33. Prototypes

When we look whether an individual belongs to a category, what do we compare its features with? What ‘represents’ the category? Earlier, I denied that there always is an essence, or necessary and sufficient features that something must have to be assigned to a category. So how does it work? Several proposals have been made for this.

Hilary Putnam proposed a linguistic division of labour, where specialists know the ‘real’ meaning, such as, for water, its chemical composition (H₂O). Ordinary people can refer to the specialists in case of doubt, but for everyday activity they use what Putnam recognized as a stereotype, such as water being clear, potable, boiling at 100 degrees Celsius, freezing at 0 degrees, subject to expansion when freezing (causing water ducts to burst). In fact these features do not always apply (stereotypes often fail): under pressure the boiling point is higher and the freezing point is lower. At greater heights air pressure is lower hence boiling point is lower. One goes beyond the stereotype when conditions require it. However, specialists also can be wrong, and new scientific discoveries can shift ‘real’ meaning, even though that seems unlikely before it happens.

Johnson-Laird used the notion of default to clarify how conventional criteria for meaning could work. In a default, features are assumed until there is contrary evidence. This fits with philosophical pragmatism: we assume something as given until we run into misfits or novel openings. In other words: all ideas are defaults. The stereotype serves as a default. 

Wittgenstein offered the idea of typical cases that form a norm and one handles boundary cases in comparison to the norm. Different cases, individuals of a universal, may not have any universally shared feature but a chain of family resemblance. Proximate members of a family have common features while distant members don’t. X is in the same class as Y not because they have something in common but because there is an intermediate Z that has one feature in common with X and another with Y.

Eleanor Rosch proposed the idea of a prototype, which is a salient exemplar of a class that connects others in the class. Class membership is decided on the basis of resemblance to a salient case, or a typical case, which serves as a prototype. The prototype depends on culture and natural conditions. For example, apparently for the Dutch the prototype of a bird is a sparrow, and for the British it is a robin.

The idea of comparison with a salient case or prototype is an ancient one. It goes back to the old notion of a paradigm, used by Socrates, as an exemplary case to mimic. Aristotle recognized the exemplary cause in his multiple causality: the prototype that a carpenter imitates in constructing a chair.

There is wisdom in this notion. One can see management as imposing strict, universal rules, or as setting an example to be imitated, or a ‘role model’, with some leeway for interpretation according to circumstances.