. Problems about meaning can be put into two main and interrelated groups: What is meaning, in the sense in which words and sentences have meaning? What different kinds of meaning are there, and how are they related to various other notions?
Discussions of what meanings are have mainly concerned words and sentences, which differ from each other: We construct sentences, and can understand ones we have not seen before. Sentences have meaning because of the words in them, while words only have meaning because they are fitted to play a role in sentences. Different words may have meaning in different ways, but whether all words have meaning is disputed. Do proper names (see below)? Does ‘to’ in ‘I want to go’?
During the earlier part of the twentieth century notions like naming, referring to, or standing for have dominated such discussions. Naming (denotative, referential) theories of meaning say that a word’s meaning is what it names or stands for, or else its relation to that. Cf. the slogan unum nomen unum nominatum: ‘For every name there is exactly one thing named’. Proper names are taken as the primary case. Hence the nickname ‘Fido’-Fido theory. The word ‘Fido’ has as its meaning the dog Fido which it names. On this theory a general word like ‘dog’ could stand for the UNIVERSAL, doghood (Russell), or the class of dogs, or different dogs on different occasions (also Russell). ‘Red’ could stand for the colour red, ‘runs’ and ‘running’ for the action of running, even perhaps ‘if’ for the notion of doubt or conditionality.
Ideational theories make words stand for ideas or thoughts etc. (Aristotle, Locke). They provide a single kind of thing (ideas) for very different kinds of words to stand for, though usually without explaining the notion of standing for.
All these are relational or correspondence theories. They say that a word’s meaning is or involves a thing (physical, mental or abstract) to which it is related. Largely through the influence of Wittgenstein they have been much attacked, though they have also experienced a certain revival recently.
This attack on relational theories implied that meanings are not ‘things’, and led to the slogan ‘Don’t ask for the meaning, ask for the use’. Use theories explain meaning in terms of use. They provoke questions about whether the connexion between meaning and use is the same for words and for sentences, whether meaning is as wide a notion as use (see below on speech acts), and whether what matters most is actual use or rules for use. This last distinction is one version of that between de facto and de jure theories. De facto theories explain meaning in terms of what happens or is the case, e.g. how people actually do use words. De jure theories explain it in terms of norms, rules, conventions, standards, or in general what ought to be the case, e.g. they might claim that ‘I didn’t do nothing’ cannot properly mean ‘I didn’t do anything’, despite Cockney usage.
Causal or stimulus/response theories explain the meaning of a word or sentence in terms of its effect on the hearer or the cause of the speaker’s uttering it. They are examples of de facto theories, stemming from BEHAVIOURISM, and they claim the advantages of a scientific approach. They become naming theories too, if they say that the object named is the meaning while the object’s effects are what make it the meaning. A causal theory of names and ‘natural kind words’ (like ‘lion’ ‘water’) has recently been advanced by S. Kripke and H.Putnam, in opposition to description theories like Russell’s (see below). The same authors have advanced a causal theory of REFERENCE, whereby what if anything one refers to is partly determined by the things to which one stands in certain causal relations.
A theory of sentence-meaning also stimulated by science is the verification theory, whereby a sentence’s meaning is the method of verifying it; cf. logical POSITIVISM for this, and for the operationalist theory of word meaning. Picture theories are analogues for sentences of naming theories for words and like naming theories they are correspondence theories. They are especially associated with LOGICAL ATOMISM, and with the correspondence theory of TRUTH. On picture theories sentences, whether true or false, have meaning because they picture possibilities; true sentences picture those possibilities which are facts. But can picture theories cater for the stating involved in stating facts?
A common modern theory says that to give the meaning of a standard indicative sentence is to give its TRUTH CONDITIONS. These, however, are not given for sentences individually but holistically, by constructing a ‘theory of truth’ for a whole language. This starts with axioms saying what the words mean and rules for combining them into sentences, in such a way that theorems can be derived, each of which says of some sentence that it is true if and only if so-and-so is the case, where ‘so-and-so’ represents what, intuitively, the sentence means. Such a theorem is called a T-sentence. An example would be, ‘“Snow is white” is true if and only if snow is white’. When, as here, what follows ‘if and only if’ is the same, without the inverted commas, as what precedes it, the T-sentence is called homophonic (literally, ‘same-sounding’). A theory of truth whose axioms yield the intuitively right T-sentences is said to satisfy Convention (or Criterion) T. See also TRANSLATION (for ‘radical interpretation’). Discussion of all this has largely concerned whether truth conditions alone can provide adequate resources for a proper theory of meaning. See REALISM. In a more limited sphere a truth conditions account of meaning has been offered as a way of avoiding positing the existence of certain properties, such as that of being present: ‘1994 is present’ will get its meaning by being true if and only if uttered during 1994, so that no property of ‘presentness’ is needed.
