Consider the following two cases
Guard Dog Case
Bob’s mom is worried about burglars. She tells him she needs a dog (but not that it’s to scare away burglars). Bob goes out and gets her a tea cup poodle. She laughs at him and says, “You don’t understand. I’m worried about burglars. That’s not a dog. I need a dog.”
Contractor Case
Bob’s dad wants to have a second story balcony built, and asks Bob to go get a contractor. Bob goes out and finds a handyman who does some freelance contracting on the side. Bob’s dad says, “My grandchildren are going to play on this patio. He’s not a contractor. I need a contractor”
These are examples where (a) someone wants an X for some purpose, (b) the stakes are high enough that the X they get is not a suitably good X for their purposes, and (c) they speak loosely and appropriately as though the thing they are given is not an X.
What are we to make of these examples?
Option One: We could say that these examples force us to radically revise our theories about dogs and contractors. We go subject sensitive invariantist about words like ‘dog’ and ‘contractor’ and say that sometimes the practical stakes are so high that certain things that qualify as dogs and contractors under normal circimstances do not qualify as dogs and contractors under high stakes circumstances. Seems like a terrible move.
Option Two: Sometimes people use words loosely. Sometimes they are interested in something that a dog or a contractor is generally good for their purposes, but when they stakes are high they want a dog or a contractor that has the good making feature they’re interested in to a really high degree. Saying “No, I want a dog” is a way of saying. “The dog you’ve given me doesn’t have the property that dogs have to a high enough degree.”
I go for option two. And I suspect we now have an interesting way to argue against one of the main motivations for subject sensitive invariantism about knowledge.
Stanley’s High Stakes/Low Stakes Cases
For those of you who are unfamiliar with some of Stanley’s cases, I’ll lay two out in this section. If you’re already familiar with Stanley’s cases you can skip down to “A Possible Reply to Stanley”
Here are two of Stanley’s cases.
Low Stakes
Hannah and her wife Sarah are driving home on a Friday afternoon. They plan to stop at the bank on the way home to deposit their paychecks. It is not important that they do so, as they have no impending bills. But as they drive past the bank, they notice that the lines are very long, as they often are on Friday afternoon. Realizing that it isn’t very important that there paychecks are deposited right away, Hannah says, “I know the bank will be open tomorrow, since I was there just two weeks ago on Saturday morning. So we can deposit our paychecks tomorrow morning.
High Stakes
Hannah and her wife Sarah are driving home on a Friday afternoon. They plan to stop at the bank on the way home to deposit their paychecks. Since they have an impending bill coming due, and very little in their account, it is very important that they deposit their paychecks by Saturday. Hannah notes that she was at the bank two weeks before on a Saturday morning, and it was open. But, as Sarah points out, banks do change their business hours. Hannah says, “I guess you’re right. I don’t know the bank will be open tomorrow”
Stanley claims that the common intuitive reactions to these cases are that Hannah is right in both. When she says she knows the bank will be open in low stakes, she’s right. When she says she doesn’t know in high stakes she’s also right.
Subject sensitive invariantists argue that a tight connection between knowledge and rational action captures both intuitions. The only thing that has changed is the importance that Hannah be right about whether or not the bank is open. Her practical investment in that proposition is much higher in high stakes. Since that’s the only thing that has really changed between the two cases, it must be that practical stakes can trump knowledge.
A Possible Reply to Stanley
Given that there is a linguistic convention of speaking as if something that is an X is not an X when the stakes are high (e.g. Guard Dog Case and Contractor Case), it seems like we can explain the apparent appropriateness of Hannah’s assertion in a similar manner. There is a property that comes with knowledge that we like to have when deciding how to act. Let’s call it evidence.
When the stakes are high we want more of this thing (evidence) that generally comes with knowledge, and so we sometimes loosely speak as if a belief is not knowledge when in fact it is (it just lacks the super high degree of evidential support we want given the practical stakes)
I have an objection Stanley could raise, but I’m going to mull all of this over for a bit. The objection will probably be the topic of my next post.
(Posted from my Android Phone)
Hi Andrew,
First time commenter (commentor? commentator?), long-time occasional reader. (Incidentally, I’m also an Ubuntu user and a some-time grad student in philosophy at the University of Rochester–in my case, the “some time” overlaps with the current time–but that’s a mere aside for now.)
