Skip to main content.
September 10th, 2009

Blome-Tillmann on IRI

George and Ringo both have $6000 in their bank accounts. They both are thinking about buying a new computer, which would cost $2000. Both of them also have rent due tomorrow, and they won’t get any more money before then. George lives in New York, so his rent is $5000. Ringo lives in Syracuse, so his rent is $1000. Clearly, (1) and (2) are true.

(1) Ringo has enough money to buy the computer.
(2) Ringo can afford the computer.

And I think (3) is true as well, though (4) is less clearly true.

(3) George has enough money to buy the computer.
(4) George can afford the computer.

But I want to focus for now on (3). It is a bad idea for George to buy the computer; he won’t be able to pay his rent. But he has enough money to do so; the computer costs $2000, and he has $6000 in the bank. So (3) is true. Admittedly there are things close to (3) that aren’t true. He hasn’t got enough money to buy the computer and pay his rent. You might say that he hasn’t got enough money to buy the computer given his other financial obligations. But none of this undermines (3).

The point of this little story is to respond to an argument Michael Blome-Tillmann makes in a paper attacking interest-relative invariantism (IRI). (He calls IRI ‘SSI’, which I think is unfortunate, since everyone agrees that knowledge is subject-sensitive. No one thinks S knows that p entails T knows that p.) Here is one of the arguments he makes.

Suppose that John and Paul have exactly the same evidence, while John is in a low-stakes situation towards p and Paul in a high-stakes situation towards p. Bearing in mind that SSI is the view that whether one knows p depends on one’s practical situation, SSI entails that one can truly assert:

(11) John and Paul have exactly the same evidence for p, but only John has enough evidence to know p, Paul doesn’t.

And this is meant to be a problem, because (11) is intuitively false.

But SSI doesn’t entail any such thing. Paul does have enough evidence to know that p, just like George has enough money to buy the computer. Paul can’t know that p, just like George can’t buy the computer, because of their practical situations. But that doesn’t mean he doesn’t have enough evidence to know it. So there isn’t a problem for SSI here.

In a footnote attached to this, Blome-Tillmann tries to reformulate the argument.

I take it that having enough evidence to ‘know p‘ in C just means having evidence such that one is in a position to ‘know p‘ in C, rather than having evidence such that one ‘knows p‘. Thus, another way to formulate (11) would be as follows: ‘John and Paul have exactly the same evidence for p, but only John is in a position to know p, Paul isn’t.’

The ‘reformulation’ is obviously bad, since having enough evidence to know p isn’t the same as being in a position to know it, any more than having enough money to buy the computer puts George in a position to buy it. But might there be a different problem for SSI here?

No; the reformulated argument isn’t a problem because the conclusion is not unacceptable. Indeed, the conclusion is a kind of conjunction that is made true all the time, on relatively uncontentious theories of evidence. Consider this example.

Mick and Keith both have evidence E, which is strong inductive evidence for p. And p in fact is true, just as E would suggest. If Keith were to conclude p on the basis of E, that would be knowledge. But Mick has been taking some philosophy classes. And as luck would have it, he has been taking classes from a smart Popperian, who has convinced him that induction is not a source of knowledge. Now if Mick concluded p on the basis of E, this would not be knowledge, because his Popperian beliefs would constitute a doxastic defeater. So Mick and Keith have exactly the same evidence for p, but only Keith is in a position to know p, Mick isn’t.

I think it’s interesting to think about ‘afford’ in this context, since it seems very likely that some kind of IRI analysis of ‘afford’ will be true. We don’t want to have any kind of contextualism about ‘afford’, at least nothing like modern day epistemic contextualism. It would be crazy to say that if my rent is $5000, and it is due tomorrow, then (2) is false, because after all, in my context someone with Ringo’s money couldn’t buy the computer and meet their financial obligations. If I’m right that ‘afford’ is interest-relative, then looking at the way ‘afford’ patterns should provide some useful evidence for or against IRI.

Posted by Brian Weatherson in Uncategorized

32 Comments »

This entry was posted on Thursday, September 10th, 2009 at 10:43 am and is filed under Uncategorized. You can follow any responses to this entry through the comments RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed.

32 Responses to “Blome-Tillmann on IRI”

  1. Juan says:

    Just a quick question about the Mick and Keith example: maybe they both have the same evidence for p, but do they both have the same p-relevant evidence? It seems that Mick has more p-relevant evidence than Keith does (that evidence being that a smart professor says that there is no inductive knowledge and that the only other evidence for p is inductive).

  2. Brian Weatherson says:

    This is a possibly controversial view of mine, but I don’t think this kind of thing (i.e. what philosophers say about the nature of evidence) is really evidence. Or, at least, I don’t think it is p-relevant evidence. The p-relevant evidence is the stuff that really and truly matters to p; the other stuff is stuff Mick (mistakenly) thinks is relevant to p.

    I think this because I draw fairly radical conclusions from Lewis Carroll style examples about regress. Change the example so that E is (p v q) & ~q. And Mick hasn’t been taught by a Popperian, he’s been taught by Graham Priest, who tells him disjunctive syllogism is bad. I still want to say that Mick and Keith have the same evidence, and indeed it is conclusive evidence for p, but Mick can’t know p because he has a defeater.

    I think any alternative view, i.e. any view on which Keith or Mick have more evidence, will lead to a hopeless Carrollian regress.

    Perhaps, though, I could have made the case simpler by leaving out the professors. After all, if Mick just believes, for no good reason, that induction is bunk, or that disjunctive syllogism is invalid, then he’ll still have a doxastic defeater, and he won’t be in a position to know the things that Keith is in a position to know.

  3. fitelson says:

    Neat example, Brian. One point of clarification about the nature of defeaters here. It sounds like Mick “has” an undercutting defeater for p. But, I guess (on your view) there must be a difference between “having” evidence and “having” a defeater? The former being more objective/external (since you claim that they “have” the same evidence, and that evidence is conclusive for p) and the latter being more subjective/internal (since only Mick “has” a defeater)? For instance, can Mick falsely (but justifiably) believe he “has” an undercutting defeater for E/p? I guess my worry is that there might be Carroll worries lurking here for defeaters (if not for evidence). I must be missing something here…

  4. Juan says:

    Brian,

    I don’t see how the Carrollian regress gets started. Suppose that I acquire evidence that all my previous evidence for p is misleading. Would you say that if we think that I have acquired p-relevant evidence then we get a Carrollian regress started?

    Having Mick believe for no good reason that induction is bunk is better in the sense that it is in that case more plausible that he has the same evidence as Keith, but it is worse for your purposes because it is now not so clear (at least to me) that Mick is not in a position to know that p. (I’m not saying that he is in a position to know, just that I find it hard to think about the relevance of our unjustified beliefs on our other beliefs.)

  5. fitelson says:

    Juan — I think our questions are related. If we make it too easy for Mick to “have” an undercutting defeater for E/p, then it becomes less clear why “having” a defeater should be probative for whether Mick knows that p (say, on the basis of E). This is one reason I’m trying to get clearer on what “having” a defeater consists in (and how it differs from “having” evidence).

  6. fitelson says:

    I forgot to add to my last post the following. “On the other hand, if we make it too hard to have a defeater (say, by requiring some sort of strong internalist access to whether something is a defeater or not), then we may run into regress worries about whether something gets to count as a defeater”.

  7. Brian Weatherson says:

    I think there is plenty of evidence you can have without knowing, or even believing, that you have it. And there are plenty of defeaters that you can have without knowing, or even believing, that you have them. It’s just in Mick’s case that he does know what defeaters he has (though he doesn’t recognise them as defeaters) and he doesn’t recognise what evidence he has.

    I also think having evidence requires knowledge. That is, for you to have the evidence that p, you must know p. You need not know that you know p, or know that p is evidence, or know that you have that evidence, but you have to know p. It’s a lot easier to have defeaters; sometimes the fact that someone nearby is out to deceive you could be enough.

    I’m in general hostile to thinking of any second-order evidence as first-order evidence. But before I say more, I want to report a conversation I overheard. I’ve replaced some of the facts with variables to protect certain people’s privacy.

    Harpo: p.
    Groucho: Why do you think p?
    Harpo: E.
    Groucho: Ah, that’s interesting. Is E evidence for p?
    Harpo: Yes it is.
    Chico: I don’t think E is evidence for p.
    Groucho: Shut up Chico, of course E is evidence for p. Let’s let Harpo add that to his reasons.
    Zeppo: Ooh look Harpo, now you’ve got two pieces of evidence for p.
    Groucho: Shut up Zeppo.
    Harpo: But that’s right. I have E, and E is evidence for p. Now do you believe p?
    Groucho: That still feels like a bit of a leap. I agree with your evidence, but something’s nagging at me.
    Harpo: But E, and E is evidence for p is excellent evidence for p.
    Chico: I don’t think E, and E is evidence for p is evidence for p.
    Groucho: Shut up Chico, of course E, and E is evidence for p is evidence for p. Let’s let Harpo add that to his reasons.
    Zeppo: Ooh look Harpo, now you’ve got three pieces of evidence for p.
    Groucho: Shut up Zeppo.
    Harpo: But that’s right. I have E, and E is evidence for p, and E, and E is evidence for p is evidence for p. Now do you believe p?
    Groucho: That still feels like a bit of a leap. I agree with your evidence, but something’s nagging at me.
    Harpo: But E, and E is evidence for p, and E, and E is evidence for p is evidence for p is excellent evidence for p.
    Chico: I don’t think E, and E is evidence for p, and E, and E is evidence for p is evidence for p is evidence for p.
    Groucho: Shut up Chico. Of course E, and E is evidence for p, and E, and E is evidence for p is evidence for p is evidence for p. Let’s let Harpo add that to his reasons.
    Zeppo: Ooh look Harpo, now you’ve got four pieces of evidence for p
    Groucho: Shut up Zeppo.

    And it went on like that for a while longer. I think Groucho leads them off on the wrong track very early on. E is evidence for p isn’t an extra piece of evidence that Harpo gets for p. He had all the evidence for p he needed at the start. Sometimes the connection between evidence and conclusion is not mediated by extra evidence. In those cases, getting extra evidence for or against the mediation isn’t actually evidence about the underlying proposition.

  8. Juan says:

    Sorry, but I’m still not seeing it. I don’t think I’ve said anything that commits me to the claim that “E is evidence for p” is an extra piece of evidence that Harpo gets for p. (Maybe he does get extra evidence for p, but that would be because he learns that Groucho thinks that E is evidence for p, and (depending on the details) this may actually be evidence for p. But I don’t think this leads to a regress at all.)

    In any case, how does this relate to my question whether I have acquired p-related evidence when I acquire evidence that my previous evidence for p is misleading? Do you want to say that I don’t? Never?

  9. Brian Weatherson says:

    In cases where E is basic evidence for p, I want to say never.

    It’s trickier if you learn, say, that a particular measuring device is reliable. It’s not clear whether you get more evidence, or you get evidence that your earlier evidence was better than you thought. But in the cases of basic evidence, I think evidence for or against their reliability is no evidence whatsoever for or against p.

  10. fitelson says:

    OK, I think I’ve got it. So, a defeater is just second-order evidence against the first-order support claim (so they don’t really have the same evidence, full stop)? And, then, you think that this second-order countersupport claim never collapses to a first-order support claim. Is that it?

  11. fitelson says:

    Clarification. Why not just say that they have different second-order evidence, but the same first-order evidence regarding p?

  12. fitelson says:

    In other words, how is “defeater” distinct from “second-order evidence that’s negatively relevant to a first-order evidential support claim”? I’m still not clear on that.

  13. Jeremy Fantl says:

    Brian,

    Here’s a potential difference between Paul and George. You can hold everything else about Paul fixed while increasing his evidence for p, and he will end up knowing that p (despite his high stakes). That is, you can make him know by increasing his evidence. You can’t make it financially possible for George to buy the computer by increasing his money in the bank (it’s already financially possible for him to buy the computer, so you can’t change whether it is possible by increasing his money). So, there seems a clear sense in which George has money enough for him to be able to buy the computer, but Paul does not have evidence enough for him to know that p.

    -Jeremy

  14. Brian Weatherson says:

    Branden,

    Yep, it’s basically just that I think second-order evidence doesn’t collapse into first-order evidence. (I think it never collapses when it is second-order evidence about the status of basic methods, and think it doesn’t collapse that often when it concerns less basic methods.)

    But I don’t think every defeater concerns second-order evidence. That’s why I wanted to use the more general language of defeaters here. I think almost any kind of evidence will generate a Mick/Keith style example.

  15. Brian Weatherson says:

    Jeremy,

    I don’t quite see the disanalogy. It seems to me all the following things are true.

    George has enough money to buy the computer.
    Paul has enough evidence to know that p.

    George isn’t in a position to buy the computer, but with more money he would be.
    Paul isn’t in a position to know that p, but with more evidence (or for that matter more money) he would be.

