Introduction: signs of the times

Language rests, like an enduring flame, by changing. Speaking, the primordial form of linguistic behavior, is a dynamic phenomenon, because it occurs in time, and solely in time. Yet in normal circumstances we feel we all understand each other without too many difficulties and, what is more, we feel that this understanding occurs more or less instantaneously, without the help of essentially incremental mechanisms of online processing. We feel, in fact, that our very knowledge of the language, both for speakers and for hearers, is not an object of time, strictly speaking, but a relatively stable resource that we can exploit to organize and express relatively stable thoughts or convictions. This (feeling of) tremendous stability is a genuine puzzle. So is the repeated occurrence of rapid, orderly change in linguistic systems. And there is a corollary problem: how could anyone, confronted with the enormous potential for (synchronic) variation exhibited by verbal communication at all conceivable levels of organization, learn any linguistic system at all (Cooper 1999)? At the same time, much of linguistic change, though certainly real and never-ending, also shows itself as fairly inconspicuous to most, or all, naive language users. Even in those domains that call for rather high levels of linguistic awareness, such as (the knowledge of) lexicon, it is often very difficult to keep track of unusual or extraordinary uses of words and of how these should affect (stretch, shrink, etc.) their other meanings, or the meanings of other, related expressions. The paradox of this situation lies in the realization that linguistic change need not rely on any native awareness, but that the explicit analytic awareness needed to reconstruct its workings may suggest — falsely, as it turns out — that the changes at hand are generally mere matters