Return to Kent Bach's Homepage

[entry, MIT Encyclopedia of the Cognitive Sciences]


Kent Bach

MEANING: centrally, the feature(s) of an expression (over and above its

form) that determine its contribution to what a speaker says in using

it; also, the content of the communicative intention of the speaker in

using an expression (even if that use departs from the expression's

meaning). Accordingly, any discussion of meaning should distinguish

speaker's meaning from linguistic meaning.

We think of meanings as what synonyms and translations have in common,

what ambiguous expressions have more than one of, what meaningful

expressions have and gibberish lacks, and what competent speakers grasp.

Yet linguistic meaning is a puzzling notion. The traditional view is

that the meaning of a word is the concept associated with it and, as

FREGE suggested, what determines its reference, but this plausible view

is problematic in various ways. For starters, it is not clear what

CONCEPTS are, what the relevant sort of association with words is, or,

indeed, that every word has any concept, much less a unique concept,

associated with it. Wittgenstein (1953) even challenged the Platonic

assumption that all the items to which a word applies must have

something in common. Unfortunately, there is no widely accepted

alternative to the traditional view (there is also skepticism about

meaning, at least as traditionally conceived, registered in various ways

by such prominent philosophers as Wittgenstein, Quine 1960, Davidson

1984, Putnam 1975, and Kripke 1982; for review of the debates they have

generated see Hale and Wright 1997, chs. 8 and 14-17). Psychological

approaches based on prototypes or on semantic networks, as well as

COGNITIVE LINGUISTICS, seem to sever the connection between meaning

and reference. The most popular philosophical approaches to sentence

meaning, such as truth-conditional, model-theoretic, and POSSIBLE

WORLDS SEMANTICS, also have their limitations. They seem ill-equipped

to distinguish meanings of expressions that necessarily apply in the

same circumstances or to handle non-truth-conditional aspects of


Here are some foundational questions about meaning, as difficult as they

are basic:

1. What are meanings?

2. What is it for an expression to have meaning?

3. What is it to know the meaning(s) of an expression?

(More generally, what is it to understand a language?)

4. What is the relationship between the (or a) meaning of an

expression and what (if anything) it refers to?

5. What is the relationship between the (or a) meaning of a

complex expression and the meanings of its constituents?

An answer to question 1 would say whether meanings are psychological,

social, or abstract, although many philosophers would balk at the

question, claiming that meanings are not entities in their own right,

and insist that answering 2 would take care of question 1. An answer to

3 would help answer 2, for what expressions mean cannot be separated

from (and is perhaps reducible to) what people take them to mean. And

question 4 bears on 3. It was formerly assumed that the speaker's

internal state underlying his knowledge of the meaning of a term

determines the term's reference, but Putnam's (1975) influential TWIN

EARTH thought experiments have challenged this "internalist" or

"individualist" assumption. In reaction, Chomsky (1986, 1995) and Katz

(1990) have defended versions of internalism about knowledge of language

and meaning.

Question 5 points to the goal of linguistic theory: to provide a

systematic account of the relation between form and meaning. SYNTAX is

concerned with linguistic form (including LOGICAL FORM, needed to

represent scope relationships induced by quantificational phrases and

modal and other operators), SEMANTICS with how form maps onto linguistic

meaning. The aim is to characterize the semantic contributions made by

different types of expression to sentences in which they occur. The

usual strategy is to seek a systematic, recursive way of specifying the

meaning(s) of a complex expression (a phrase or sentence) in terms of

the meaning(s) of its constituents and its syntactic structure (see

Larson and Segal 1995 for a detailed implementation). Underlying this

strategy is the principle of semantic COMPOSITIONALITY, which seems

needed to explain how a natural language is learnable (but see Schiffer

1987). But there are difficulties for compositionality, e.g. regarding

the philosophically puzzling cases of conditional sentences and

propositional attitude ascriptions, as well as various constructions of

concern to linguists, such as genitive and adjectival modification. For

example, although 'Rick's team' is a so-called possessive phrase, Rick's

team need not be the team Rick owns--it might be the team he plays for,

coaches, or just roots for (not that any relation will do). Or consider

how the force of the adjective 'fast' varies in the phrases 'fast car,'

'fast driver,' 'fast track,' and 'fast race' (see Pustejovsky 1995 for a

computational approach to such problems).

The study of speaker's meaning belongs to PRAGMATICS. What a speaker

means in uttering a sentence is not just a matter of what his words

mean, for he might mean something other than or more than what he says.

For example, one might use "You're another Shakespeare" to mean that

someone has little literary ability and "The door is over there" to mean

not only that but also that the person should leave. The listener has to

figure out such things, and also resolve any AMBIGUITY or VAGUENESS in

the utterance and identify the references of any INDEXICALS AND

DEMONSTRATIVES. GRICE (1989) ingeniously proposed that communicating

involves a distinctive sort of audience-directed intention: that one's

audience is to recognize one's intention partly on the supposition that

they are intended to recognize it. This idea, which has important

applications to GAME THEORY (communication is a kind of cooperative

game), is essential to explaining how a speaker can make himself

understood even if he does not make fully explicit what he means, as in

IMPLICATURE. Understanding a speaker is not just a matter of

understanding his words but of identifying his communicative intention.

One must rely not just on knowledge of their linguistic meaning but also

on collateral information that one can reasonably take the speaker to be

intending one to rely on (see Bach and Harnish 1979 for a detailed

account). Communication is essentially an intentional-inferential

affair, and linguistic meaning is just the input to the inference.




Bach, K and R. M. Harnish, 1979, *Linguistic Communication and

Speech Acts*. Cambridge: MIT Press.

Chomsky, N., 1986, *Knowledge of Language*. New York: Praeger.

Chomsky, N., 1995, 'Language and nature.' *Mind* 104: 1-61.

Davidson, D., 1984, *Essays on Truth and Interpretation*. Oxford:

Oxford University Press.

Grice, P., 1989, *Studies in the Way of Words*. Cambridge, Mass.,

Harvard University Press.

Hale, B. and C. Wright, eds., 1997, *The Blackwell Companion to

the Philosophy of Language*. Oxford: Blackwell Publishers.

Katz, J. J., 1990, *The Metaphysics of Meaning*. Cambridge, Mass.:

MIT Press.

Kripke, S., 1982, *Wittgenstein on Rules and Private Language*.

Cambridge, Mass: Harvard University Press.

Larson, R., & G. Segal. 1995, *Knowledge of Meaning*. Cambridge,

Mass.: MIT Press.

Lyons, J., 1995, *Linguistic Semantics: An Introduction*. Cambridge,

Eng.: Cambridge University Press.

Pustejovksy, J., 1995, *The Generative Lexicon*. Cambridge, Mass.:

MIT Press.

Putnam, H., 1975, 'The meaning of "meaning".' In *Language, Mind, and

Knowledge*, K. Gunderson, ed. Minneapolis: University of Minnesota

Press, pp. 131-193.

Quine, W.V., 1960, *Word and Object*. Cambridge, Mass.: MIT Press.

Schiffer, S., 1972, *Meaning*. Oxford: Oxford University Press.

Schiffer, S., 1987, *The Remnants of Meaning*. Cambridge, Mass.:

MIT Press.

Wittgenstein, L., 1953, *Philosophical Investigations*. New York: