While the traditional view was that in order to understand language and our linguistic practices we must explain meaning, the 'pragmatic turn' emerging within the writings of various philosohpers of the second half of the twentieth century caused a basic change of the perspective: the tendency is to concentrate directly on explaining the linguistic practices and leave the need for explaining meaning to emerge (or, as the case may be, not to emerge) subsequently.
Just as all men," he continues, "have not the same writing, so all men have not the same speech sounds, but the mental experiences, which these directly symbolize, are the same for all, as also are those things of which our experiences are the images."
The pragmatic turn Notice the shift of focus brought about by the Wittgensteinian view: we abandon the assumption that explaining meaning must necessarily precede investigating our linguistic conduct; now we concentrate directly on explaining the conduct and leave the need for explaining meaning to emerge subsequently – or, as the case may be, not to emerge.
The relevant patterns are forced upon us not (directly) by natural selection, but by the ongoing demands of our peers.
However, the traditional logic with its extensional semantics was quickly deemed to be insufficient - the range of natural language phenomena which could be directly captured by its means was only had to be found scanty.
Two senses of 'semantics' I think that it is of crucial importance to point out immediately that the term semantics is used to cover what are in fact two different enterprises, only one of which is directly relevant for linguistics and philosophy.
Take for example the large body of studies concerning the nature of definite and indefinite descriptions, which have persuasively shown that to see these locutions directly in terms of classical, Fregean quantification is inadequate and may be severely misguiding.
Sameness of truth values is an equivalence; and it directly reifies into the two truth values.
We may define function f with its domain equal to the set of all natural numbers by defining f(0) and giving the recipe how to compute f(x') out of f(x), or we may base the definition on a more complicated system of rules generating natural numbers, but it is impossible to define such a function directly, bypassing generating rules altogether.
Maybe there is a God who is able to grasp infinity directly and who thus perceives our positing meanings in this way; but for us the compositional way is the only way. Any more direct differentiation is our mythical creature that is licensed by set theory which embraces actual infinity and which can only be talked about, but never really demonstrated.