Assertion : the context sensitivity dilemma

It looks as though many philosophers assume that the intuitive variability of proper assertion with practical stakes motivates the following dilemma: either (1) we embrace a knowledge norm of assertion (KNA), and are forced into a view that takes knowledge, or ‘knowledge’ to be sensitive to practical stakes, or (2) we stick to our classical invariantist (CI) guns, but then KNA goes out the window and we get practical sensitivity in the normativity of assertion. Let us dub this The Sensitivity Dilemma. This dissertation aims to bring this implicitly assumed dilemma to centre stage in order to then take a step back. It is argued here that the Sensitivity Dilemma is a false dilemma: a biconditional knowledge norm of assertion, I argue, is perfectly compatible with Classical Invariantism. And, more ambitiously, the dissertation aims to offer independent reason to believe that, if Classical Invariantism and KNA are true, shiftiness in assertability is exactly what we should expect. To this effect, I put forth a functionalist rationale for KNA, in a classical invariantist framework. I argue that not only are the data at hand friendly to CI and KNA, but, if we look at the main epistemic function of assertion, KNA readily follows. I begin by arguing that the Sensitivity Dilemma rests on deontic equivocation. To this effect, Chapter #1draws an important distinction between epistemic norms and mere norms with epistemic content. In the light of this distinction, I argue that a knowledge norm for assertion need not imply context sensitivity of either knowledge/knowledge attribution or proper assertion. Now, say that it turns out that the knowledge norm of assertion, in its biconditional form, is perfectly compatible with the shiftiness data. Does that also mean that KNA is correct? The answer, of course, is ‘no’. After all, empirical adequacy is shared by several of the competing views on the market. We need further reasons to believe KNA is the correct account. Chapter #2 looks at several extant attempts to provide a rationale for KNA, and finds them wanting. In Chapter #3, I offer an alternative answer to the rationale question: assertion, I argue, is governed by a particular epistemic norm in virtue of serving a particular epistemic function. More precisely, according to the proposed account, a biconditional knowledge norm of assertion drops right out of the assertion’s epistemic function of generating testimonial knowledge. Chapters #4 and #5 defend, in turn, the necessity and sufficiency directions of KNA against the classical objections in the literature. I argue that: (1) The necessity claim involved in KNA scores better than weaker norms when it comes to both accommodating linguistic data and explaining how a speaker can be blameless, yet in breach of the norm, and (2) The sufficiency direction of KNA survives the intuitive need for more than knowledge in cases put forth by Jessica Brown and Jennifer Lackey, since the latter is not sourced the epistemic norm governing assertion, but in further norms with epistemic content stepping n and raising the bar. Last but not least, in Chapter #6 I argue that several theoretical virtues, such as simplicity and prior plausibility, favour my functionalist account over extant competing explanations of the shiftiness data