    George can’t afford the computer.
    Paul doesn’t know that p.

    The analogy there seems fairly tight to me.

  16. Jeremy Fantl says:

    I’m questioning whether Paul has enough evidence to know that p in every relevant sense in which George has enough money to buy the computer. One thing standing in the way of Paul knowing is the high stakes. Another is the evidence-shortage. How do we know that the evidence shortage is one thing standing in the way? Because appropriately raising the evidence and holding everything else fixed makes Paul know. Assuming that there is a sense of “having enough evidence to know” that is equivalent to “one’s evidence-shortage not standing in the way of one’s knowing” there is a sense in which Paul doesn’t have evidence enough for him to know that p.

    In that same sense, George has money enough to be able to buy the computer. He also has enough money to buy the computer. Increase his money and hold everything else fixed and it does not all-of-a-sudden become true that George buys the computer (or that he can buy it).

  17. fitelson says:

    Thanks, Brian. I think I’m getting it now. That’s helpful. Sorry for clogging things with so many posts [I shouldn’t try to blog from my phone — I’m still just getting the hang of this blogging thing in the first place!]

  18. jonathanweinberg says:

    Brian, since it seems that your discussion turns fairly centrally on some particular aspects of your views on evidence, I wonder if maybe there should be, from where you sit, more charitable ways of understanding Blome-Tillman’s (11)? I haven’t read the paper, but if he isn’t thinking of evidence in quite the same way you are, then maybe some factors that he is counting as evidential are ones that you are not. And those other factors that they count as evidential, but you don’t, should get packaged into (11). That is, something like the following might be a fair reinterpretation:

    (11*) John and Paul have exactly the same evidence for p, and exactly all the same higher-order evidence and belief-states as well, but only John knows (or is in a position to know, or whatever) p, and Paul doesn’t (or isn’t, or whatever).

    I don’t have any intuitions as to whether IRI should be committed to this (for your “affords”-like reasons), and none of these principles strike me as either intuitive or counter-intuitive. But I think it may be immune to your Mick/Keith-type cases. (Say, does using Mick/Keith and John/Paul counterexamples deployed against one count as being “Rolling Stoned or Beatled til you’re blind”?)

  19. Brian Weatherson says:

    Jeremy,

    I don’t see any sense in which Paul doesn’t have enough evidence to know in which George also doesn’t have enough money to buy the computer. Paul can’t know p given his practical interests, though the amount of evidence he has is sufficient, in principle, to know p. George can’t buy the computer given his practical interests, though the amount of money he has is sufficient, in principle, to buy the computer.

    If we increase George’s money to $7000 it does become all of a sudden true that he can buy the computer. In that sense of ‘enough’, if we increase his money to $7000 it does become all of a sudden true that he has enough money to buy the computer.

  20. Brian Weatherson says:

    Jonathan,

    I think to get a version of (11) that is immune to independent counterexamples, you’ll have to add so many qualifications and clarifications of the type you suggest that you’ll end up making it a fairly explicit denial of IRI. Maybe MBT wants to say that straight-up denials of IRI are so intuitively true that we should reject IRI. (Compare Moore on scepticism!) But that’s not the way his argument works.

    I should stress that my quirky views on second-order evidence are not necessary to get the argument going against the revised version of (11). Anyone who thinks that there are non-evidential defeaters will think there’s nothing wrong with MBT style conjunctions. For instance, here’s a version based on a variant of Harman’s dead dictator example.

    Lou and Andy both get evidence E, which is strong evidence for p. Nothing in the Lou’s situation is unusual; if he were to infer p from E he would come to know p. But Andy has just missed a phone call from a friend who would tell him (falsely) that ~p. The friend left a voicemail saying ~p. Andy trusts the friend, and with good reason, since the friend is very reliable. If Andy were to infer p from E, that wouldn’t be knowledge, since the missed phone call and not yet retrieved voicemail is a Harman-style defeater. But since Andy hasn’t yet got either the call or the voicemail, he has the same evidence as Lou. So Lou and Andy have the same evidence, but Lou is in a position to know p, and Andy isn’t.

    Not everyone agrees with Harman’s dictator case, like not everyone agrees with me about second-order evidence. But there are tons of cases like this, and thinking that what you are in a position to know depends only on what evidence you have requires rejecting all of them. Indeed, it requires having a very very narrow conception of what the grounds for knowledge are. (I think it basically requires a JTB theory of knowledge.) So while any particular objection to this argument of MBT’s might be controversial, I don’t think it should be controversial that the argument fails.

