Is sympathetic atheism incoherent?

Hi Ted,

(P2) For any subject S if S knows (or justifiedly believes) that there are gratuitous evils then any close world with less evil contains gratuitous evils.

I'm pretty sure P2 is false. Let amount of evil E be gratuitous and small and let N be non-gratuitous evil and large. The fact that E is a small amount of evil does not entail that I do not know state N + E is gratuitous and N - E is not. The fact that P is a small decrement in pain might entail that I do not know that P* + P is painful and P* - P is not, since I am not so sensitive. Similarly for a small decrement in observable evil e: since my perception is not that sensitive, I might not know that N + e is perceptibly more than necessary and N - e is not. But knowing that E is gratuitous does not require any special sensitivity on my part. All I have to know is that E is pointless, however small E happens to be. If I could know that the small amount E is gratuitous, then take away that small amount E, and there remains non-gratuitous evil N.

Hi Ted,

The difficulty here is that 'close' is vague. To keep (P2) plausible, 'close' has to be restricted to a fairly narrow scope. But on a narrow reading of 'close' it becomes far from obvious that (P4) is incompatible with (iii).

Hi Mike,
Thanks for this. I was assuming that one knows that there are gratuitous evils only by an aggregative judgment—roughly, that there are too many bad things and not enough good things. Your example seems to ride on some evils being such that there can’t be goods that offset that evil. I tend to think that every evil can be redeemed, but I’ll definitely need to discuss this in the paper.

Hi Alan,
You’re right, of course, that ‘close’ is vague. (P2) is to be read as a way consistent with the margin of error principle. Read that way, worlds that don’t contain gratuitous evils are worlds containing non-negligible differences from the actual world. I think this makes (P4) true. Think about it this way. Given the margin of error principle and the assumption that the sympathetic atheist knows or justifiedly believes that there are gratuitous evils, worlds in which there are not gratuitous evils are worlds in which the evidence is not-insignificantly different from the actual evidence. Since the sympathetic atheist can notice not-insignificant changes in evidence the sympathetic atheist will think that the world lacks the property of containing gratuitous evils.

I'm curious about the concept of sympathetic atheism itself. Is there anyone who discusses it by that name, and has it also gone by other names?

Does the sympathetic atheist take (iii) to mean that the world would be relatively similar, but without gratuitous evil? So, in all possible worlds where God exists, there is no gratuitous evil? I think she would have to, but it's not made explicit, so I'm just wondering.

David,

From what I recall Rowe’s “friendly atheism” is similar to sympathetic atheism, where the similarity condition (iii) is understood in terms of evidential similarity.

Bradley,

That’s right. Basically the idea is that what makes sympathetic atheism *sympathetic* is condition (iii) but what makes it atheism is condition (ii). The basic problem is that the reason given in condition (ii) obeys a margin of error principle which conflicts with the condition (iii).

Hey Ted,

Let me say one other word on behalf of the SAists (though I am not an official representative). Why wouldn't a defender of SAism read (iii) epistemologically rather than metaphysically? Suppose your argument is sound and none of the most similar worlds to @ is a world without gratuitous evil. That's of course a metaphysical claim. Defenders of SAism should respond that, if (iii) is read epistemologically, then there need not be any similar worlds in which there is no gratuitous evil. (iii) requires only epistemological similarity. So were there a God, then the world would not appear all that different vis a vis the extant evil. But it might be metaphysically very different vis a vis that evil, since all of that evil would have to be necessary for some greater good. The fact that all that evil is necessary for some greater good does not entail that we would observe any connection between the good and evil. It might well be beyond our ken, as the skeptical theists are wont to say. Further, I do think that (iii) is most plausibly read epistemically.

Mike,

I agree with you that the way out of the argument I give is to opt for epistemic similarity. But then I’ve got another argument that that way is incoherent as well. The main idea is that the sympathetic atheist should think the evidential theist is an epistemic peer. Since epistemic peers have similar evidence and competently handle the evidence, the sympathetic atheist’s evidence violates the margin of error principle. It might be thought that the sympathetic atheist could deny the evidential theist is an epistemic peer but this surrenders the sympathy part of the sympathetic atheism.

