God’s Knowledge and the Evidence Thesis
December 4, 2011 — 11:54

Author: Andrew Moon  Category: Concept of God  Tags: , ,   Comments: 46

In a forthcoming paper, I defend the view that knowledge does not require believing on the basis of evidence. In other words, I argue against what I call the “Evidence Thesis”, which states:

(Evidence Thesis) S knows that p at t only if S believes that p on the basis of evidence at t.

How does the evidence thesis relate to evidentialism, formulated and defended by Earl Conee and Richard Feldman? Well, their view is about epistemic justification, and it states that the doxastic attitude one is justified in having is the one that fits the evidence. Evidentialism is a popular view, and we can see that it is distinct from the evidence thesis. However, VERY MANY evidentialists endorse the evidence thesis. So do VERY MANY internalists. On the other hand, almost no externalist will endorse the evidence thesis. So long as one’s true belief was produced in the right way (e.g., by a reliable process, with safety, with sensitivity, by properly functioning faculties, by an exercise of the right sort of ability, etc.), the belief counts as knowledge. Despite the fact that some people seem to presume the truth of the evidence thesis, we can see that a great many theories of knowledge (the externalist ones) entail that it is false. And I argue that it is false in my paper. So, my argument both provides support for the many externalist theories of knowledge and also gives many evidentialists and internalists a reason to revise their views.
In this post, I want to try out a counterexample against the evidence thesis.

Suppose God exists. (Even if God doesn’t exist, we can use God in a counterexample against the evidence thesis.) God knows all true propositions; God does not believe propositions on the basis of evidence; therefore, the evidence thesis is false.

When I first reflected on God’s knowledge, as pertaining to the evidence thesis, I was inclined to think that God did not believe truths on the basis of evidence. It’s not that God is appeared to bluely when looking at the earth and, on that basis, believes that the earth is blue. God just knows and believes (without requiring extra perceptual evidence) that the earth is blue. To use some other examples, God just knows and believes (without requiring extra evidence) that 1+1=2, that there are exactly 14,000,000 angels, or that Cain sinned. God does not need an a priori insight, or it does not need to seem to God that these propositions are true.
That was my pre-theoretical understanding. But it’d be nice to have some arguments, and they don’t clearly go one way or the other for me. I suspect that most evidentialists and internalists would just say that God does know those propositions on the basis of evidence. I’d deny it and tell them so. And then we’d sit there staring at each other, unhappy that the other person is unconvinced. (In fact, this has happened to me, on two separate occasions, with different internalists!)
So, here are some questions that I’d like to ponder and get readers’ reactions to:
1) Are there are any theological reasons to hold one view over the other? Is there evidence from Biblical, Quranic, Jewish or other possibly revelatory texts?
2) Does God believe some propositions on the basis of other propositions that God believes (or on the basis of God’s other beliefs)?
3) Does God gain perceptual knowledge on the basis of perceptual experience (or a sense datum or being appeared to a certain way) in the way we do? (I recall reading Anthony Kenny saying that the medieval philosophers dealt with this question. Any help from them?)
4) Does God gain memorial or a priori knowledge on the basis of intellectual or memorial seemings?
5) Is it plausible to believe that God could have all these perceptual experiences and intellectual and memorial seemings all at once?
Then again, I think Alston didn’t even think that God knows individual propositions, but I don’t think I ever understood his view. Anyway, any thoughts on (1) – (5) are welcome!

Comments:
  • Nate Shannon

    There are indeed theological perspectives on this question, but it depends on your theological perspective. If you’re an Open Theist, I think you’d say that God’s knowledge must be at least in some way evidentially grounded, at times. If you hold to middle knowledge, God’s knowledge of future contingents would appear to require discursive – if instant – reasoning, based on evidence, if the contingent facts of possible worlds maybe count as evidence. If you are an “Anselmian” theist, you would say that the doctrine of simplicity implies that God’s knowledge is distinguishable but inseparable from his being and from every other divine attribute. The ‘a se’ God knows all things because before creation there only was God, and everything which came to be came to be by virtue of divine creation/sanction.
    That being said, I’m not sure any of this will do any good if what you’re interested in is the evidential thesis regarding man’s knowledge. God appears to know things exhaustively, a kind of knowledge unique to the creator of all things. I think the knowledge situation is displayed well in Genesis, when after having created the garden, God places Adam into it. God places Adam into a knowledge situation, surrounded by facts which are pre-known (exhaustively) by God, and are afterward known by the creature. I have wondered what the epistemological significance was of Adam naming the animals; an exercise of vice-regency? Whatever that is, this scenario is a good illustration of the creator/creature distinction epistemologically.

