Page 96 - Encyclopedia of Philosophy of Language

Language and Mind
Bibliography
Blackburn S 1984 The individual strikes back. Synthese 58: 281-301
Gilbert M 1983 On the question whether language has a social nature. Synthese 56: 301-18
Kripke S 1982 Wittgenstein on Rules and Private Language. Harvard University Press, Cambridge, MA
McDowell J 1984 Wittgenstein on following a rule. Synthese 58: 325-63
Pitcher G (ed.) 1966 Wittgenstein: The Philosophical Inves- tigations. Macmillan, London
Travis C 1989 The Uses of Sense. Oxford University Press, Oxford
Wittgenstein L 1967 (trans. Anscombe G E M ) Philosophical Investigations, 2nd edn. Basil Blackwell, Oxford
Wright C 1986 Does Philosophical Investigations 1.258-60 suggest a cogent argument against private language? In: McDowell J, Pettit P (eds.) Subject, Thought and Context. Oxford University Press, Oxford
The notion of representation is a familiar, if philo- sophically problematic, one. It becomes more prob- lematic, and less familiar, when modified with the epithet 'mental.' Nevertheless, the notion of mental representation is crucial both in cognitive psychology and cognitive science. It is also crucial in linguistics itself, at least for those who accept Chomsky's views that grammars describe part of the contents of the minds of language users, and that linguistics is cor- rectly construed as part of cognitive psychology.
1. Representation and Mental Representation
It is a fundamental assumption of cognitive psy- chology and cognitive science that explanations of behavior make reference not only to inputs and out- puts but to information encoded in the mind. In order to provide an information processing account of a particular ability it is necessary, therefore, to describe how inputs, outputs, and stored information are in- ternally encoded. It is natural to think of these encod- ings as depending on a mental representation scheme or language. Following Jerry Fodor (1975) this lan- guage is usually referred to as the language of thought, though different mental faculties may use different representational schemes. Although this view of men- tal processing is widely accepted, it raises a very difficult question: what is a mental representation?
One can make a start on answering this question by considering everyday types of representation that are easier to understand. A simple two-dimensional town map represents space spatially. In general, however, there need be no such direct correspondence between what is represented and how it is represented. British Ordnance Survey maps represent the third spatial dimension using contour lines, and they represent things in the landscape by symbols that may (church with a tower) or may not (coach station) resemble what they represent. In such a representational
scheme, the correspondence between what is rep- resented and the elements of the scheme must play a role in both the production and the use of particular representations—only in aberrant cases will it not do so. In particular, resemblance is not sufficient for rep- resentation, as a consideration of portraits, par- ticularly those of identical twins, shows. Causation is crucial in determining what something represents. Indeed, some philosophers (e.g., Fodor 1990) have suggested purely causal theories of how a mental state comes to represent something in the world.
There is little difficulty understanding how maps work. But maps requirepeople to create and interpret them. We, as mapmakers, create the representational schemes that allow us to make particular maps and, thus, to achieve our navigatory goals. And we, as map users, have the goals that make maps useful. Mental representations differ from maps in both respects. First, the meanings of the elements of a system of mental representation are not arbitrarily stipulated. They arise from natural effects that the environment has on people or animals. However, not every effect that the world has on an animal gives rise to a mental representation. For an effect to be a representation, it must have the function of providing information about what it represents. Second, although natural effects can have representational functions imposed upon them, mental representations typically have functions that derive from the natural goals of people or animals. Furthermore, an account of mental rep- resentations cannot be based on the idea of a person inside the head setting them up and using them— homunculus theories cannot explain cognition.
If the job of cognitive scientists is to discover the representational schemes used by the mind, that job is very different from a mapmaker's. Mapmakers decide what to represent, taking into account how their maps will be used, and they stipulate a representational
74
Representation, Mental A. Garnham