I’ve been writing a lot about safety accounts of knowledge recently (here and here). It’s time for more.
My concern is that DuncanPritchard’s Safety Account of Knowledge doesn’t easily avoid Kripke’s Fake Barn Country counterexample to Nozick’s Sensitivity Principle. Pritchard is aware that Jonathan Kvanvig has already raised this worry, but Pritchard’s response to Kvanvig seems unsatisfactory.
Let’s recall a version of Nozick’s Sensitivity Principle
Sensitivity Principle
If S knows P, then S does not believe P in the nearest possible world where P is not true.
Now we are to imagine Bob is traveling through Fake Barn Country. Fake Barn Country used to have a lot of regular barns that were a tourist attraction, but they’ve started to fall apart. Instead of building new barns, they build barn facades. Whenever a barn collapses, the citizens of fake barn country put up a fake brown barn in its place. Real barns are always painted red, but the red paint destroys the cheap barn facade material so they cannot make fake red barns. Now consider the following two propositions.
(RB) There is a red barn in the field.
(B) There is a barn in the field.
Bob is sensitive to (RB). In the nearest possible world where (RB) is not true, Bob doesn’t believe it. Because in the nearest possible world where (RB) is not true, there is a brown facade in its place. So, by SP Bob knows (RB). However, Bob is not sensitive to (B). In the nearest possible world where there is not a barn in front of Bob, he still believes that there is because of the facade in it’s place. So, if the Sensitivity Principle is true Bob knows that there is a red barn in the field, but he doesn’t know there is a barn in the field. This is an egregious violation of closure.
Now let’s recall Pritchard’s version of the Safety Principle.
Safety Principle
If S knows (contingent proposition) P, then in nearly all (if not all) nearby possible worlds in which S forms the belief that P in the same way that S forms the belief that P in the actual world, that agent only believes that P when P is true.
Jonathan Kvanvig argues that safety principles fare no better in barn country (Kvanvig 2004). In the original thought experiment above he would argue that (RB) is safe for Bob because in almost all of the nearby worlds Bob believes (RB) when (RB) is true.
However, Pritchard argues that Kvanvig doesn’t really have a version of the safety principle, as Pritchard construes it, in mind. According to Pritchard, Kvanvig has in mind a safety principle that only focuses on nearest possible worlds not nearby worlds. After noting this confusion, Pritchard says,
With this in mind it seems that the agent in this example does not have a safe belief in the target proposition, since in an environment where there is ban deception going on there will be a wide class of nearby possible worlds where, for example, the agent is looking at a [red] barn facade and yet is nevertheless forming a belief that she is looking at a [red] barn (it could be for instance that this is one of the barn facades that the townsfolk haven’t got around to painting [brown] yet). (Pritchard 2005, 186)
It is not, however, that easy. All we need to do is tweak the worlds so that fake red barns are not in nearby worlds. When Richard Feldman spells out Kripkean counterexamples to sensitivity principles he spells out worlds where the red paint ruins fake barns. (Feldman 2002, 89) We could even go so far as to build it into the laws of physics of the world that fake red barns are very difficult to obtain. The counterexample would run as Kripke’s original and Bob would believe in all nearby worlds that there was a red barn only when there was a red barn, but Bob would not believe in all nearby worlds that there was a barn only when there was barn. It’s very difficult to see how Pritchard could maintain that fake red barns are in nearby worlds.
A rather long post, to make a minor point.
References
Feldman, R. (2002). Epistemology. Prentice Hall.
Kvanvig, J. L. (2004). “Nozickian Epistemology and the Value of Knowledge”. Philosophical Issues, 14(1), 201-218.
Pritchard, D. (2005). Epistemic Luck. Oxford University Press.
I don’t believe in barns. But wanted to say I like the new look.
Thanks.
Are you coming to LA?
Justin,
I am. I just sent you an email.