    December 4, 2011 — 14:01
  • Andrew Moon

    Thanks Nate.
    On your first paragraph,
    Right, if you’re open theist or molinist, you can answer my (2) by saying that God does believe some things on the basis of other beliefs.
    I must confess that I lost you in your first paragraph when you brought up the Anselmian theist. More specifically, I’m not sure how those points were relevant to my post.
    On your second paragraph,
    The Evidence Thesis is about knowledge simpliciter, not just knowledge that humans have. I think that this is what epistemologists generally are interested in. For example, this is one of the objections that Evan Fales raises to Alvin Plantinga’s proper functionalist theory of knowledge; God wasn’t designed, so God’s faculties don’t function properly, so God doesn’t have knowledge according to proper functionalism.
    Of course, we could restrict the Evidence Thesis to creaturely knowledge. That would get rid of the problems for both proper functionalism and proponents of the Evidence Thesis. Anyway, I’m interested in whether there’s a problem for the evidence thesis when we take it to be a thesis about any knowledge.

    December 4, 2011 — 14:24
  • Andrew:
    Hello. Hope you don’t mind if I make a couple of brief, conceptually disconnected comments.
    I suspect that this objection has already been raised and addressed, but what about the criticism that the Evidence Thesis (ET) seems to fail to satisfy its own criterion? Is there some available evidence which could plausibly substantiate the ET?
    It seems to me that one might encounter a certain difficulty when applying ET to an essentially omniscient being like God, for many theists are inclined to affirm a doctrine of God according to which there’s no time t at which God lacks knowledge of some true proposition. (Maybe ET doesn’t require a temporal index.) But this doctrine doesn’t appear to entail that God never acquires knowledge, for someone like William Lane Craig suggests that, with each passing moment, God might know it’s now this particular time and no longer is it that particular time. So perhaps this further suggests that God’s acquisition of knowledge only pertains to circumstances in which He acquires knowledge of contingent truths. If so, then I think it’s plausible to explain God’s acquisition of contingent truths by appealing to God’s immediate knowledge of His own actions, since He’s the one who brings about all contingent states of affairs. This consideration seems relevant to question (2) above, for God may “come to believe” the proposition There is an actual world on the basis of His believing some other proposition, like I just actualized a possible world.
    Do you think ET involves any commitments about knowledge acquisition?
    Thanks.

    December 4, 2011 — 15:58
  • Nate Shannon

    “Simple” and “a se” theism presents the boldest challenge to the legitimacy of “knowledge simpliciter” which would be common to creator and creature alike. There’s plenty more to what I mentioned above (there is a traditional distinction between God’s knowledge ad intra and ad extra, of himself and then of everything else, for example, and the distinction between the two worked out in terms of voluntary condescension, a concept demonstrated in the incarnation). Or perhaps the challenge is to what I understand as a missing premise in your evidential thesis: what is true for God’s knowledge is true for human knowledge, or something along those lines. I think that is a whopper, and it is where I’d aim my critical energy if I were attempting to counter the thesis.
    As for an objection like that to proper function, goodness – I don’t think Plantinga ever intended that to be “knowledge simpliciter.” He draws from a Reidian anthropology which begins with man as created by God. The “design plan” is obviously borrowed from a theistic framework. Sure, “God is not designed,” but no one said he was. “Therefore he has no knowledge” is a whopping non sequitur of which I’m sure Plantinga is innocent. I suppose I would object to the idea of “knowledge simpliciter,” ultimately because I reject “being simpliciter.”
    Thanks again

    December 4, 2011 — 18:16
  • Gordon Knight

    I have always thought God’s knowledge consisted of immediately grasping the facts in questiosn. God knows I am typing this in virtue of having immediate acquaintance with my act of typing. This strikes me as the clearest sort of evidence, but if by evidence you mean propositional reasoning, then what God does (so I with great hubris suppose) is something different.

