"The 'NP' is one of the least controversial grammatical units that linguists work with. The NP is often assumed to be universal, and appears to be robust cross-linguistically (compared to 'VP' or even 'clause') in that it can be manipulated in argument positions in constructed examples. Furthermore, for any given language, its internal structure (order and type of modifiers) tends to be relatively fixed. Surprisingly, however, the empirical basis for 'NP' has never been established. The chapters in this volume examine the NP in ordinary interactions from diverse languages, including little-studied languages as well as better-researched ones, in a variety of interactional settings. Together, these chapters show that cross-linguistically, the category NP is not as robust as has been assumed: it is realized only in temporally unfolding human interaction, its structural status thus constantly being negotiated in terms of participants' social agendas"-- |