Categorization

Classical View

The first theories of categorization can be traced to Plato and Aristotle. In their view, categories had clear boundaries defined by common properties and were uniform in respect to centrality. This meant that all members of a particular category were equal and no member had any special status. All category members are united only by shared attributes and are distinguished from other categories when they do not share the same attributes. Although this view of categorization is not really conceived of using empirical research on thought or interaction, it is more or less sufficient for speakers in day-to-day interaction. This model, however, does not work as well when categories are challenged and the process of categorization is explicitly at issue. Categories, particularly in social organization and interaction, are not always static things and members of particular categories are not always equal.

In the 20th century, this classical view of categorization was challenged, beginning with the work of German philosopher Ludwig Wittgenstein. Wittgenstein suggested that categorization is based not only on common attributes, but also on family resemblance between members of a category. Wittgenstein used the example of the category “games,” showing that there is no common attribute between all games. They are instead like family members, similar to one another in a wide variety of ways. Wittgenstein also suggested that categories can have central and noncentral members, and that there are good and bad examples of a category (i.e., members that are more typical of a category than others). Wittgenstein’s work, however, did not focus exclusively on social categorization and was not based on empirical evidence of categorization in talk or cognition.

Prototype View

Key research in the field of cognitive science challenged the classical understandings of categorization with empirical data. Eleanor Rosch’s research was particularly important in developing a cognitive model of categorization, showing how categorization works not only in an abstract way but also through doing empirical research into the processes humans use when they categorize things. Rosch did not see categories as fixed boundaries between objects or ideas, but proposed that there was a lot of “fuzziness” around the process of categorization. Categories are not so much rigid and fixed, but have edges that are hard to define. Instead, Rosch suggested that within categories prototypical members—or particularly good examples—can be found.

Rosch’s prototype theory of categorization took Wittgenstein’s notion of central and noncentral category members ever further. Rosch worked in Papua New Guinea with the Dani tribe to show tribe members were able to distinguish between colors, even though they do not have words for colors beyond black and white. Her research showed that people did not think about categories as lists of attributes that different potential members did or did not have, but instead people have an idea of a prototypical member of a category. People compare other potential members of the [Page 120]category to this prototype to decide whether or not something is a good fit for a category. Rosch suggested that cultures and individuals discover correlations and build categories based on the correlations. If someone says “fruit,” for example, one might think immediately of an apple rather than a tomato. This suggests that something about an apple makes it a better prototype of “fruit” than tomato. That does not mean, however, the prototype of the category of “fruit” is the same in every culture or that the category of “fruit” would be found in every culture. In tropical climates, where there are no naturally occurring apples, people might think of other fruits as prototypical of the category. The correlations and categories come from people’s lived experience and therefore differ based on the complex interaction of these experiences.

...