    December 4, 2011 — 18:57
  • Andrew Moon

    Marc,
    On your first point, evidentialists will likely say that the evidence thesis is supported by evidence, and that their belief in it is based on evidence. For defenses, you could just read many articles by Conee and Feldman. (If you want more specific references, just ask me!)
    On the second point, right, if God is outside of time, and so God does not know propositions at times, then anything involving God would not be a counterexample to the evidence thesis. If you hold a view like Craig’s or other philosophers who think that God is in time, then this discussion is relevant. I think that God does know propositions at times. For example, like Craig, I think that God knows, right now, that Andrew Moon is typing.
    Nate,
    I don’t think I understood everything you said in that first paragraph (no need to explain further if you don’t want to), but I think that I don’t have anything interesting to say in response.
    So, I think that in the sentences, “God knows that p” and “Fred knows that p”, knowledge is being ascribed in both cases. Furthermore, the meaning of ‘knows’ does not change from the one sentence to the other. I suppose you disagree?
    Gordon,
    We want an understanding of ‘evidence’ that is in accord with the sort of thing that paradigm evidentialists in epistemology care about.
    Now, I had a hard time following exactly what you were saying. Maybe this will help. Do you think that ‘graspings” (in your first sentence) and “acquaintance” (in your second sentence) are the same thing?

    December 4, 2011 — 20:18
  • Dennis Trate

    This may be outside your scope, but I’d be interested to hear more explanation on the move corresponding human epistemology to divine epistemology.
    At least in the Augustinian tradition, God’s knowledge seems to differ from humanity’s by a matter of kind, not degree. “Has God hair and nails?” Biblical evidence is difficult to nail down since it’s seldom clear whether the author speaks anthropomorphically.

    December 4, 2011 — 20:38
  • I wonder if this ends up in a sort of tautology. I think the following proposition sounds plausible: It is necessarily the case that God’s knowledge of a proposition is sufficient evidence for belief in that proposition. I think this would obviously work in cases of regular people were they to find out some of God’s knowledge. I’m not as sure it would work with God.

    December 4, 2011 — 23:53
  • Dani Rabinowitz

    Andrew
    Thanks for these interesting questions. Here are a few comments.
    1) As I understand externalists about knowledge, they don’t think ET false, they merely think that their favored condition involves evidence in one way or another such that their condition does a better job than ET on the whole. In other words, it would surprise me if any externalist would say that knowledge bears no relation to evidence and hence that’s why their condition does not mention evidence. Relying on evidence, for example, is a reliable belief forming mechanism. Or, when one uses evidence in the correct manner one usually thereby satisfies the sensitivity, safety, relevant alternatives conditions etc. Evidence plays a central role in Williamson, for example.
    2) Some externalists think ET false but not for the reasons you think. Williamson, for example, in his reply to Breuckner in Williamson on Knowledge, explains that knowledge that p and evidence for p are often acquired simultaneously i.e. one does not believe p on the basis of evidence for p.
    3) In the medieval tradition the nature of God’s knowledge was often taken to be direct in the manner of what we now call a priori knowledge. (See Plantinga on a priori knowledge for an accessible discussion thereof.) Some medievalists e.g. Gersonides thought God did not know contingently true propositions. This may both solve your problems but also bust your project at the same time. Probably the most relevant part of medieval philosophy of relevance to your paper will be claims by those who thought “God identical with his knowledge” e.g. Maimonides and Aquinas (?).

    December 5, 2011 — 5:08
  • Dani Rabinowitz

    Andrew
    You might also wish to consider moves by those, such as contextualists, that there is no one unique knowledge relation that holds fixed across all contexts. AS such, one might think that a similar line of reasoning could apply to your problem: there is one kind of knowledge relation (and those in its vicinity) that is appropriate for contexts in which human knowledge is being discusses, and another knowledge relation for contexts in which God’s knowledge is being discussed. Naturally this thought requires far more unpacking that what I’ve done here.

    December 5, 2011 — 7:56
  • Sorry, my wording of that was a little off – I said ‘belief’ but meant ‘knowledge’. Essentially, what I mean is that when you’re the type of being that is all knowing, then in some sense there really is no better evidence that could be available beyond you knowing it. It would certainly be sufficient as evidence for one of us.

    December 5, 2011 — 8:57
  • I don’t know almost anything about the evidence literature, but I wonder if you’re not assuming an internalist view of evidence as some sort of a seeming or other internal state.
    But if we are consistently externalist, can’t we embrace the Evidence Thesis? What is my evidence for thinking there is a screen in front of me? It is the photons coming from the screen, the reliability of my faculties, etc. This is the evidence even if I don’t know anything about it–this is an externalist theory, after all. And what is God’s evidence that snow is white? It is the whiteness of snow.

