‘Know-that’, like so many natural language expressions, exhibits patterns of use that provide evidence for its context-sensitivity. A popular family of views – call it pragmatic invariantism – attempts to explain the shifty patterns by appeal to a pragmatic thesis: while the semantic meaning of ‘know-that’ is stable across all contexts of use, sentences of the form ‘S knows [doesn’t know] that p’ can be used to communicate a pragmatic content that depends on the context of use. In this paper, the author argues that pragmatic invariantism makes inaccurate predictions for a wide range of well-known use data and is committed to attributing systematic pragmatic error to ordinary speakers. But pragmatic error is unprecedented, and it is doubtful that speakers are systematically wrong about what they intend to communicate.
Contextualism, Skepticism, and Warranted Assertability Maneuvres. 85–104 of: Keim-Campbell J. O’Rourke M. & Silverstein H. (eds.) Knowledge and Skepticism. Cambridge, MA: MIT Press.
See for instance Blome-Tillmann (2014); Cohen (1986 1987 1999); DeRose (1992 1995 2009); Ichikawa (2010); Kompa (2002); Lewis (1996); MacFarlane (2009); Neta (2003); Schaffer (2004a); Schaffer & Szabó (2014).
Brown (2006) and Rysiew (2007) in places liken knowledge claims to cases of what Kent Bach calls ‘implicitures’ (Bach 1994). The differences between Gricean conversational implicatures and Bachian implicitures won’t matter to the features of implicatures I discuss here.
Hazlett (2009) and Pritchard (2010) also also argue that the pragmatics adds to moderate invariantism’s rejection of scepticism while Brown (2006) Lutz (2014) and Locke (2015) do not explicitly argue for mpi’s potential to address sceptical worries. As I point out below there are good reasons to think that if mpi applies to non-sceptical ordinary knowledge claims it applies to sceptical ones as well.
See Kindermann (2013) and Montminy (2009b) on relativism and Kindermann (2012) for the universal claim.