In cognitive Sciences, the Théorie of the prototype is a model of gradual Catégorisation, in which certain members of the category are regarded as more representative than others.

* For example, when one asks to provide an example of the concept of “piece of furniture”, the term chair is more frequently quoted than, say, stool .

Definition

The prototype term of was proposed by Eleanor Rosch in 1973 in its study entitled Natural Categories ( Natural Catégories ). It was defined initially like a stimulus, which takes a projecting position in the formation of a category because it is the first stimulus which one will associate with this category. It then redefined it like the the most central member of a category, functioning like a cognitive point of reference .

Note:: This term does not have here the same direction as in an industrial context, where it means the first specimen built , before the mass production.

It should be noticed that the prototype of a category constitutes a subcategory, and not a single authority. Thus the prototype of cat could be male ordinary cat for example, but in no case the individual cat Felix .

Prototype vs CNS

Such as she was formulated in the years 1970 by Eleanor Rosch in particular, the theory of the prototype constituted a radical separation of with CNS (Requirements and Sufficient) of the Logique aristotelician, which led to approaches ensemblists of the intensionnelle Sémantique . Instead of a model définitionnel (for example a bird can be defined by the features, and to fly), the theory of the prototype regards a category “bird” as based on various attributes having an unequal statute: for example a Rouge-gorge would be a better prototype of bird than, say, a Pingouin. This led to a gradual design of the categories, which is a central concept in many models of the cognitive Sciences and cognitive Sémantique, such as for example in work of George Lakoff ( Women, fire and dangerous things , 1987) or of Ronald Langacker ( Cognitive Grammar , vol.1/2,1987/1991).

In traditional lexicology, the lexical direction relèv of linguistics whereas encyclopedic knowledge arose with pragmatic the .

The prototypic approach in particular makes it possible to consider features such as white plumage to define a swan (even if there exist black swans, this feature is regarded as generally relevant ), or invariable to define an adverb (whereas in the expression: of all little girls , the word all , although adverb, is bent; the prototype of the French adverb would be preferably a candidate ending in - lies ). One speaks about typical properties , in opposition to the requirements .

The principal comparative data between the two theories are gathered in the table, made up below starting from the work of Georges Kleiber. It should be noted that the theory of the prototype evolved/moved, of the fact in particular of the pioneers them-even: to see further standard Version and wide Version .

Tests of cognitive categorization

The concept of prototype emerges in particular from investigations near members of the same speech community. It is thus more socio- psychological and statistics that related to the expertise. Kleiber Here an extract of the resulting classification:

If one can have an opinion different on this list for reasons from cultural identities, the important point is that such a gradual categorization has good lucks exist in all the cultures. An additional proof that certain elements of a category are privileged compared to others was brought by experiments holding account of the following factors:

1. Response time. The questions implying of the prototypic members (for example a robin is a bird ) caused answers faster than for the nonprototypic members.

2. Starting. When they were packaged by the category of higher level (superdordonné), the subjects could identify two identical words more quickly. Thus, when one introduced initially the category of the “pieces of furniture”, they more quickly detected equivalence between two occurrences of the word “chair” or “cooker”.

3. Exemplarity. When it was asked to name some representative examples, the prototypic objects were more frequently advanced.

Scopes of application

The theory of the prototype interests in the first chief the lexicographers and lexicologists, on the level of the Définition; it has also implications in Artificial intelligence.

Traditionally, the privileged field of experimentation of categorizations is that of the natural species (animals, plants…). Contrary to categorization aristotelician, the theory of the prototype does not present however opposition between the natural categories (dog, bird…) and the Artefact S (toys, vehicles…)

Following work of Rosch, the prototypic effects were studied largely in fields such as the Cognition of the Couleur S (Brent Berlin and Paul Kay, 1969), and also for more abstract concepts. One subjects, for example, “up to what point this account is can ask a case of lie”. Similar work was undertaken on the actions (verbs like: to look at, kill, speak, go ), the adjectives such as large , the prepositions, the conclusive ones. Academic cases usually treated by the CNS were also studied under the angle of the prototype, like the terms bachelor or mother .

The concept of prototype was finally introduced into all the aspects of linguistics: phonology, morphology, syntax, discursive grammar and textual linguistics, which summarizes as follows: that in Nunggubuyu (language aboriginal of Australia), the term meaning bird includes the bats and the Sauterelle S. On the other hand, the concept of frequency (of use) seems closely related to that of basic category, but is difficult to define exactly.

Other problems emerge when one applies the concept of prototype to lexical categories others that of the name. The verbs, for example, seem to constitute a challenge with a clear application of a prototype: it is difficult to distinguish in as more or less central members.