    December 5, 2011 — 9:04
  • Andrew Moon

    Alex,
    I think I do understand ‘evidence’ in an internalist way.
    And doesn’t seem plausible? Are you familiar with BonJour’s case of Norman the clairvoyant? Norman has a reliable clairvoyant ability, but he has no idea about it. He suddenly believes that the president is in New York; his clairvoyant ability causes this belief, and so the belief is reliably produced. But it seems odd to say that Norman has evidence for his belief, or that Norman’s belief is based on any evidence, where the evidence is that his belief was reliably produced. (I’ll note that BonJour intended this case for another purpose.)
    Also, it should be noted that there are externalist and internalist theories of what makes one bit of evidence support a belief. Some say that E supports proposition p because a believer with properly functioning faculties would believe p upon gaining evidence E. On the other hand, internalists tend to think that the supports relationship between E and p is a necessary one. (Kvanvig calls the externalist theory the doxasticist view, and the internalist theory the propositionalist view.)

    December 5, 2011 — 9:49
  • Building (perhaps) on what Alex Pruss said, suppose we take evidence possession to be a perfection, so that a perfect being would possess the maximal possible amount of evidence. On this view, it seems natural to say that the whiteness of snow – or perhaps the proposition that snow is white – is included in God’s evidence. Since God is maximally perfect, and evidence possession is a perfection, pretty much everything gets to be part of God’s evidence!
    Being slightly more careful, we should probably not say that God possesses *everything* as evidence. Instead we should say that he possesses as evidence *everything that can be evidence*, or perhaps *everything that can be had by God as evidence*, or perhaps even *everything that can be had by God as evidence at that particular time*. This raises the issue of what it takes for something to be possessable as evidence, or possessable by God as evidence, or possessable by God as evidence at a given time. If any old proposition will do here, so that any proposition can be possessed (by God) as evidence, then perhaps this vindicates the view that God’s evidence includes the proposition that snow is white.

    December 5, 2011 — 10:07
  • I don’t know if internalism or externalism is true. But I naively find it fairly plausible that what one should say about whether Norman has evidence goes corresponds with what one should say about whether Norman knows.

    December 5, 2011 — 11:18
  • Dani Rabinowitz

    Andrew
    Thanks for these interesting questions. Here are a few comments.
    1) As I understand externalists about knowledge, they don’t think ET false, they merely think that their favored condition involves evidence in one way or another such that their condition does a better job than ET on the whole. In other words, it would surprise me if any externalist would say that knowledge bears no relation to evidence and hence that’s why their condition does not mention evidence. Relying on evidence, for example, is a reliable belief forming mechanism. Or, when one uses evidence in the correct manner one usually thereby satisfies the sensitivity, safety, relevant alternatives conditions etc. Evidence plays a central role in Williamson, for example.
    2) Some externalists think ET false but not for the reasons you think. Williamson, for example, in his reply to Breuckner in Williamson on Knowledge, explains that knowledge that p and evidence for p are often acquired simultaneously i.e. one does not believe p on the basis of evidence for p.
    3) In the medieval tradition the nature of God’s knowledge was often taken to be direct in the manner of what we now call a priori knowledge. (See Plantinga on a priori knowledge for an accessible discussion thereof.) Some medievalists e.g. Gersonides thought God did not know contingently true propositions. This may both solve your problems but also bust your project at the same time. Probably the most relevant part of medieval philosophy of relevance to your paper will be claims by those who thought “God identical with his knowledge” e.g. Maimonides and Aquinas (?).

    December 5, 2011 — 12:07
  • Andrew Moon

    Hi Dani,
    It’s good to hear from you, as it’s been awhile since Purdue!
    Thanks for the earlier comment and this one. Regarding this latter one, regarding your point (1), all externalist theories of knowledge, at least all that I’m aware of, entail that ET is false. By externalist theory of knowledge, I mean a theory of knowledge espoused by someone who is known by most people as an externalist (e.g., Armstrong, Dretske, Goldman, Sosa, Plantinga, Nozick, Greco, Zagzebski, etc.). Do you disagree?
    On (2), thanks for the reference! I didn’t know he said that.
    On (3), thanks for the info. From what you said, I didn’t find anything that would threaten my view that ET is false.

    December 5, 2011 — 13:22
  • Marc Belcastro

    Dr. Moon:
    Echoing Dr. Pruss, I’m not really acquainted with the evidence literature either, so what articles by Conee and Feldman would you recommend? Thanks. I’m especially interested to see what kind of evidence could support the claim that S’s believing that p on the basis of evidence is a necessary condition for S’s knowing that p. To my mind, only some rather impressive evidence could establish this claim.
    Also, how do evidentialists and advocates of the evidence thesis generally define “evidence”? Suppose I know, a priori, that the proposition There are truths is true. Call this p. According to evidentialists and the evidence thesis, since I know p, I know p because I believe p on the basis of evidence. But what might serve as evidence for a proposition—like p—which I know, a priori, is true? Perhaps I arrived at my knowledge of p by reasoning that, if I deny p, I actually generate a true proposition q, and so I ought to believe that p is true because it can’t be false. Would this reasoning process itself constitute (knowledge-conferring) evidence for p?