With the general question of what meaning is belongs also the question of synonymy, i.e. when words or sentences have the same meaning.
Turning to the second of the original questions, about kinds of meaning, and related notions, we can for suitable words distinguish between intension and extension (cf. INTENSIONALITY AND INTENTIONALITY), and between various allied notions. Thus Mill distinguishes connotation from denotation. A word denotes the things it applies to, and connotes the attributes it implies that those things have. ‘Man’ connotes, perhaps, the attribute being a rational animal, and denotes all men. ‘Connotation’ can refer to the relation or to what is connoted, and ‘denotation’ similarly. The comprehension of ‘man’ is the whole set of properties shared by all men, or else the set which (logically) must be shared by them.
Frege distinguishes Sinn and Bedeutung, standardly translated sense and reference respectively, though other translations exist, and ‘meaning’ has been used for each of them at different times. The phrases ‘the evening star’ and ‘the morning star’ have the same reference, Venus, but different senses. Frege uses this to explain why ‘The evening star is the morning star’ is not trivial, like ‘Venus is Venus’.
The relations between these terms are complex. Roughly, a term denotes, independently of occasion, all those things we can refer to on a given occasion by using certain phrases containing the term. ‘Cats’ denotes cats in general, and we can use a phrase like ‘that cat’ to refer to, say, Tiddles on some occasion. Strictly it is we who refer to Tiddles by saying ‘that cat’, but the phrase ‘that cat’ can itself be said to refer to Tiddles on this occasion. (‘Denote’, especially, is often used loosely.)
A term has divided reference (Quine) if, like ‘shoe’ but unlike ‘water’ or ‘red’, it can be used, without additions like ‘piece of’, to refer to different objects.
The connotation/sense distinction is difficult, and controversial, but an example may illustrate one way of making the distinction. Is it contradictory to say, ‘The queen of England is not queen at all but illegitimate’? Yes, if ‘the queen of England’ is interpreted as having connotation, since the statement then implies that the person referred to by the phrase has the property of reigning. But the sense of the phrase can be used to pick out the person, without committing the speaker to the truth of this implication, for a hearer, for example, who believes that Elizabeth II is legitimate. If the phrase is thus interpreted as still having sense, but not connotation, the statement is not contradictory. If Elizabeth II were not legitimate, the sentence could be used to say so without contradiction, and might be useful for ensuring that the supposed hearer knew who was being talked about. Sense and reference apply to subject expressions and sometimes to predicates and to sentences (Frege thought the reference of a predicate was a concept, and of a sentence was its TRUTH-VALUE). Connotation and denotation can apply to subject expressions and predicate expressions, but not to sentences. Sense is close to the non-technical notion of sense.
Proper names, which provide the model of meaning for naming theories, raise problems about whether they have connotation or sense or both. ‘That man there’ picks someone out as being a man and being in a certain place, but ‘John’ makes no obvious reference to any of its bearer’s properties. In what sense are proper names words? Do they form part of a language? They are hardly meaningless, but they do not usually appear in dictionaries. Russell followed Mill in thinking they lack connotation. However, he thought this only of logically proper names, i.e. those names which were not abbreviated descriptions, as he thought ordinary proper names in fact were. He thought ‘Socrates’ was not really a name at all, but an abbreviation for, for example, ‘the philosopher who drank hemlock’. (Cf. DESCRIPTIONS.) This is the view criticized by Kripke and Putnam (above). A logical subject is either the subject of a sentence in a logically ideal language, i.e. a language where a sentence’s real and apparent subjects coincide, or it is what such a subject refers to.
Under the influence of logical positivism, cognitive, descriptive or factual meaning has been distinguished from other kinds, notably emotive, evaluative, and prescriptive: see NATURALISM. Arising out of these distinctions and generalizing from them is Austin’s theory of SPEECH ACTS, which distinguishes between the meaning of what is said and what he calls the ‘illocutionary force’ of saying it. Attempts have been made to analyse the meanings of some words, including ‘good’, ‘true’, ‘probable’, in terms of speech acts they are used in making, and to ground meaning itself in illocutionary force (a kind of use theory). It is here that the different kinds of IMPLICATION (presupposition, implicature, etc.) and their roles become relevant.
It is also important to distinguish, especially concerning TOKEN-REFLEXIVES, between what meaning words and sentences have in general and what they mean, or what the speaker means by them, on a given occasion. This raises the question how meaning is related to intention, for people can ‘mean something’ both in the sense of referring to something (‘I mean you’) and as intending to do something (‘I meant to come yesterday’). They may also ‘mean’ something in the sense of suggesting something beyond what they mean in the direct sense (‘In saying “Time’s getting on” he meant it was late, and he also meant it was time for you to go’. Cf. IMPLICATION again). An interesting question, discussed by Wittgenstein, concerns what it is to mean what one says.