The main idea is that the sympathetic atheist should think the evidential theist is an epistemic peer.

Ted, nice suggestion. But certainly epistemic peers can disagree over how to weigh evidence, no? I think this is exactly Bill Rowe's line. That, as he weighs the evidence, it favors atheism, but there are alternative and reasonable ways to weigh evidence on which theism comes out on top. You can coherently say that there are other reasonable ways to weigh evidence on which theism is favored without being incoherent, I think. Is this something you're denying? Just as, say, coherentists about justification can concede that foundationalists are not irrational without being incoherent. No?

Mike, Maybe an analogy will help. Suppose you and I are counting beans in a coffee bag. On your count there are 1011 beans; on my count there are 1021 beans. We can both hold the other’s count is reasonable, but I don’t think we can hold that we are justified in believing that our count is accurate. Since, assuming we’re epistemic peers apropos coffee beans, we each possess a defeater that our count is right (in addition to the defeater that it’s always hard to make aggregative judgments). A similar point holds with sympathetic atheism. Rowe’s right that there are reasonable and differing ways to tally the amount of good (or evil) in the world. But given that those that tally are epistemic peers (at least by the sympathetic atheist’s light) it undercuts the justification for believing that their tally is correct.

Ted, maybe you got this example from Christensen? His example is adding up a total on a bill, but the idea is the same. Rich Feldman has been defending a similar idea too.

But of course a single example can't settle the issue, since the equal weight view is self-defeating in a certain way. You know that there are peers of yours who disagree (e.g., Tom Kelly, Ralph Wedgwood, Dick Foley, Peter van Inwagen, etc.), so you shouldn't hold onto the view.

. . . Suppose you and I are counting beans in a coffee bag. On your count there are 1011 beans; on my count there are 1021 beans. We can both hold the other’s count is reasonable, but I don’t think we can hold that we are justified in believing that our count is accurate.

We have a fairly simple case here, but even then I don't obviously have a defeater. Suppose I did count the beans, and indeed twice. If you count differently, I have to assume that you've been as careful as I have been on this occasion, that you are a serious about you bean-counting as I am on this occasion, that you were not distracted on this occasion, etc., etc.. Otherwise, I have no defeater.
In the case of gratuitous evil, things are obviously much more complicated. There are so many psychological reasons(indeed, many rise to the level of theories) that explain why someone would succumb to theism, misread evidence to favor theism, or wrongly weigh evidence in favor of theism, that I can--now playing the role of Sympathetic Atheist--see how my epistemic peer got things wrong. People whom I concede are my epistemic peers need not be my psychological peers. Nor need they be my sociological peers, where sociological explanations work to make understandable the theist's tendency to skew evidence in their favor.
Here I am taking the line that I need not maintain that Smith is not my epistemic peer to explain Smith's misreading or mis-weighing of the evidence. I can assume that he is epistemically every bit as good as I am, but he is not my psychological peer or not my sociological peer or....[pick your favorite relevant perspective] peer. These are what explain his misreading.

Jon, Thanks for mentioning the Christensen and Feldman references. I’ll have to go find those articles again. The self-defeat issue is really interesting. In the end I’ll probably say the same thing Unger says in response to the claim that skepticism is self-defeating. Also, the self-defeat issue is endemic to lots of philosophical views—Williamson asserts that knowledge is the norm of assertion but he doesn’t know that. This isn’t a justification; just an exculpation.

Mike, I don’t think you can draw the distinction you want to. If there are psychological features that bias an individual’s judgment then that individual isn’t an epistemic peer.

I don’t think you can draw the distinction you want to. If there are psychological features that bias an individual’s judgment then that individual isn’t an epistemic peer.

Hi Ted,
Yes, this is sort of what I expected. Here are some examples.

Jones is every bit as good a chess player as I am, but he suffers a bit from a agoraphobia. The agoraphobia does not make him a less skillful player--it does not cause him to know less about chess or to understand less about chess or to be less insightful--but it does interfere with the exercise of those skills. He is still my chess peer. Similarly, a temporary loss of consciousness would not make him any less my chess peer: he still has all the skill I have, though for the moment he cannot exercise it.