Resemblance of family

The “family of the plays” (Wittgenstein)

The prototype concept of is traditionally associated with the distrust expressed by Wittgenstein vis-a-vis the traditional concept of category (even if this one did not use the prototype term ). This influential theory led to a prospect for the semantic components as possible contributors rather than necessary to the significance of the texts. Its discussion of the category of the plays is particularly perspicacious and constitutes an obligatory canonical reference (philosophical Investigations 66,1953):

“Let us consider for example the activities that we call “plays”. I the plays of table, the card decks, the plays of ball, the Olympic Games, etc Which do they have hear all in common? Known as steps “they must in common have something, without what one would not call them plays”, but observe and seek if there exists something which is common to them to all. Because if you study them will not be a common point to them, but similarities, relations, and that in great number. I repeat it: do not think, observe! Look at for example the plays of table, with their various relations. Pass now to the card decks: you will find there many correspondences with the first group, but much of common points are erased, while others appear. If we consider then the plays of ball, much of what is common is preserved, but much also is lost. Are the plays “amusing”? Compare the failures with the crab-louse. Or is there always a concept of gaining and loser, or competition between players? Think of the success (with the charts). In the plays of ball there is gaining and of the losers; but when a child launches his ball against the wall and catches up with it, this characteristic disappeared. Look at the roles held by the address and the chance; and with the difference between the skill in the failures and the address with tennis. Think now of plays as Ring has Ring O' Pinks ; here one finds the element “recreation”, but how much other characteristic features disappeared! And we can traverse many the other groups of plays in the same way, by seeing how the similarities emerge and are erased. And the result of this examination, here: a network complexes similarities overlapping and intersecting; sometimes total similarities, sometimes of the similarities of détail.
Je do not see a better expression to characterize these similarities that of resemblance of family, because various resemblances between the members of a family: conformation, the features, the color of the eyes, the step, the temperament, etc, overlap and intersect same manner. And I will say: the “plays” train a family. ”

Outdistance and convexity conceptual

Obviously, the concept of resemblance of family calls a concept of conceptual distance , which is closely related to the idea of gradual units, but that does not go without problems.

Recently, Peter Gardenfors ( Conceptual Spaces , MIT Close 2000) worked out a possible implementation of the theory of the prototype in terms of multidimensional spaces of features, in which a category is defined starting from the conceptual distance. The most central members of a category are “between” the peripheral members. Gardenfors postulates that the “natural” categories more have a convexity of conceptual space, in the direction where, if X and are elements of a category there, and if Z is “between” X and there, then Z has a good probability of also belonging to the category.

Multiple prototypes

However, if we mention the concept of play above, does there exist a single prototype or several? Recent linguistic data resulting from studies on the colors seem to indicate that the categories can have more than one focal element: in Tsonga for example, the term rihlaza refers to a continuum green-blue, but seems to have two prototypes, a focal blue and a focal green. It is thus possible to have simple categories having of the multiple prototypes, disjoined, in which case they could constitute the intersection of several convex units rather than a single unit.

Combinations of categories

All around us, we find examples of objects like a large man or a small elephant which combine one or more categories. This constituted a problem for extensionnelle semantics, where the semantics of a word such as red must be defined by the whole of the objects having this property. It is clear as this as well does not apply to modifiers such as small : a small mouse is something of very different from a small elephant .

Is the prototype of a man of 1,85m or a skyscraper of 120m? and Taylor 1988. The solution emerges from the contextualisation of the concept of prototype compared to the object prone to modification. In the case of composed such as of the red wine or a red face (of embarrassment), red can be comparable with difficulty with its prototypic direction, but indicates only one semantic slip starting from the respective prototypic color of the wine or the face. This joined the concept, introduced by Ferdinand de Saussure, of the purely differential definition of the concepts: “ not positively by their contents, but negatively by their relationship with the other terms of the système' ”

There remain other problems, for example that to determine which constituent category will contribute to which semantic feature. In the example of a bird of company ( fart bird ) 97, of company indicates the habitat applicable to the made up concept (a cage rather than nature), whereas bird indicates the type of natural protection (plumage rather than peeling).

Standard version and wide version

From 1978, and on the initiative of the pioneers themselves, then other linguists, the theory of the prototype evolved/moved: according to the ones (Lakoff…), it is only about one prolongation, according to others (Kleiber

  • Polysemia

the wide version takes into account the Polysémie, by supposing that there can be at the same time plurality of the references and intuitive unit of the significance ( blackbird would be monosemantic, and polysemous bird ). But there still, the criteria of determination of the membership of a category become variable then and little lights.

Conclusion

The theory of the prototype aroused an interest considerable and made up an undeniable projection in lexical Sémantique. It remains of topicality, in spite of the unsolved problems which it raised and the divergent versions to which it gave birth, going until a questioning of the concept even of prototype.

However, the theory did not really make it possible to answer the fundamental question up to now: why does one classify an object in such or such category?

See too

Random links:Corentin Celton (subway of Paris) | Geyster | Research center and of information for the development | Dull plover | Albert Counson