Prototype Theory has been a controversial issue within and among different disciplines including linguistics, psychology, cognitive science, computer science and philosophy. Linguistics focuses on the appropriateness of Prototype Theory as a theory of meanings. Psychology and cognitive science look for experimental data to explain cognitive thinking processes in the brain and human perception. Computer science investigates the mathematical representations of concepts and their hierarchical structures. Philosophy questions the existence of entities and the way how these entities exist in different possible worlds. In this essay, I would like to investigate and dissect Prototype Theory slice by slice from different cross-disciplinary perspectives.
Get Help With Your Essay
If you need assistance with writing your essay, our professional essay writing service is here to help!
My focus of discussion will be based on the criticisms of Prototype Theory made by Laurence and Margolis (1999). They point out four problems of Prototype Theory: (1) The Problem of Prototypical Primes; (2) The Problem of Ignorance and Error; (3) The Missing Prototype Problem and (4) The Problem of Compositionality. In this essay, I argue that Prototype Theory is not a comprehensive and appropriate theory of meanings.
Firstly, for the Problem of Prototypical Primes, Laurence and Margolis (1999) propose from the perspective of cognitive science that “typicality effects do not argue for prototype structure since even well-defined concepts exhibit typicality effects” (p. 32), in which “well-defined concepts” means that people know and can produce the definitions of the concepts readily. Laurence and Margolis (1999) believe that typicality judgments are not about degrees of membership while prototypes are explained by typicality judgments based on speed and accuracy of categorization reflecting people’s views on the degree instances of an exemplar instantiate a category. A philosopher Wittgenstein (1999) suppose that the concept game cannot be defined by properties that are shared by all games but the concept game exhibits prototypical effects which correlate with family resemblance. The more properties a member of a category shares with the other members, the more prototypical it is. In linguistics, some linguists think that Prototype Theory is a better alternative theory than Classical Theory as it can account for prototype effects which are defined as having fuzzy category boundaries and central and marginal members in a category. Cognitive psychologist Rosch (1999) devised some experimental tasks of goodness-of-example ratings in terms of reaction times and priming. The findings show prototypicality as a consequence of human perception when members of a category are ranked differently and responded in different speed.
I argue that Prototype Theory is a good theory of meanings on the ground of its ability to bridge the indirect link between language and the world through the mind by prototype effects. The above views of different disciplines introduce the problem of whether typicality or prototype effects can be a solid foundation on which Prototype Theory is grounded. This also questions whether Prototype Theory is an appropriate alternative to the Aristotelian Classical Theory as a theory of meanings. Psychological findings account for prototype effects as a result of human perception. Prototypicality is therefore a feature in human thinking when it involves categorization of concepts from the world through the mind by language. This feature is applicable to most occasion no matter definitions are involved or not. Hence, I do not agree with Laurence and Margolis that prototype structures are mutually exclusive to definitions of concepts. Prototypes show the typical representation of the concepts and this does not contradict to whether people have the knowledge of the definitions of concepts. Wittgenstein illustrated that the concept game cannot be defined but people can still grasp the concept of game and its meanings by prototypes. This comes to the problem of concept formation and concept recognition. Knowledge and understanding of meanings of concepts can stem from the generalizations of prototypes. According to the Ogden & Richard’s Triangle, in linguistic terms, words are directly linked to thoughts (concepts), which in turn are directly linked to the world. In cognitive terms, language links with cognition through categorization, and cognition links to the world by perception. Prototypes bridge the link between the undefinable concepts in language through the protypicality feature in human perception of concepts in the mind to the possible world in which the concepts exist.
Secondly, for the Problem of Ignorance and Error, I argue that Prototype Theory is flawed in meaning determination in atypical and marginal members of complex concepts in exchange for cognitive merits. Laurence and Margolis (1999) propose that “concepts with prototype structure fail to cover highly atypical instances and incorrectly include non-instances” (p. 32). They use the concept grandmother to illustrate the problem. Prototypical grandmothers are old, gray-haired with glasses, kind to children and like baking cookies. However, the problem is that a person can satisfy these properties without being a grandmother and a person can be a grandmother without satisfying these properties. I think George Lakoff provided a good explanation on this problem. Lakoff (1999) illustrates the Idealized Cognitive Model (ICM) that some categories are characterized by clusters of cognitive models. Some concepts are so complex that no definitions can cover the full range of cases. Lakoff (1999) used the concept mother to illustration the clusters of models in terms of birth, genetic, nurturance. marital and ancestral models. Hence, there are many subtypes of mothers like stepmother, surrogate mother, adoptive mother, foster mother, biological mother and donor mother. As far as I am concerned, my prototypical understanding of mother is a woman who contributed my genes, gave birth to me, raised me and is my father’s current wife. More importantly, my prototype of mother is the same woman with the above attributes. Laurence and Margolis’s satisfaction of properties demonstrates how a subordinate prototypical model of the concept grandmother fails to fulfill properties of other cognitive models. Therefore, the authors are right to say that prototype representations lack sufficient richness to include all birds or all tigers, and they are too rich to embody information that includes things that are not birds or tigers. Birds without feathers are still birds. Three-legged, tame and toothless tigers are still tigers. They are just the atypical members of the category.