Smith is every bit as good a musician as I am. But he fears crowds. The fear of crowds does not make him less skilled as a musician. But it does inhibit the exercise of those skills. Smith is still my musical peer.

Sue is every bit as insightful epistemically as I am. She knows as much, she draws relevant inferences at least as well and imagines counterexamples at least as skillfully. But she suffers from acousticophobia. Noisey places inhibit the exercise of those skills. The fear of noisey places does not make her less epistemically skillful. They rather inhibit the exercise of those skills. She is my epistemic peer, with all of the knowledge and ability that I have.

I think Mike is right that the response, that any psychological features biasing an individual's judgment means that the individual is not my peer, creates problems. If we're pressing the notion that hard, it seems like (given how even slight differences can create biases) we're getting close to saying you can't be an epistemic peer unless you are exactly equivalent to me in every respect relevant to epistemic matters; which is not something we are generally in a position to know. All we would know is that there are a lot of people whose status compared to me is simply indeterminate: maybe their my peers, maybe I'm slightly more competent, maybe they are, I don't know and have no way of ever knowing. The term becomes useless, and your claim that the sympathetic atheist should think the evidential theist his peer (rather than simply having no idea) seems wrong.

On the other hand, if we understand 'epistemic peer' in such a way that epistemic peers can differ in psychological biases, and thus in how they are inclined to judge in particular contexts, as it seems we need to if we are to find the term useful at all, Mike's argument seems to be right. If nothing else, we would have to allow that epistemic peers can have temperamental differences, and that the sympathetic atheist could reasonably suspect that some of his epistemic peers might be temperamentally likely to overlook a few salient evidences, or to be temperamentally inclined to accept (incorrectly) this or that apparent sign of non-gratuitousness at face value, through no serious fault of their own. And the real-world sympathetic atheist would have to factor in slight differences in education, personal experiences, etc.; and these lead to the same result as when we consider temperament.

Hi Mike, In each of those cases there’s a difference of abilities. In the chess case, you have the ability to play in crowded places whereas Jones’ lacks that ability. If you and Jones make the FIDE championship, my money is on you. To increase my confidence that you’ll win I’ll bring along some friends to watch. Mutatis Mutandis for the other cases. In the debate between the sympathetic atheist and the evidential theist the gambit you’re playing has the sympathetic atheist view the theist as (e.g.) psychologically impaired to rightly assess the evidence. That’s perfectly consistent with recognizing that the evidential theist can handle other sorts of evidence and is very competent, etc. But when it comes to gratuitous evils there’s some (e.g.) psychological mechanism at works that explains why an otherwise intelligent, competent individual can’t follow the evidence. Such an individual may very well be my epistemic peer in all other respects but not in this one.

Brandon, I think it’s possible to justifiedly believe that some individuals are your epistemic peers. Of course, epistemic peers sometimes differ in important ways—one believes some claim whereas another rejects it. Peter van Inwagen has some neat examples of epistemic peers in his paper “Is it wrong everywhere, always, and for anyone to believe anything on insufficient evidence.” He and Lewis are epistemic peers and I find it hard to take seriously that either has justification for believing the other is (e.g.) psychological impaired in the relevant respect. Here’s the link: http://comp.uark.edu/~senor/wrong.html

On the troubling self-defeat issue Jon raised, here’s a different response. The argument I gave doesn’t require the equal weight view in complete generality. It just requires that in cases of reasonable differences among epistemic peers in aggregative situations there’s a violation of the margin of error principle. It’s not clear that that succumbs to self-defeat. It’d have to be that the reasons for/against that principle constitute an aggregative situation. If it does I’ll take the Unger route.

Ted,

There is no difference in abilities, there is a difference, in certain circumstances, in the capacity to exercise abilities. It's silly to think that Kasparov is a less able chess player--or not my chess peer (can you imagine?!)--simply because he is hampered by agoraphobia and I can defeat him when he is really under that strain. It really just confuses things to blend exogenous psychological or contextual matters relevant to exercising a skill together with having a skill.