  21. Jeremy Fantl says:

    If we increase George’s money to $7000, it is not all-of-a-sudden true that he can buy the computer. He already can. It’s not appropriate or wise for him to. But it’s possible. So, George’s lack of money, while it stands in the way of it being appropriate for him to buy the computer, does not stand in the way of it being possible for him to buy the computer. He has money enough for it to be possible for him to buy the computer.

    If we increase Paul’s evidence sufficiently, it is all-of-a-sudden true that he knows. So, in that sense, he does not now have evidence enough to know.

  22. Brian Weatherson says:

    That’s a difference I guess, but I don’t see why it should be relevant. The use of ‘can’ there is a purely ability based ‘can’; it isn’t normative. Since ‘knows’ means (among other things) ‘properly believes’, and George can’t properly buy the computer, his situation seems parallel to Paul’s, who can’t properly believe that p.

  23. Jeremy Fantl says:

    Hi Brian,

    It’s relevant because we can interpret Blome-Tillman’s (11) in that sense:

    John and Paul have exactly the same evidence for p, but Paul’s evidence-shortage stands in the way of him knowing, while John’s evidence-shortage doesn’t stand in the way of him knowing.

    Interpreted this way, 11 seems to me counterintuitive (though not, I obviously think, unacceptable). It’s also not obvious to me whether you’ll want to say that Mick’s evidence-shortage is standing in the way of him knowing. But I’m not sure about that.

  24. Jeremy Fantl says:

    One more point:

    Even if Mick’s defeater is not p-relevant evidence, it still seems to bear on Mick’s strength of epistemic position with respect to p. So, we can modify my modified Blome-Tillman’s (11) accordingly:

    John and Paul have exactly the same strength of epistemic position with respect to p, but Paul’s epistemic weaknesses with respect to p stand in the way of him knowing, while John’s epistemic weaknesses with respect to p don’t stand in the way of him knowing.

    This also seems entailed by IRI. It also seems counterintuitive. And it even less clearly is a “kind of conjunction that is made true all the time.”

  25. Brian Weatherson says:

    John and Paul have exactly the same evidence for p, but Paul’s evidence-shortage stands in the way of him knowing, while John’s evidence-shortage doesn’t stand in the way of him knowing.

    True, but not counterintuitive. Compare,

    George and Ringo have exactly the same amount of money, but George’s money-shortage stands in the way of him (properly) buying the computer, while Ringo’s money-shortage does not stand in the way of him (properly) buying the computer.

    And I’m really not sure that all defeaters affect “strength of epistemic position”.

    If a student knows ~~p, and then attends a lecture of mine where they learn about why there are reasons to be sceptical of double negation elimination, they might no longer be in a position to know p because my lecture would be a defeater.

    Compare them to someone who doesn’t have the concept of entailment clearly worked out in their mind, but happily infers p from ~~p.

    The first student is, I think, in at least as good an epistemic position with respect to p. It’s not that there’s something relevant the second student knows – after all, they don’t have a clear concept of entailment, so they don’t clearly know that ~~p entails p. It’s just they’re in a position to infer p knowledgably from ~~p, because of the absence of defeaters.

    Here’s another way to put the point.

    If “strength of epistemic position” means “alike in every knowledge-relevant respect”, then John and Paul don’t have just as strong an epistemic position. Paul’s interests weaken his epistemic position.

    If “strength of epistemic position” means “alike in every knowledge-relevant respect except for interests” then they are alike in epistemic position, but now MBT’s conjunction is simply a statement of anti-IRI, not a premise in an argument against IRI.

    If “strength of epistemic position” means “alike in evidence and defeaters”, then I’m back to denying that Paul and John are alike in strength of epistemic position, since I think Paul’s interests are a practical defeater.

    What else could “strength of epistemic position” mean? It’s not a folk notion after all, and I don’t even see a stipulative definition of it that helps MBT’s argument.

  26. jonathanweinberg says:

    “Maybe MBT wants to say that straight-up denials of IRI are so intuitively true that we should reject IRI.” Looking now at that section of the paper, that strikes me as a not-implausible take on what’s going on there. One might be able to duck around the “that’s simply a statement of anti-IRI” objection by claiming that looking at IRI in these terms really brings out its falsity. (In part because, as he does argue, there’s no room in these cases for us to be led astray by the subtleties of modal operators, etc.)