    December 5, 2011 — 15:50
  • Trent Dougherty

    Did not read any of the above, but Williamson and Plantinga both basically endorse the evidence thesis.

    December 5, 2011 — 20:02
  • Andrew Moon

    Hi Marc,
    haha, feel free to call me ‘Andrew’. A great intro to C&F’s thought (and widely read) is their 1985 article “Evidentialism”, which I believe came out in Phil. Studies.
    A really nice exchange occurred between Plantinga and Feldman, which was very formative in my thinking, and which I think is a nice introduction to these issues. I’d recommend reading pp. 357-361 from Plantinga’s paper “Respondeo” in the book “Warrant in Contemporary Epistemology”, 1996, edited by Jonathan Kvanvig. (This is an excellent volume, btw.) Then read C&F’s response to Plantinga in their (also widely read paper) “Internalism Defended” (2001). This is the very end of a discussion that began between Plantinga and Feldman in 1993, and Plantinga gives the references to the earlier works in the 1996 work. I think it’s all good philosophy and instructive, but you can get away with starting at the 1996 work, I think.
    On your second paragraph, for the many ways that internalists understand ‘evidence’, take a look at my forthcoming paper, which I linked above! I think that I lay out many of the options pretty clearly (and I give tons of references too). Of course, feel free to ask more questions.

    December 5, 2011 — 22:43
  • Andrew Moon

    Marc,
    Oh, actually, I think pp. 185-193 from Plantinga’s Warrant and Proper Function are very helpful for understanding evidence. I’d recommend reading that after you read Conee and Feldman’s “Evidentialism” (since Plantinga is responding to them).

    December 5, 2011 — 22:48
  • Dani Rabinowitz

    I’m with Trent here; belief that is in some manner appropriately connected with evidence is necessary for knowledge even for the externalist. It’s just that formulating an anti-luck condition in a manner that does not mention evidence is a much better way of treating the Gettier problem than a condition based entirely on evidence. In other words, an evidence based clause will not handle Russell’s stopped clock case (unless the condition is “cooked” in a way to handle the unique nature of evidence in such cases). So I think it is misleading to think that all externalist theories of knowledge entail that ET is false. I suspect that Trent’s new book will deal with these issues more extensively. And as Trent says, and as I said earlier, evidence is at the heart of Williamson’s safety condition given that he thinks evidence = knowledge.

    December 6, 2011 — 4:16
  • Nate Shannon

    Hi Andrew,
    Reading the comments, I see God has dropped out of it, and I’m still convinced that is for the best. The way I see it, in your argument you’ve (1) assumed that God’s knowledge is not like man’s knowledge, and you’ve also (2) assumed that God’s knowledge is indeed like man’s knowledge. What I mean is that (1.1) you’ve treated God as a different kind of being – you’ve bracketed his existence. Then you go on to discuss what kind of knowledge he has. Unless you’re just talking about a guy you’ve only heard about and never met, whose name is God (maybe a Norwegian name…) then you’re talking about some kind of personal, eternal – or at least everlasting – transcendent in some way, spiritual being. Maybe even a triune creator. A being like that will have knowledge vastly different than our own, because his mind and experiences are vastly different than our own, as would be his relationship to the facts which constitute evidence. (1.2) Secondly, if God’s knowledge were not significantly different than our own, why introduce him into the argument? What advantage could there be? Why not just say, “My neighbor down the street, does he have knowledge on the basis of evidence?” Or – knowledge simpliciter – “My goldfish..” or “that goose” etc. Since you introduce God into the argument, it is a going assumption that his knowledge is different (i.e. you’re not operating with knowledge simpliciter), but then there is a missing premise, (2)the one which affirms knowledge simpliciter. I suspect, therefore, that your approach here harbors a significant inconsistency.

    December 6, 2011 — 6:09
  • Nate Shannon

    I meant to conclude by saying that I am still dubious that ‘God knows that p’ and ‘Fred knows that p’ do not make use of significantly distinct notions of knowledge; and that the fact of this difference (or variety or ambiguity) is in part assumed already when you introduce God into the discussion. That’s what I’ve been trying to get at, I suppose.
    Thanks again

    December 6, 2011 — 6:52
  • Andrew Moon

    Trent,
    I think that Williamson would reject ET for the reason that Dani already gave in his December 5, 2011 12:07 PM comment.
    Trent and Dani,
    Take Plantinga’s theory (stated loosely): S knows that p iff 1) S’s true belief that p is formed by properly functioning faculties, 2) in an appropriate environment, 3) according to a design plan aimed at truth, and 4) formed reliably. To say that knowledge requires evidence on this theory is to say that conditions (1)-(4) entail that S believes on the basis of evidence.
    Do you think that (1)-(4) entail that S believes on the basis of evidence? That seems implausible to me. And we could say the same for other externalist theories. The reasoning seems straightforward to me.

