PART TWO
CONDITIONS ON KNOWLEDGE:
CONCLUSIVE REASONS, SENSITIVITY, AND SAFETY
Overview of Part Two
© 2017 Mark McBride, CC BY 4.0 https://doi.org/10.11647/OBP.0104.09
Our Overview of Part Two — in which, by turn, the status of conclusive reasons, sensitivity, and safety as necessary conditions on knowledge is considered — can be refreshingly brief. In large part this is because the bulk of the heavy lifting in setting out the overall narrative of the book has been completed. (The final chapter — looking at an application of safety to the domain of legal proof — is, of necessity, something of an outlier.)
It simply remains to (re)affirm the following. (1) Our Interim Review made clear that a rejection of closure is one — but not the only — means of responding on behalf of the dogmatist to the serious objection to dogmatism considered in Chapter Five. Insofar as conclusive reasons or sensitivity is necessary for knowledge, (K-Closure) fails. (2) Our Introduction outlined, in a very broad-brush manner, Martin Smith’s (2009) proposal to explain (a Part One issue) transmission failure (of knowledge) in terms of the safety condition (a Part Two issue). This proposal is considered in detail in the conclusion. Let us note for now, though, that safety’s status as a necessary condition on knowledge is essential if Smith is to establish his proposal.
Do I then, in the following chapters establish the status of any of the aforementioned conditions as genuinely necessary for knowledge? Not quite. Taking first the closely related conclusive reasons and sensitivity conditions, my aim is simply to make them as plausible as they can be as necessary conditions on knowledge. I explicitly disavow the challenge of rendering them bulletproof to counterexample(s). To the extent that I have rendered them more plausible, I will also have rendered more plausible the (K-Closure)-rejecting (and, possibly, (J-Closure)-rejecting) response of the dogmatist to Chapter Five’s arguments.
The following objection to this as a dogmatist response may be raised at this point: the root idea behind standard dogmatisms is that seemings necessarily constitute prima facie justification for their contents. This is dogmatism’s core. The implication would appear to be that seemings provide this justification even if they aren’t backed by conclusive reasons and even if the resulting belief wouldn’t be sensitive. Therefore, dogmatism about justification and knowledge is incompatible with requiring conclusive reasons or sensitivity. (This objection, and my ensuing reply, carries over, mutatis mutandis, to the safety condition, though safety is, by contrast, preserved across competent deduction.)
My reply is that, while the premise(s) is true, this conclusion doesn’t follow therefrom. To be sure, conclusive reasons or sensitivity as necessary conditions on knowledge is no part of the core of dogmatism; but there is nothing incompatible about augmenting dogmatism therewith. Two routes for securing this compatibility (between which I can remain agnostic) present themselves. First, one might (arguably, taking these conditions at face value) insulate conclusive reasons or sensitivity from any role in justification — that is, from any load-bearing weight in justification. Then, straightforwardly, the core of standard dogmatisms can be preserved, compatibly with conclusive reasons or sensitivity as necessary conditions on knowledge. (On this first route, this dogmatist response will only bear on (K-Closure).) Second, one might, plausibly, give conclusive reasons or sensitivity a role in justification — that is, give it load-bearing weight in justification. Still the core of standard dogmatisms can be preserved: seemings still necessarily constitute prima facie (or pro tanto) justification for their contents, but such justification would fail to be ultima facie (or all-things-considered) should conclusive reasons or sensitivity not be met. (On this second route, this dogmatist response will bear additionally on (J-Closure).)
A final note on Chapters Six and Seven, and the relationship between conclusive reasons and sensitivity. It must be acknowledged up front that these two putative conditions on knowledge are very similar. It is correspondingly no surprise that my initial fixes for the two conditions are also very similar. Several remarks are therefore in order. First, as I note in Chapter Seven, while the two requirements are similar, Dretske (2005: 24, n. 4) himself is at pains to point out situations in which the requirements come apart. I think there is further interesting work to be done exploring this matter. Second, while the initial fix I propose for the two conditions is similar, in Chapter Seven’s treatment of sensitivity I go far beyond that initial fix to explore more wide-ranging modifications of sensitivity. Those more wide-ranging modifications have obvious potential to be carried over to conclusive reasons also. Finally, and most generally, the chapter ordering reflects the chronological sequence in which they were written (reflecting, in turn, a maturation of my thoughts on these conditions). Ideally, within the overall structure of the book, these two chapters should be read as a pair. While general morals are offered in Chapter Six for the future of conclusive reasons as a condition on knowledge, those morals are offered within the comparatively narrow context of the debate between Dretske and Hawthorne. Chapter Seven, meanwhile, while engaging with Hawthorne, operates at a more detached level, and explores sensitivity in abstracto.
Regarding safety, again, I aim to have made it as plausible as it can be as a necessary condition on knowledge, and take myself to have staved off a group of putative counterexamples thereto. Nonetheless, I also concede I cannot be taken to have definitively established (my) safety (condition) as a necessary condition on knowledge. Once again, though, to the extent that I have rendered it more plausible, I will also have rendered more plausible Smith’s explanation of transmission failure (of knowledge) in terms of the safety condition.
Finally, the weakness of my dialectical aims with respect to these putative conditions on knowledge should be emphasised. I’m attempting to defend these conditions as necessary for knowledge. In fact, it is slightly weaker than that still: I’m attempting to make these conditions as plausible as they can be — render them in as plausible a form as possible — as necessary conditions on knowledge. That aim is independently interesting, and also (for reasons already given) has a payoff for upshots of allowing for basic knowledge. My aim is avowedly not to provide an analysis of knowledge with one or more of these conditions forming a component part. In this context, consider a recent fascinating case presented by John Williams and Neil Sinhababu (2015 et al) — the Backward Clock:
You habitually nap between 4 pm and 5 pm. Your method of ascertaining the time you wake is to look at your clock, one you know has always worked perfectly reliably. Unbeknownst to you, your clock is a special model designed by a cult that regards the hour starting from 4 pm today as cursed, and wants clocks not to run forwards during that hour. So your clock is designed to run perfectly reliably backwards during that hour. At 4 pm the hands of the clock jumped to 5 pm, and it has been running reliably backwards since then. This clock is analogue so its hands sweep its face continuously, but it has no second hand so you cannot tell that it is running backwards from a quick glance. Awaking, you look at the clock at exactly 4:30 pm and observe that its hands point to 4:30 pm. Accordingly, you form the belief that it is 4:30 pm.
This innovative case is set up to be one in which the belief formed is sensitive (and presumably one for which one has conclusive reasons) and safe, but yet which is intuitively non-knowledge — the belief is too luckily true. Now, to be sure, the core cases I use as motivation to prompt fixes to conclusive reasons and sensitivity in Chapters Six and Seven are cases in which these conditions are met but which are clearly non-knowledge. Moreover, as I note in Chapter Eight (after stressing my weak dialectical aims), being readily able to invent cases of non-knowledge meeting my proposed safety condition is prima facie troubling; and I go on to consider possible such cases in the ballpark of the core case motivating my proposed fix to safety in that chapter. However, the fact remains: fascinating as the Backward Clock is, it does not dent my limited dialectical ambitions with respect to these conditions.