. In the part of the study that deals with categorization, the findings of cognitive linguistics are used.

Depth of categorization

Categorization of people is made in all societies at all times, but the resulting categories differ. Some categories are experienced as 'deeper' than others are. Characteristic of deep categories is that they are experienced as natural and that it is felt to be important whether or not an individual belong to a deep category. WOMAN and MAN are deep categories. Examples of less deep categories would be LEFT-HANDED / RIGHT-HANDED PERSON or, even less deep, RIGHT-FOOTED / LEFT-FOOTED PERSON. Categories of people constructed as 'others' can vary in depth, too. Thus, WOMAN, which, it can be argued, is the prototypic category of 'others', is a deep category, while LEFT-HANDED PERSON is a less deep, but (at least until recently) also a category of 'others'. The depth of a certain category is a cultural construct, but not a random one. It is interrelated with other ideas of what is important in and about people.

Apart from the intuition that members of a certain society can have about the naturalness and importance of a categorization, the depth might be revealed by the use of language in one or both of the following two ways:

- The category might or might not be on the basic level. A deeper category is more likely to be a basic-level category than a less deep one. The various criteria used by Rosch (1978) and Lakoff (1987:46) and can be used to judge if the category in question is a basic-level one.

- Lexicalization might show, in different ways, that a category is deep. A person with the medical condition of epilepsy could for instance be called an 'epileptic' or a 'person who suffers from epilepsy'. The choice of the single noun seems to imply a deeper categorization.

Results and problems

As part of the study described above, it was analyzed how categories, based on the various facts mentioned, were formed during the three points in time. For each category studied, I tried to judge if it was a deep or less deep one. An example is the various categories formed on the basis of the fact that people (are thought to) have committed crimes. Here MURDERER and THIEF seemed to be deep categories used at all three points in time. The next question asked was whether there was a difference between the three points in time regarding the frequency of the use of such deep categories in similar contexts.

Some problems involved in making the judgements about the depth of categories, were the following ones:

- Most studies of basic level categories and prototype effects deal with categories of concrete objects, such as animals or furniture. The categories analyzed here are much more abstract ones. Therefore, only some of the criteria of basic-levelness mentioned by Lakoff and others, are applicable. The most important difference seems to be that we use our senses in a much more indirect way when determining whether or not an individual should be included in, for instance, one of the categories based on the fact that people commit crimes, than when we judge whether or not an individual should be included in a category like DOG.

- Not all types of categories come in neat taxonomies, like categories of animal or furniture.

- There is a general problem to do with the fact that a language contains dead stuff, which, like frozen metaphors, is only a reminiscence of ideas held previously in a society. Thus, the word 'epileptic', 'epileptiker', existed in Swedish at all three points in time. Other facts support the judgement that EPILEPTIC was a deep category in the early 1930's. People classified as epileptics were, together with other categories, treated as abnormal and, among other things, denied the right to marry without special permission. People suffering from epilepsy are not thought of or treated that way today, but the term 'epileptiker' is still in use. Is this a sign of the term being a relict of earlier ideas, likely to fall out of use in the future? Or is its usage a sign of the existence of ideas that we would perhaps not like to admit are still alive in the Swedish society?

These problems made it impossible to make a judgement on the depth of categorization for some of the categories that appeared in the corpus. But for others it was possible to make a judgement with the help of the devices discussed above.

The importance of the depth of categorization

The importance of the depth of categorization has to do with the feeling that deep categories are 'natural' and with the fact that people tend to see it as important whether or not an individual belongs to a deep category. Categories of 'others' that are deep might be very difficult to challenge for exactly those reasons. An important question to ask is how the depth of categorization is related to the second component of the model referred to above, disdain. Are the categories that are the victims of disdain mostly deep categories, or is there no such connection? This question is studied empirically in the wider research project, and tentative results from that part of the study are presented in this paper.


Banakar, Reza (1993) "Det offentliga samtalet om etnokulturella frågor" (''The public discourse on ethno-cultural issues'') Häften för kritiska studier, no. 1-2 1993

Bauman, Zygmunt (1993) Modernity and Ambivalence Polity Press

Beauvoir, Simone de (1970) The Second Sex Alfred A. Knopf

van Dijk, Teun A. (1993) Elite Discourse and Racism Sage

Gilroy, Paul (1987) 'There Ain't No Black in the Union Jack'. The cultural politics of race and nation Hutchinson

Katz, Jonathan Ned (1996) The invention of heterosexuality Penguin

Knowles, Caroline (1992) Race, Discourse and Labourism Routledge

Lakoff, George (1987) Women, fire, and dangerous things University of Chicago Press

Rosch, Eleanor (1978) ''Principles of Categorization'' in Eleanor Rosch and Barbara Lloyd, eds., Cognition and Categorization Hillsdale

Ungerer, Friedrich & Hans-Jörg Schmid (1996) An Introduction to Cognitive Linguistics Addison Wesley Longman