From the perspective of psychology, the above reveals the cognitive notion that how a concept is deployed determines what items fall under it. Taylor (1989) suggests that prototype categories are flexible, in comparison to Aristotelian categories, to be able to accommodate new and unfamiliar data. Therefore, new entities and new experiences can be readily associated to a prototype category as peripheral members without having to create new categories and redefinition of existing categories. This structural stability and flexible adaptability is a merit of Prototype Theory in cognition. Gazzaniga, Heatherton and Halpern (2010) define categorization as grouping things based on shared properties, which serves to reduce the amount of knowledge we must hold in memory and is an efficient way of thinking. When it comes to atypical instances, this merit is achieved by serving cognitive functions not for meanings of words, but for structured organization and categorization without exploiting cognitive resources.
Thirdly, for the Missing Prototype Problem, I argue that Prototype Theory cannot cover all senses and concepts when they lack prototypes. Concept possession is viewed differently from different disciplines. According to Frege’s semantic theory, senses are the cognitive content of linguistic expressions. In linguistics, the meaning of linguistic expressions derives from two sources including the language they are part of and the world they describe. Sense is the relationship between expressions within the language. Sense is fundamental of meanings. In psychology, according to Gazzaniga, Heatherton and Halpern (2010), concepts are mental and schematic representations that groups or categorizes objects, events, or relations around common themes. As opposed to the psychological view of concepts being mental representations, concepts are some abstract and non-mental entities in philosophy. No matter how concepts are viewed in different perspectives, there is one thing in common that concepts have their own content to convey meaning.
Laurence and Margolis (1999) propose that many concepts lack prototypes and they are not associated with typicality judgments so people fail to represent central tendencies. The authors quote Jerry Fodor’s example, showing that there are no prototypical properties of grandmothers most of whose grandchildren are married to dentists. This reveals the problem that many complex concepts lack prototype structure and prototypical properties. Although it seems natural for people to have representative exemplars of concepts coming to mind easily, it does not necessarily mean that this applies to all concepts. It is still possible for people to grasp a concept without knowing a prototype. For instance, some abstract concepts like belief, sincerity, utopian, love, hatred and adversity lack prototype structure. These abstract concepts are metaphysical but not epistemic. They do not possess prototypical properties but their meanings are still conveyed by language. Some examples indicating exclusivity or unknown and supposed novelty also lack prototype structure. For instance, non-human, substance weighing more than one kilogram, and not Chinese are exclusive examples. Examples like new inventions, Martian and doomsday demonstrate unknown and supposed novelty. Therefore, concept possession does not necessarily require prototypes. This is an inevitable weakness of Prototype Theory.
Find Out How UKEssays.com Can Help You!
Our academic experts are ready and waiting to assist with any writing project you may have. From simple essay plans, through to full dissertations, you can guarantee we have a service perfectly matched to your needs.
View our academic writing services
Fourthly, I argue that Prototype Theory cannot account for complex conceptual combinations. For the Problem of Compositionality, Laurence and Margolis (1999) propose that the Prototype Theory cannot account for compositionality adequately since the prototypes of complex concepts are not generally a function of the prototypes of their constituent concepts. This can be supported by Osherson and Smith’s (1981) objection to fuzzy set theory’s model of compositionality. Fuzzy set theory is a theory in computer science that sheds light on Prototype Theory in linguistics.
According to Laurence and Margolis (1999), fuzzy set theory builds on the notion of graded membership; a fuzzy set is a function that assigns to each item in the domain of discourse a number between 0 and 1, measuring the degree to which the item is in the set. The higher the values, the higher degree of membership. I think that fuzzy set theory is so computational that it ignores the fact that prototypical features involve knowledge of the real world in terms of culture and experience. The methodology fuzzy set theory adopted to explain the measurement of similarity and degree of membership reveals the setbacks of Prototype Theory-the inability to apply for compositional conceptual combinations. Osherson and Smith’s (1981) use an example of conjunctive concepts-striped apple to explain fuzzy intersections of compositional concepts of striped and apple. The authors point out that a very good instance of a striped apple will inevitably be a poor instance of an apple. Fodor and Lepore (1996) suggest that prototypes are not always a function of the prototypes of their constituents and they come to the conclusion that concepts cannot be prototypes as prototypes are not compositional and concepts must be compositional. They illustrate their arguments by pet fish. Prototypical pet fish is goldfish. Prototypes of pet can be cats and dogs. Prototypical pet fish makes poor examples of both pet and fish. Hence, this supports the notion that the prototypes of complex concepts are not generally a function of the prototypes of their constituent concepts. As far as I am concerned, prototypes are subject to recursion but prototypes of compositional concepts complicate the matter in determining meanings. Prototypes are recursive because prototype of a superordinate level furniture can be chair. Prototype of chair can be dining room chair and the prototypes of the subordinate levels go on and on. When concepts are no longer simple but compositional like striped apple, recursion of prototypes stops. Prototypes of striped can be evenly or randomly distributed, and in vertical, horizontal or diagonal directions. Prototypes of apple can be red, round and juicy. When these prototypes of striped and apple have to merge with each other, the compositionality is an odd conceptual combination where prototypicality can no longer account for it.
In spite of its ability to bridge the indirect link between language and the world through the mind, Prototype Theory is not comprehensive enough to be an appropriate theory. It is not versatile enough as its only strength lies in accounting prototype effects.
(2000 words)
Cite This Work
To export a reference to this article please select a referencing style below:
