multiple realizability
The thesis that a mental state is the type it is independent of the physical realization of that mental state. See functionalism.

In general, if two events can be tokens of the same mental type (e.g., they are both thoughts about Vienna) while being tokens of two different physical types (e.g., one is a pattern of neural activations in a mammal's brain, one is a certain distribution of electrical currents in a creature made out of silicon), then the mental type in question is multiply realizable. The multiple realizability thesis is largely a negative thesis: a physical system realizes a mental state not in virtue of the particular stuff it is made of. Most philosophers attracted to the negative thesis of what mental states are not realized in virtue of are also attracted to the positive thesis that physical systems realize mental states in virtue of the abstract pattern of causal relations that parts of that system bear to each other.

In other words, multiple realizability relies on the functionalist contention that there is a higher-level functional description of physical states in terms of their causal role which abstracts from their lower-level physical constitution. It is supposedly with such functional properties that mental properties can be identified.

The ubiquity of multiple realizability convictions in cognitive science is difficult to underestimate. Fodor (1981) finds the position a natural result of a Turing machine functionalist perspective, and argues that identical 'programs' can run on physically different 'machines' and that neuroscientific data are therefore irrelevant. Putnam (1975) similarly claims: "The functional organization (problem solving, thinking) of the human being or machine can be described in terms of the sequences of mental or logical states respectively (and the accompanying verbalizations), without reference to the nature of the 'physical realization' of these states." (p. 373). And finally, Block has put this conviction as follows (Block 1980, p. 178):
[I]t is a simple matter to dream up a nomologically possible machine that satisfies [a given] machine table but does not have the designated physical state. Of course, it is one thing to say this and another thing to prove it, but the claim has such overwhelming prima facie plausibility that the burden of proof is on the critic to come up with reason for thinking otherwise.

(Note: See functionalism for a discussion of the logical relation between multiple realizability and functionalism).

Chris Eliasmith & Pete Mandik