But suppose you insist on defining epistemological ability in psychological, sociological, etc. terms. In that case the Sympathetic Atheist can rightly note that theists are more susceptible than non-theists to certain non-evidential psychological influences that lead them to embrace theism. They would certainly have lots of empirical support on their side. The charge now that Sympathetic Atheists are failing to treat theists as epistemic peers has no traction. You've advanced a concept of 'epistemic peer' that requires such agents to be psychological and sociologiical (near)equivalents. But it is just an empirical fact that some agents are more susceptible than others to various forms psychological influence that lead them to find theism congenial. If pointing out such facts means pointing out that theistic agents are not our epistemic peers, then so be it. So long as they have the facts, that's a charge that SAists needn't worry about. The dispute now comes down to whether SAists have the psychology right. And on that question, I'd bet on the SAists.

[van Inwagen] and Lewis are epistemic peers and I find it hard to take seriously that either has justification for believing the other is (e.g.) psychological impaired in the relevant respect

But there is no need to use 'impairment' here. Certainly there is some explanation for why each of these agents weighs the evidence differently. To explain that difference in terms of PvI's psychological disposition toward theism (or what it promises) or his affinity for theism (or what it promises) seems to me perfectly reasonable. It seems just the way to go when there is this sort of epistemic standoff. But on your view, if there is this psychological difference, that makes them epistemic non-peers. I don't find it at all difficult to believe that they are epistemic non-peers in that sense. I hasten to add that this is not the sense in which I use 'epistemic peers/epistemic non-peers' or a sense I'd recommend.

At the risk of being completely off the mark, why do we think there are gratuitous evils? I take it that a gratuitous evil is one that has no possibility of good resulting, or the amount of evil is more then what is necessary for good to result. Can someone provide me with an example? I am familiar with Rowe's example of the deer, but that does not work. If there are no gratuitious evils the SAism is unwarranted, is it not?

Rowe's argument isn't that it is impossible that some greater good might result from a particular evil like that suffered by the deer, but that it is reasonable to believe that no greater good comes of it.

Mike,

On the ‘impairment’ language, I think it’s appropriate. The dialectic at this point is assuming the margin of error principle so the evidence has to be such that small changes in it do not make a difference to what it supports. So when the sympathetic atheist looks at the evidence and forms the belief (justified by her lights) that there are gratuitous evils, she also should think that small changes in the evidence won’t make a difference. She is then faced with an otherwise reasonable, intelligent person that disagrees with her. Why the disagreement? It’s not the evidence. It must be the person doesn’t appreciate the force of the evidence. Given that this is ongoing and a feature of a certain class of persons, they must have certain traits that prevent them from correctly assessing the strength of evidence. Whatever the mechanism at work, it is non-evidential in nature and prevents them from properly assessing the evidence. Hence, impairment. I think this is what the atheist should say but it’s not sympathetic atheism.

Abilities are probably contextual. In most contexts you and Jones have the same ability but not in every context. In the Kasparov example suppose we develop a variation on chess called “agora chess”, exactly the same rules and pieces as chess but played in an agora. Kasparov’s is a chess stud but he can’t play agora chess to save his life. I think though that we’re focusing on the wrong sorts of examples here. The issue with the SA and ET is whether the ET suffers some non-evidential mechanism that hinders his otherwise good ability to follow the evidence. If so then the ET isn’t competent to handle the evidence. We recognize that otherwise intelligent people aren’t competent to handle specific kinds of evidence. We dismiss jurors, for instance, when there’s some personal conflict. One reason we do this is that their personal interest can conflict with a proper weighting of the evidence.

Dave, you are correct and I should have been clearer. However, it seems to me that Rowe's argument is such that it can be interpreted that either atheism or theism can be reasonable based on the evidence. He does not rule out, if fact he admits the possiblity that theism is correct, only that the evidence suggests, to him, that it is not. It seems as if he is in the position of Wisdom's travelers who engage in an endless debate on which perspective best fits the evidence, with both using the evidence to support their respective, contradictory, positions. I would suggest that if this is so, then the rational alternative is to suspend judgment and remain agnostic.