    December 6, 2011 — 11:39
  • Andrew Moon

    Nate,
    You say that I say,
    “in your argument you’ve (1) assumed that God’s knowledge is not like man’s knowledge, and you’ve also (2) assumed that God’s knowledge is indeed like man’s knowledge.”
    Now how do I do that? And is the “like” here supposed to not change meaning from (1) to (2)?

    December 6, 2011 — 12:08
  • Dani Rabinowitz

    Andrew,
    As for my comment on Williamson, he only says that sometimes knowing p and acquiring evidence that p can occur simultaneously. A good many other times the evidence comes first.
    I can’t speak for Plantinga, but let me hazard a guess: conditions (1)-(4) do not entail that ET is false because believing p on the basis of evidence can satisfy all of those conditions.

    December 6, 2011 — 12:38
  • Andrew Moon

    Dani,
    “conditions (1)-(4) do not entail that ET is false because believing p on the basis of evidence can satisfy all of those conditions.”
    Right, believing p on the basis of evidence can satisfy those conditions. But that’s not the issue. Let me put it more clearly. Consider,
    A) Necessarily,
    (A) clearly seems false. And if (A) is false, then ET is false.

    December 6, 2011 — 13:59
  • RyanB

    Here’s a quick and rough argument I’ve toyed with for the conclusion that God’s knowledge requires evidence:
    1. God is maximally intellectually excellent.
    2. Maximal intellectual excellence requires comprehensive understanding.
    3. Comprehensive understanding requires comprehensive satisfaction of curiosity.
    4. Comprehensive satisfaction of curiosity requires evidence-possession–even evidence possession of a very first-personal exeriential sort.
    5. So, God’s knowledge requires evidence possession–even evidence possession of a very first-personal experiential sort.
    1 and 2 should be rather uncontroversial. Something very like 3 is defended by Jon Kvanvig in his recent work on understanding. Understanding is supposed to be that at which the ubiquitous motivation of curiosity aims. 4 is motivated by cases where a person’s curiosity is not satisfied without her having evidence for her belief of a very first-personal, experiential sort. Jon also mentions some cases which motivate this conclusion, like his cross-dresser example.
    I’m not sure the argument as stated shows that all of God’s knowledge must be evidence-based, though. But perhaps the argument could be fixed up to get this conclusion.

    December 6, 2011 — 21:07
  • Marc Belcastro

    Andrew:
    Thanks again for the very helpful references. I appreciate it, as I’ve been meaning to get my hands on some material by Feldman for a while now.

    December 7, 2011 — 11:23
  • christian

    hi andrew.
    “Despite the fact that some people seem to presume the truth of the evidence thesis, we can see that a great many theories of knowledge (the externalist ones) entail that it is false.”
    this isn’t right. if the basing relation is a causal relation, then many externalist theories are compatible with the evidence thesis. yet if you make the basing relation an inferential relation, then many internalists will deny the evidence thesis.

    December 7, 2011 — 17:53
  • Regarding your first question, I would think that there are passages in the Qur’an which are, at the face of it, suitable for construing as internalism. Hence it is not accidental that muslim theologian are historically internalist, unlike Muslim mystics, such as Rumi, who seem to be against internalism and hence sometimes mistakenly called anti-rationalist. Here are some passages from the Qur’an:
    And they say, ‘None shall enter Paradise except that they be Jews or Christians.’ Such are their fancies. Say: ‘Produce your proof, if you speak truly.’ (2:111).
    Who originates creation, then brings it back again, and provides you out of heaven and earth. Is there a god with God? Say: ‘Produce your proof, if you speak truly (27:64)