I like the cases that you’ve presented, and I also like the way you’ve put them to use. Concerning the “Guard Dog Case” and the “Contractor Case,” I agree with your description of them in the paragraph beginning, “These are examples . . . “, and I also favor your “Option Two.” I also like the way that you’ve used these cases as part of a reply to SSI-ists. So geez, with all of this liking and agreeing, why the heck am I even commenting?
Well, here’s why. This post in particular made me think of Earl( Conee)’s contribution to the debate concerning epistemic contextualism in the Contemporary Debates in Epistemology volume (edited by Sosa and Steup). Putting it roughly (and attempting to recall all of this off the top of my head), Earl suggests there that ordinary talk employing “knowledge” is loose talk by which the speaker attempts to convey something like “trustworthiness on the subject for all relevant intents and purposes,” and that cases like those presented by contextualists and SSI-ists are cases wherein the knowledge-attributor (or knowledge-denier) “gets serious,” sets aside his childish things, and undertakes the task of more carefully talking about knowledge. I’m hereby noting (because I find it interesting) that your view seems to go in exactly the opposite direction–on your view, the talk in the SSI-ist-type cases is “loose” (as you put it, “there is a linguistic convention of speaking as if something that is an X is not an X when the stakes are high”), and the talk in “ordinary” cases is spot-on about knowledge.
I just wonder if you have any thoughts/comments about this. My initial reaction was to think that your view is more plausible, given the evidence of a widespread linguistic convention of speaking in the way you indicate (I can’t help but think of Crocodile Dundee here: “You call that a knife? This is a knife!”). But then I also begin to think that there are analogs of Earl’s view, too (I say, “Jon’s a genius!”, a friend responds, “Well, his IQ is actually 135 [or whatever], so technically he’s not,” and I say, “Yeah, yeah, I know, but you know what I mean”). So, is there some way to tip the scales one way or the other? I think there may be–for example, when the dog-, contract-, and knife-talk happens, it’s obvious to me (and to other competent hearers) that the speaker is engaging in the linguistic convention you mention, and we all readily (I think) endorse that view when we hear it mentioned; not so with knowledge-talk, and in fact I think we’re generally more ready to endorse Earl’s view here–but I worry (for you) that the scales tip in Earl’s direction.
Jason
Jason,
That is VERY interesting. I hadn’t thought about the asymmetry like that.
And it would certainly augment the reply I’m worried about from my most recent post.
I’ve been jotting down thoughts about that reply this morning, and I’ll have to incorporate this now. I’ll probably post it here by the end of the day. Thanks for bringing this up.
p.s. I hope things are going well for you at Rochester.
p.p.s Have you upgraded to 9.04 (Jaunty Jackelope)…it’s awesome
Andrew,
Thanks. I’m looking forward to your next post!
After I wrote my comment here, I noticed that you mentioned (in the post following this one) a possible reply that an SSI-ist might have to your suggestions. I’ve been thinking about it a bit since then. I do think there are some similarities between that reply and the sort of thing I mentioned in my comment, but I also think there’s at least one important difference: the (Earl-ish) view that I mentioned does not hold (in tandem with the SSI-ist) that the relevant ordinary-context knowledge-attributions are, strictly speaking, true (though they may convey truths). In other words, this view won’t help the SSI-ist; it’s subject insensitive. It’s still a (fallibilist, I emphasize!) view that’s at odds with your own, though.
Oh, and things are going quite well for me in Rochester! It’s a great place to be, as you know. And yes, I have upgraded to Jaunty (I did so on the day it was released). My review of it is a bit more mixed than yours; it is awesome in many respects (boot-up time, particularly), but my two computers have an Intel graphics card and an ATI Radeon Xpress graphics card, respectively. There were serious issues with both of these cards (particularly the Intel) in Jaunty because of the new X server. Fortunately, I resolved the issues–but I do worry about the more casual users. In any case, I’m now quite happy with Jaunty (and given that this was an X server issue, Ubuntu development itself wasn’t centrally to blame anyway); it’s the only OS I’m currently using .
I see your point regarding the differences. Still thinking about a potential reply. Thanks for these comments.