This conference will examine the problems of pervasive context-sensitivity in natural language, from both a philosophical and a psychological perspective. Some indicative questions which may be explored at the conference include: What exactly is 'the problem of pervasive context-sensitivity'? Is there a single problem to be studied here or are there different phenomena in play across different cases? Similarly, is there one mechanism which should handle all instances of context-sensitivity or do natural languages contain different ways of accommodating different types of context-sensitivity? Are the kinds of thought experiments Charles Travis has famously used in arguing for pervasive context-sensitivity robust? In light of recent scepticism about the role of intuitions as data for philosophical theories, are Travis-cases open to challenge in this regard? Are there intercultural differences in intuitions about Travis-cases (of the sort that have been argued for in experimental work on reference; see Machery et al. 2004, Nisbett et al.2001)? In what ways do contemporary theories of context-sensitivity agree or disagree with each other? Are there any surprising points of consensus? What exactly are the different theories disagreeing about?