    December 7, 2011 — 19:59
  • Andrew Moon

    Ryan,
    That’s an interesting argument, and it’s the sort of thing I was looking for. Thank you. I think I’d disagree with (4). You say, “4 is motivated by cases where a person’s curiosity is not satisfied without her having evidence for her belief.” But even if there are such possible cases, this wouldn’t show that all knowledge requires this. And ET is a statement about all knowledge.
    Marc,
    when it comes to epistemology, you don’t get much better than Feldman!
    hi christian,
    On your first sentence, right, but most of them aren’t. (See my responses to Alex, Trent, and Dani.) On your second sentence, like many internalists, I don’t interpret basing to be inferential. That makes ET a more powerful view and harder to attack (but it’s still false!).
    Yaser,
    thanks! I think I’ve seen those verses before. I’ve never heard of Rumi, but I think I’m familiar with any literature connecting these areas of epistemology with Muslim philosophers. The verses are interesting too.
    On your sentence sentence,

    December 7, 2011 — 21:41
  • Todd Long

    Andrew,
    I don’t know why you dismissed Gordon’s point about acquaintance as you did. Surely Fumerton counts as a paradigm evidentialist, and he thinks that acquaintance with a fact yields excellent justification. It is true that, in his _Metaepistemology and Skepticism_, Fumerton ends up a little skeptical about whether we *humans* actually have acquaintance with facts that yield the evidence we need to refute external world skepticism, but it is clear that he thinks that one’s having acquaintance with facts would yield excellent justification. And I can’t think of a reason that any other paradigm evidentialist would disagree. After all, what could be better evidence for p than acquaintance with the fact that p? Such acquaintance would surely indicate the truth of p to one; so, it would count as evidence for p.

    December 7, 2011 — 23:42
  • Andrew Moon

    Dear Todd,
    I was a little puzzled that you thought I dismissed Gordon’s point. Why do you think that?
    Maybe it will help to let you know that I do take acquaintance seriously, I agree that many evidentialists would think it’s evidence, I have read Fumerton’s book, and I was thinking about Fumerton’s view when I read Gordon’s comment. However, I’m currently not sure if grasping a proposition (or fact) is identical with being acquainted with it (or the fact), and Gordon mentioned both. Hence, I wanted to first get clear on what Gordon was saying before I responded. Hence, I ended my comment with a question asking for clarification. If he responded, then I’d proceed in dialoguing. Otherwise, other folk have kept me busy.

    December 8, 2011 — 0:22
  • christian

    andrew.
    if basing is not a causal relation, and if it is not an inferential relation, what is the basing relation?

    December 8, 2011 — 0:27
  • Andrew Moon

    Christian,
    Well, that’s a hard issue that requires looking at the nice, bulky literature on the basing relation! Let me know if you’re interested in some references. Btw, I’m inclined to think that a causal relation is a necessary condition for the basing relation, but I don’t think it’s sufficient for it. I didn’t mean to communicate that I thought otherwise.

    December 8, 2011 — 0:45
  • christian

    hey andrew.
    i agree this is a hard issue. i guess i wonder when *you* say that the evidence thesis is false, what it is that you mean by ‘on the basis of’ as it occurs in the very thesis you say is false. at this point, and precisely because the issue is, in fact, a very hard issue, i simply have no idea what the thesis is that you think is false, and precisely because i have no idea what you mean by ‘on the basis of’ as it occurs in the sentence that expresses the thesis you claim is false.
    and to respond to a potential reply: i do not think there is some pre-theoretic understanding of ‘on the basis of’ that we can appeal to in this case.
    moreover, notice that your opponent could respond to your argument against the evidence thesis by saying: “well, i just have a different sense of ‘on the basis of’ in mind when i argue for the evidence thesis.” you could of course push your opponent to articulate this sense, as you should, but then, i guess, she should do precisely the same thing to you. and so the question: “what do you mean by ‘on the basis of’?” seems to me to be very important, that is, if we are to assess whether the evidence thesis is true or false.
    this seems like a pertinent question to me, unless i am missing something.

    December 8, 2011 — 3:32
  • Andrew Moon

    Hi Christian,
    Yeah, it is pertinent. How about I say this? When Conee and Feldman distinguish between well-foundedness and justification, they appeal to cases where a belief is based on evidence and when it is not. When Pollock distinguishes between a justifiable belief and a justified belief, he does the same thing. Bergmann and a whole host of others do this when distinguishing propositional and doxastic justification. It seems that all these philosophers (externalists and internalists) are appealing to the same sort of basing relation; notice that nobody (nobody in the literature that I know of) disputes the cases. Furthermore, Lehrer, Swain, and others go on to appeal to different thought experiments to disprove different accounts of basing (causal theories, metabelief theories), and the cases tend to be agreed upon (I think… it’s been awhile).
    What can we learn from this? Given that there’s little disagreeement on the cases themselves, it seems that epistemologists have a grasp of a common property picked out by “based on”. And now I’ll say that that’s what I mean by ‘based on’. That’s what I think opponents have in mind, that’s what I am appealing to in ET, and that’s how I understand basing when I argue against ET.