    There is an interesting methodological question in the neighborhood: if one is engaging in this sort of philosophy, what sorts of moves are available when someone has argued for p, and you just think that their argument for p can’t rise above p’s own inherent counter-intuitivity? It may be very hard to avoid this danger of either looking like you’re just begging the question against the p-arguer, or, for that matter, of actually begging the question against them.

  27. Jeremy Fantl says:

    Brian,

    I find principles are counterintuitive for knowledge but not for proper buying. I don’t have an argument for that. But I do think the proper buying (and similar examples) suffice to show that the knowledge-principles aren’t incoherent or unacceptable.

    More importantly, you said

    ‘If “strength of epistemic position” means “alike in every knowledge-relevant respect except for interests” then they are alike in epistemic position, but now MBT’s conjunction is simply a statement of anti-IRI, not a premise in an argument against IRI.’

    This is EXACTLY the point I made when I refereed this paper. (Am I allowed to say that online?) Blome-Tillman’s response is in footnote 27. And Jonathan’s comment @26 is ultra-relevant. Here’s a possible result: the straight-up counterintuitiveness of IRI is sufficient to make simple appeals to cases inadequate for a defense of IRI. But that leaves open whether more principled defenses can succeed.

  28. Brian Weatherson says:

    Jeremy,

    I’m inclined to agree that IRI is a little counterintuitive when applied to cases. And I think that tells against a case-based defence of it. So I’m antecedently disposed to think the principle-based defence is better.

    The difference between a case-based defence and a principle-based defence is a matter of degree to a large extent. But I think there’s a good sense in which the arguments that Hawthorne gives, and which I give, rely less on intuitions about cases than the arguments which Jason gives, and which you and Matt give. So maybe this is just an argument in favour of doing things the way Hawthorne and I do!

    Jonathan,

    This kind of issue comes up a lot in philosophy of logic. That’s not too surprising, because once you start debating what the right logic is, what moves are and aren’t allowed is a little tricky. I think some of the best arguments in those fields are question-begging. But I would put two constraints on question-begging arguments.

    1) It might be rational to believe ~p simply because p is so counterintuitive, even if p solves various problems. But that doesn’t mean one should publish papers simply saying p is really counterintuitive!

    2) Relatedly, if one wants to argue against a proposition p that is supported despite being counterintuitive, it is good to look at the reasons people have for believing the counterintuitive thing, and seeing where they fail. (See, for instance, David Lewis on how ex falso isn’t a relevance-failure in the most important sense of relevance, or Stephen Read on how the classical inference rules don’t violate the best formulation of harmony principles.) That’s one of the annoying things about this paper. There’s a straight-up insistence that IRI is counterintuitive, without any acknowledgment of the puzzles that some of us (esp Hawthorne and me) have argued IRI is needed to solve.

  29. Jeremy Fantl says:

    Brian,

    I think we can agree on almost everything, then, with one exception being your claim that Matt and I rely on intuitions about cases. I’m pretty sure we explicitly deny that such reliance is appropriate.

  30. Brian Weatherson says:

    My bad. I was misremembering your papers. I’m sorry for that.

  31. Juan says:

    I’m a big fan of Jeremy and Matt’s (“Jeremy’s and Matt’s”?) papers on this topic (which doesn’t mean, of course, that I agree with everything they say!). But I think that Brian is right on when he says that “[t]he difference between a case-based defence and a principle-based defence is a matter of degree to a large extent.” I think that J&M’s knowledge-to-action principle is appealing precisely because (and to the extent that) it reflects our judgments about a number of cases (how I would defend myself from not stopping at the supermarket to buy yams before thanksgiving, etc.)

  32. Jeremy Fantl says:

    Thanks, Juan. The worrisome cases — the ones that at best leave the interest-relative invariantist in a tie with the moderate invariantist and the contextualist — are the stakes-shifting cases in which we’re supposed to, by changing the stakes on a subject, vary whether the subject is intuitively said to know. I’m more happy about other sorts of intuitive evidence (e.g. about the linguistic data you mention, but especially about various principles — like if you know it’ll have the best results, you’re rational to do it; and if you know there is conclusive reason to do it, then you’re justified in doing it).

Leave a Reply

You must be logged in to post a comment.