Other then thinking that the deer's imagined suffering is gratuitous, what reason do we have that any suffering is gratuitous such that the SA's postion would be succesful over the theists. I do not see how this can happen, but I remain open to the possiblity that you, or someone, can provide me with a clear example of gratuitous suffering. If factual, not hypothetical, examples of gratuitous suffering are not available, then Ted is correct and SA is incoherent.

If so then the ET isn’t competent to handle the evidence.

Ted, I don't see how this follows. The introduction of the non-evidential bias doesn't give us any information about the person's competence to handle the evidence. If I and my epistemic peer get different answers to the bean-counting problem, it's not because of competence. By definition, in fact; epistemic peers are equicompetent. Rather, it's because of a mundane quirk in the actual count that competence doesn't rule out -- double-counting, skipping, etc. And occasional double-counting or skipping doesn't automatically mean a failure of competence in counting, particularly if it's a massive counting problem in which mistakes can on rare occasions be made by entirely competent persons. Likewise, the sympathetic atheist of the sort Mike suggests is at no point committed to thinking the evidential theist is not competent to handle the evidence well; he needs only to recognize that there is a quirk in the actual handling of it (e.g., too optimistic an assessment of certain evidences), and this quirk will have a causal explanation in the psychological makeup of the evidential theist, just as the counting quirk will have a causal explanation in the manner of counting.

(Note, too, that the SA need not hold that the ET's handling was bad, only that it was short of optimal. He could, in fact, concede that the ET's handling of it was genuinely excellent. All he needs to say was that it was flawed in a minor way that sufficed to lead to the wrong conclusion. Kasparov in agora chess might still give Mike one massive run for his money. If he suffers agoraphobia, that doesn't indicate his incompetence to play chess under those conditions but an utterly astounding competence, even if he loses.)

Brandon, I’d agree with you if the ET’s mishandling of the evidence were merely an occasional error. But the ET continually and systemically mishandles the evidence. The SA posits that this error is caused by some non-evidential mechanism. I don’t think the SA should view this as just the occasional quirk.

But the point, I take it, is that the ET doesn't mishandle the evidence on the SA's view; he simply fails to handle it optimally for that particular question. As I pointed out, the SA can regard the ET's handling of the evidence as genuinely excellent (perhaps much better than that of many atheists); perhaps even an important influence on his own. He's only committed to saying that there's a better handling of it, not that the ET's handling of it is in any sense bad. The ET isn't (as your formulation of the SA's position put it) "horribly wrong"; the failure to get the right result is entirely understandable.

By 'quirk' I didn't necessarily mean anything occasional; I meant it in a broader sense it sometimes has -- as when, for instance, I say that someone has a personal quirk. Personal quirks are (1) relatively harmless; (2) have manifest results; and (3) give a distinctive slant to one's behavior. So, I suggest, the SA can hold that ET's who are his epistemic peers tend to have personal quirks leading them not to draw the best conclusion. Indeed, the SA could even regard the quirk as generally admirable, and in many other areas making a valuable contribution to the inquiry, or even producing optimal results. It's just in this case that it turns out to produce a barrier to the optimum handling of the question.

Brandon, you may have in mind some sort of deontological claim regarding mishandling of the evidence. I’m not saying that the ET is blameworthy for his handling of the evidence. Given the presence of the non-evidential mechanism, it’s wrong to hold the ET blameworthy for his belief. But since the evidence in this case is not in the penumbra and it lays in the area that favors SA the ET lacks justification. That the ET doesn’t see this is due to a trait that prevents him from rightly assessing the evidence. Think about it this way: suppose you and I look at a color wheel. I lack, while you do not, the ability to see shades of red. I look at the color wheel and think that there are no shades of red on the wheel. You look and see shades of red. We disagree. Furthermore, you’re justified in believing that I lack this ability. In this case you should think that I’m just not competent to assess whether a color wheel contains shades of red. In this case it’s possible that I’m not to be blamed for the belief and, further, that I do the best I can. But due to this inability, I simply can’t tell whether shades of red are present. Similarly, for how the SA views the ET’s handling of the evidence.