    December 8, 2011 — 11:37
  • christian

    well ok andrew.
    sounds like you’re saying that you’re talking about what *they* are talking about. and here ‘they’ selects some group of esteemed epistemologists. and you seem to suggest they agree that they are using a common coin, even if it hasn’t been analyzed.
    since i don’t know the literature i will defer to you. but i would be awfully surprised, shocked, if there is some common account of basing that these esteemed epistemologists have in mind in their papers. if the past is a good guide to the future, we should expect just the opposite.
    in any case, here is a worry for your proposal (if i’m understanding it correctly): god is omnipotent. thus god can do anything possible. so god can create evidence for his beliefs, even if this evidence is redundant evidence. thus, even god can believe p on the basis of evidence.

    December 8, 2011 — 22:46
  • Andrew Moon

    Hi Christian,
    The things I was talking about regarding basing is pretty standard stuff in the literature on epistemic justification. More than just the esteemed epistemologists are involved in the discussion; rag tag, common epistemologists and graduate student epistemologists are included here.
    You conclude, “thus, even god can believe p on the basis of evidence.”
    Yeah, I’m inclined to think that God both can and does believe things on the basis of evidence. I think it’s plausible given Molinism, and I am a Molinist. (This was in Nate Shannon’s first comment.) Of course, none of that tells in favor of ET or my case against ET, the main topic of the thread.

    December 11, 2011 — 22:01
  • Christian

    Hey man,
    I guess I’m confused.
    “In this post, I want to try out a counterexample against the evidence thesis.
    Suppose God exists. (Even if God doesn’t exist, we can use God in a counterexample against the evidence thesis.) God knows all true propositions; God does not believe propositions on the basis of evidence; therefore, the evidence thesis is false.”
    I suggested God can believe propositions on the basis of evidence.
    Then you said:
    “Yeah, I’m inclined to think that God both can and does believe things on the basis of evidence. I think it’s plausible given Molinism, and I am a Molinist. (This was in Nate Shannon’s first comment.) Of course, none of that tells in favor of ET or my case against ET, the main topic of the thread.”
    But if God can (and does) believe a proposition on the basis of evidence, then the existence of God would put no pressure on ET. Or weren’t you interested in whether there might be theological reasons to resist or accept ET?
    By the way, I think this is an interesting post!

    December 12, 2011 — 18:04
  • Andrew Moon

    Christian,
    Glad you find it interesting!
    Ah, okay, I see your line of thought now. You’re right that what you say, if it were correct, would count against my claim in my opening post that God does not believe anything on the basis of evidence.
    After the Molinism point, however, I rather quickly became convinced that God does believe some things on the basis of evidence. However, I do not need the strong claim that God never believes without evidence; it only needs to be possible for God to believe (and know) without evidence; if it’s possible, then we still have a counterexample to ET. I’m inclined to think that this is possible in the area of God’s a priori knowledge, or perhaps his (in Molina’s terminology) natural knowledge (knowledge of what is possible). Okay, I think we should be connecting now. I appreciate the clarification!

    December 12, 2011 — 18:14
  • Hey Andrew,
    In your post you claim that “almost no externalist will endorse the evidence thesis”. When I read this I had the same thought as Alex. Sure, there are some externalists who will deny that we need evidence to know, but many others will just give an externalist account of evidence. In your response to Alex you suggest that an externalist account of evidence is implausible. Is this why you think that many externalist will reject the evidence thesis—because they too believe that an externalist account of evidence is implausible? As a factual matter, that just doesn’t seem to me to be the case.
    For what it’s worth, this is why it has always annoyed me that Feldman characterizes externalist views as “non-evidentialist” in his book “Epistemology”, which is otherwise a totally awesome book.

    December 13, 2011 — 18:16
  • christian

    got it. agreed!

    December 13, 2011 — 19:01
  • Andrew Moon

    Hi Dustin,
    Good to hear from you. People could refute my claim by just giving some counterexamples of paradigm externalists who would endorse the evidence thesis, but there are too few. I think those few would be people like John McDowell or Snowden. They endorse externalist accounts of evidence.
    On the other hand, look at everybody whose accounts entail the falsity of ET: Armstrong, Dretske, Goldman, Sosa, Plantinga, Nozick, Greco, and Zagzebski. (If you’re interested, see my December 6, 2011 11:39 AM and the ensuing discussion with Dani for why I think Plantinga’s theory entails the falsity of ET.) I hope that clarifies.

    December 13, 2011 — 22:08