In fact not only words, sentences, and people can be subjects of the verb ‘mean’ but also actions, works of art, and natural events, states and processes. ‘Mean’ can have natural objects or events as its subject when there are symptoms (‘Those spots mean measles’) or causes (‘clouds mean rain’) or things having value or importance (‘That locket means a lot to me’). ‘His life has no meaning’ seems to come between this last sense and the sense of pattern, order, idea in which works of art have meaning. Actions have meaning either in these ways or in the way sentences have it. Grice distinguishes the various kinds of meaning into two main groups, natural, the group applying natural events etc., and non-natural, the group applying to people and symbols, words and sentences, etc.
Aristotle, De Interpretatione, esp. chapters 1–4, trans. with commentary in J.L.Ackrill, Aristotle’s Categories and De Interpretatione, Oxford UP, 1963. (Basis of Aristotle’s theory of meaning, though only elementary.)
L.J.Cohen, The Diversity of Meaning, Methuen, 1962 (revised 1966). (Chapter 2 introduces de facto/de jure distinction.)
E.Daitz, ‘The picture theory of meaning’, Mind, 1953, reprinted in A.Flew (ed.), Essays in Conceptual Analysis, Macmillan, 1956.
D.Davidson, ‘Truth and meaning’, Synthèse, 1967, revised in ‘Radical interpretation’, Dialectica, 1973, both reprinted with other relevant material in his Inquiries into Truth and Interpretation, Clarendon, 1984. (Leading exponent of truth-conditions theory. See also for a contrasting view M.A.E.Dummett, ‘What is a theory of meaning? II’ in G.Evans and J.McDowell (eds), Truth and Meaning, Oxford UP, 1976, reprinted with other relevant material in his The Seas of Language, Clarendon 1993.)
G.Frege, ‘Sinn und Bedeutung’, 1892, trans. in P.Geach and M. Black, Translations from the Philosophical Writings of Gottlob Frege, Blackwell, 1952, and also in H.Feigl and W.Sellars (eds), Readings in Philosophical Analysis, Cornell UP, 1949.
N.Goodman, ‘On likeness of meaning’, Analysis, vol. 10, 1949, reprinted in M.Macdonald (ed.), Philosophy and Analysis, Blackwell, 1954. (Difficulties over synonymy.)
H.P.Grice, ‘Meaning’, Philosophical Review, 1957, reprinted in P. F.Strawson (ed.), Philosophical Logic, Oxford UP, 1967, and also, along with other relevant material, in Grice’s Studies in the Way of Words, Harvard UP, 1989. (Criticizes causal theory and develops theory using intention (which might, however, itself be called a causal theory).)
J.N.Keynes, Formal Logic, 1884, 4th (revised) edn, 1906. (Full details of traditional terminology of intension and extension. For later terms cf. also R.Carnap, Meaning and Necessity, Chicago UP, 1947, chapter 3.)
S.Kripke, Naming and Necessity, 1980 Blackwell (original version, 1972). (Causal theory of names, natural kind words, and reference. Cf. also S.P.Schwartz (ed.), Naming, Necessity and Natural Kinds, Cornell UP, 1977, including inter alia articles by H.Putnam and G.Evans.)
J.Locke, An Essay concerning Human Understanding, 1689, book 3. (An intentional theory.)
J.S.Mill, A System of Logic, 1843, Longmans, Green, book 1, chapter 2, esp. §5. (Connotation/denotation distinction. Classic discussion of names. Cf. final footnote on senses of ‘connotation’.)
A.W.Moore (ed.), Meaning and Reference, Oxford UP, 1993. (Includes several classic items.)
Plato, Cratylus. (Earliest surviving connected discussion of meaning.)
W.V.O.Quine, Word and Object, Wiley, 1960, § 19. (Divided reference.)
B.Russell, An Inquiry into Meaning and Truth, Allen and Unwin, 1940, esp. chapters 1–7, 13–15. (Complex theory of meaning of words and significance (as he calls it) of sentences. Cf. his My Philosophical Development, Allen and Unwin, 1959, chapters 13, 14.)
D.Wiggins, ‘The sense and reference of predicates’, Philosophical Quarterly, 1984, reprinted in C.Wright (eds), Frege: Tradition and Influence, Blackwell, 1984. (Relevant on sense and connotation, and on the interpretation (and correction) of Frege.)
L.Wittgenstein, Philosophical Investigations, Blackwell, 1953, esp. §§ 1–43. (Classic criticism of naming theories in favour of use theory.)
This is the complete article, containing 2,558 words
(approx. 9 pages at 300 words per page).