Ted,

Why couldn't the SAist believe that, (1) from his perspective, he sees the color wheel better than the ET, (2) he sees why the ETist comes to his conclusion about the colors on the wheel, which is perfectly rational from the ETist perspective. And (3) he further sees why the ETist believes that it is the SAist perspective that is trammeled, rather than his own. In that case, certainly the SAist can sympathetically assess the position of the ETist.

Mike,

There are two ways of handling the case. First, the SAist possesses evidence the ETist doesn’t. The SAist has experiences representing shades of red and understands the significance of those experiences; the ETist does not have experiences that represent shades of red. In this case the similarity of evidence condition fails. The second way is that both the SAist and ETist have experiences representing shades of red but the ETist suffers some mechanism that blocks the mental uptake of those experiences (similar to the case of blindsight). In this case the similarity of evidence condition is met but the competency condition fails. And though the ETist isn’t to blame, he lacks justification for his belief. The SAist can in a sense have sympathy for the ETist but the sympathy is more like sympathy towards those that just can’t see that the rule modus ponens is valid. It’s not like the mutual respect one has towards an epistemic peer.

There are two ways of handling the case. First, the SAist possesses evidence the ETist doesn’t. The SAist has experiences representing shades of red and understands the significance of those experiences; the ETist does not have experiences that represent shades of red. In this case the similarity of evidence condition fails. The second way is that both the SAist and ETist have experiences representing shades of red but the ETist suffers some mechanism that blocks the mental uptake of those experiences (similar to the case of blindsight).

I didn't have in mind either of these. There are other possibilities. Suppose neither the SAist nor the ETist views the wheel from an ideal point of view. From the ETist point of view, it appears that the SAist is in a worse viewing position than he is. And the SAist knows that it appears that way to the ETist. From the SAist point of view, it is clear that the ETist is in a worse position to view the wheel. It is clear to both that neither is viewing the evidence from an ideal point of view. Neither one weighs the evidence wrong, neither one displays any incompetence. The evidence from both points of view is skewed. But conditions are better for the SAist. The SAist position is coherent. He understands that were he in the position of the ETist, he too would think that his epistemic position vis a vis the wheel is the better of the two. Recall, that's exactly what it looks like from the ETist position.

"From the SAist point of view, it is clear that the ETist is in a worse position to view the wheel."

Granting this, I’m not sure what the rest adds. If it’s clear to the SAist that the ETist is in a worse position to view the wheel then the ETist either doesn’t have the same evidence as the SAist or isn’t as competent as the SAist to handle the evidence.

Granting this, I’m not sure what the rest adds. If it’s clear to the SAist that the ETist is in a worse position to view the wheel then the ETist either doesn’t have the same evidence as the SAist or isn’t as competent as the SAist to handle the evidence

This interpretation of "having the same evidence" strikes me as a little silly. I mean, no kidding, no two people ever have the same evidence on ANYTHING. No two people ever have the same quality of evidence on anything. That never happens; it's a pure idealization. That exalted observation is going to play some central role in this argument?

I don't think the color case is at all similar to the SA's case, because it is far too simple to convey anything parallel to the SA's claim.

Suppose instead that we have exactly the same ability to discriminate color, but I simply glance at it and you study it closely. The difference between my quick glance and your close study is not a difference of ability, and this is true even if I am inclined (by temperament, culture, or whatever) to glance at rather than closely study things that I see. Your ability to discriminate color (or, for that matter, to study things closely) does not outstrip mine. But if I'm not inclined to pay close attention to visual matters, then even if I had an ability greater than yours to discriminate color, I'm still at a disadvantage in any particular case of color discrimination; my ability to discriminate color would have to far outpace yours in order to compensate for this disadvantage.

We can even complicate the matter further. Suppose that you sometimes second-guess yourself in things you closely study. Then it might be the case that my quick glance sometimes carries the advantage, and not vice versa; but, having carefully examined yourself and considered the type of situation in which you find yourself, you are quite certain that in this particular type of case, even if in no other, your close study will tend to result in your being more correct than my quick glance.

