The semantics of an expression may have two *aspects*:
(1) its meaning (an "intension")
(2) its reference (or "extension")
The meaning of an expression--"expression meaning"--must be distinguished from "speaker meaning".
There are different kinds of speaker meaning, among them ...:
(1) what speakers mean by an expression
(2) what a speaker meant (intended to communicate, or to express) in uttering an expression
What an expression means is derived from cases of the former variety: roughly, what an expression means is what its users mean by it when using it for communicative purposes.
The word "hut" is an expression type. When an instance of this type is uttered on a particular occasion, it is a so-called "token" of this type. The linguistic meaning of an expression token is (equal to) the meaning of the expression type it belongs to. The German expression type "Hütte" means 'hut', and each token of that German expression type means 'hut', too.
The meaning of a complex expression is completely determined by the meanings of the parts it is composed of, together with the way in which the parts are composed. This is called the principle of compositionality. Often, however, an expression which looks like a complex expression really is an idiom. As such, it has a meaning which is not a function of its parts and their composition. To the idiomatic meaning of "leave s.o. in the lurch" the principle of compositionality does not apply. (Idiomatic expression typically have both meanings: (a) their idiomatic meaning, and (b) a compositional meaning.)
Often, sentences are "semantically incomplete", which means that their meaning contains 'gaps' which prevent their having truth conditions. For example, the sentence "He left me in the lurch" is semantically incomplete, and thus has no truth conditions. One reason is that it contains an indexical, "he": the sentence expresses that there is one particular entity which left someone else in the lurch, but does not specify which entity that is, which prevents the sentence from having determinate truth-conditions. The same holds, mutatis mutandis, for the indexical "me".
Kaplan's "character" (semantic meaning), and "semantic content"
Semantically incomplete sentences have no truth-conditions. But according to the predominant conception of sentence meaning, the meaning of a sentence consists (or is completely determined by) the conditions of its truth. In order to avoid cases in which meaningful sentences have no truth-conditions, most philosophers of language endorse the construal of an entity called "semantic content" (the term was introduced by David Kaplan). The "semantic content" of a sentence is what emerges when all semantic 'gaps' in a sentence token's meaning are filled (taking the relevant information from the context of utterance), such that a truth-evaluable entity emerges.
Along with the term "semantic content", Kaplan introduced the notion of a sentence's "character". A sentence (type)'s "character" is its linguistic meaning--an entity which often, if not always, is semantically incomplete, i.e., does not specify fully determinate truth-conditions. The German sentence "Sie sind albern" means "You are silly". It is not truth-evaluable, because there is a semantic gap: "you" is indexical, and does not determine any particular entity. Consequently, it remains unclear to which entity the predicate "silly" is being ascribed. As was just said, the "semantic content" of a sentence token is what emerges when the "gaps" in the sentence's meaning are filled with reference to the relevant context of utterance. When someone says "You are silly" to Napoleon Bonaparte, then the semantic content is, roughly, 'Napoleon Bonaparte is silly'. This semantic content is true when Napoleon Bonaparte is silly: it does have truth-conditions.
expression
[under development]
o Expression (conventional/occasion -- cf. Grice) meaning ("Bedeutung") / speaker meaning ("Gemeintes")
=> The role of context; Kaplan's 'semantic content'; context-invariance of expression meaning ('Bedeutung')
[Motivation: Cappelen & Lepore 2010: 'content' with minimal (!) context-sensitivity]
? Distinction reference (referring expressions, such as 'bachelor') / extension (non-referring 'nominals', such as 'green')
[Motivation: ?]
o Presupposition (narrower sense (Frege, Strawson) / broader sense)
[Motivation: Russell's analysis of 'denoting phrases']
[Distinction extension/bearer; pace Tarski 1944.]
[Motivation: Tarski 1944?]
----------------------------------------------------------------------------------------------------------------------------------------------------------------
o Analysis of 'linguistic meaning'; analysis of 'speaker meaning' / 'communication'; the Gricean programme
o 'Implicature' (etc.)
o the failure of the truth-conditional approach to meaning; varieties of non-truth-evaluable meaning
[Motivation: @ethical judgements; (TC and) @moral norms; TC and charged terms]
----------------------------------------------------------------------------------------------------------------------------------------------------------------
o reference:
o Distinction description / (rigid) designation [various terminological alternatives]
o The 'causal' construal of reference (suits proper names, such as 'Homer')
o The 'descriptive' construal of reference (suits descriptive nominals, such as 'murderer of Caesar')
o opaque reference
[Motivation: Kripke 1980 (result far from spectacular, given o.r.)]
o Descriptive nominals and 'essence' (nominalist conception; distinction to realist conception)