Scenarios like these, distant though they still are, are certainly closer to what we get with the SA than your suggested scenario. The SA is not saying that the ET is color-blind; he is saying that the ET is naturally inclined not to discriminate to the necessary precision (which is radically different from saying that he lacks the ability to discriminate to that precision if he makes the effort to abstract himself from that inclination; that would be to deny that he's an epistemic peer at all).

Mike,

Someone flipped a fair coin in Las Vegas last night. I think it’s pretty fair to say that we have the *same* evidence about whether it landed heads. But anyway the spirit of what you say is quite right. All that’s required is that the similarity of evidence condition is satisfied. Here’s a different way of running the relevant part of the argument: Assume the similarity of evidence condition is satisfied and there’s disagreement. Either the disagreement is between evidence peers or not. If disagreement is between epistemic peers then either the relevant margin of error principle is violated or not. If the disagreement between epistemic peers and the margin of error principle is not violated then the presence of disagreement undermines the possibility of justification in the original case. Sorry to be brief, but I have to go build a fence. More later.

Someone flipped a fair coin in Las Vegas last night. I think it’s pretty fair to say that we have the *same* evidence about whether it landed heads.

I do too, but we weren't talking about what it is "pretty fair to say", right? We were talking about whether any two people ever have exactly the same evidence, since that is how you were interpreting the principle. But we don't have such evidence even in this simple case. For instance, who did you hear the story from, the same person or source I did? I doubt it, but suppose so. Do you have exactly the same credence in this person or source as I do? Right down to the last decimal point? Do consider him/it exactly as reliable as I do? I doubt it, but suppose you do. Was your assessment of his level of sincerity exactly the same as mine? I doubt that too...and so on. We always weigh evidence differently in thousands of ways having to do with the source of the evidence and our own particular physical and mental state at the time we receive the evidence. That seems undeniable.
Anyway, concerning your newer argument, I guess I'd have to see your formulation of the error principle. Also, I'm not sure that it matters (to the larger point you want to make) whether the perception of the less-well-positioned perceiver is sufficiently skewed to meet the standards of the error principle or whether it matters instead that the other perceiver has a justified belief that it is so skewed. My guess is that it is the latter that matters.

Mike (and Brandon),

The Vegas coin flipping example was intended to be an example in which we both have no evidence for a claim. Ergo, we have exactly the same evidence in for the claim. Anyway, this point really doesn’t matter for the overall dialectic. On more significant issues, one thing to keep in mind is that as the margin of error principle is precisified so the epistemic peer claim is precisified (i.e., the similarity of evidence & competency claims). I think this is the direction we’re headed. You (and Brandon) are pushing on the idea that similarity of evidence admits of small changes and some small changes can be epistemically significant but not undermine the epistemic peer claim. I’ll reply that if small changes are epistemically significantly then we need we need precisify the margin of error claim in which case it’s no longer reasonable to think the similarity of evidence condition is met. You pull on one thing and another thing changes position. Anyway, I’ll try to make this explicit in the paper which I’ll send you when I write it. Thanks for the great discussion!

The Vegas coin flipping example was intended to be an example in which we both have no evidence for a claim.

Ted, please send a copy of the paper along. I'd love to see it.
On the Vegas coin flip, we do we have evidence. Aside from the points above, there are no perfectly fair coins. I have no doubt that every existing coin is to some degree biased: but how biased? In what direction? My credences will differ from yours. There is also a series of credences each of us has for skeptical hypotheses. I place some very small credence in the BIV hypothesis, a different credence in Unger's mad scientist, a near zero credence in Cartesian hypotheses. I'm sure our credences for these are not the same. So there is some chance that I don't know that there exists a coin that was flipped, let along a quasi-fair coin. I also place a small credence that I misheard or misread that such a coin was flipped. What this comes down to is that I would not consider literally a 50/50 chance that the coin came up Heads. Neither would you. Our varying evidence about the coin, its existence, its fairness, etc., determines our credence that, if flipped, it would come up heads/tails. I was urging that in every actual case our credences will vary.