Knowing Your Theistic Belief is Warranted
March 3, 2012 — 15:08

Author: Andrew Moon  Category: Existence of God Religious Belief  Tags: ,   Comments: 25

Two assumptions:
A1: God exists
A2: Plantinga has successfully argued for one of his big conclusions in Warranted Christian Belief: that if God exists, then belief in God is likely to be warranted.
Now, suppose Smith has a properly basic belief that God exists. Smith has also read WCB and believes the conclusion of Plantinga’s argument on the basis of that argument. Can Smith appropriately reason as follows?

1) God exists
2) If God exists, then my belief that God exists is likely to be warranted.
3) Therefore, my belief that God exists is likely to be warranted.

Granted, there is some circularity going on here. Smith is using his belief that God exists as a premise in his reasoning to prove something about the epistemic status of his belief.
But many epistemologists have come to be okay with such circular reasoning. (For example, Mike Bergmann would probably be okay with it so long as Smith had no significant doubts about God exists or it’s not the case that he ought to have. And I don’t think anybody’s provided a good criticism of Bergmann’s defense in the literature.)
Q1) Is Smith’s reasoning okay? (granted my assumptions)
Q2) Anybody know of any literature that assesses the sort of reasoning Smith is engaging in (w/r/t theistic belief)?

  • Nate Shannon

    I don’t see how that argument/reasoning is “circular.” Premise 1 says “God exists,” not “I believe that God exists.” Given that it’s the first, I don’t see circularity, and I think that better represents Plantinga’s argument. If it were the second, as you suggest (“Smith is using his belief that God exists as a premise..”), then to draw the conclusion, you’d have to supply the premise (or grant the assumption, as you have) “God exists,” or the subject would have to know that God exists (you’d have to supply the premise “I know that God exists”). I wouldn’t call that circular, but it won’t get very far since it would have to be true (incontrovertibly) in the first place that God exists for the argument to be successful. That’s fine, but if it is true, then there’s not much point in an argument like this in the first place.

    March 3, 2012 — 17:11
  • I think I agree with Nate. It’s not circular (or to answer your question, its “ok), but it’s not terribly impressive or useful in any larger context. It would be far more convincing if one landed on the conclusion that one’s belief in God’s existence is warranted by appealing to evidence of God’s existence. Alas, that is the problem that has plagued the Christian metaphysic since the beginning: no one agrees on what should count as evidence for such existence.

    March 3, 2012 — 17:40
  • Peter

    Two assumptions:
    A1: Unicorns exists
    A2: My friend has successfully argued that if unicorns exists, then belief in unicorns is likely to be warranted.
    Can we appropriately reason as follows?
    1) Unicorns exists
    2) If Unicorns exists, then my belief that Unicorns exists is likely to be warranted.
    3) Therefore, my belief that Unicorns exists is likely to be warranted.
    I don’t see the value of this argument…

    March 4, 2012 — 2:00
  • This argument looks fine to me:
    1. God exists.
    2. If God exists, then Smith’s belief that God exists is likely to be warranted.
    3. Smith’s belief that God exists is likely to be warranted.
    But if the argument is OK for me, why would it not be OK for Smith? Granted, sometimes an argument is OK for one person but not for another, because different people have different background beliefs. But we can assume that Smith and I have no different background beliefs, and that doesn’t seem to affect things.
    The important thing is not to engage in double counting. Smith (or anybody else) had better not turn around and say:
    4. Smith’s belief that God exists is warranted.
    5. That a belief that p is warranted is evidence for p.
    6. So that Smith’s belief that God exists is warranted is evidence that God exists.

    March 4, 2012 — 9:33
  • Andrew Moon

    To Nate and Nate,
    Thanks for the comments. The argument DOES have epistemic circularity. Let me make myself more clear. (This will help me get my own thoughts more clear too!)
    Cognitive scientists of religion are converging on an agreement that there are some factors/faculties/processes that are in-built into humans (possibly cum the sorts of environments that humans find themselves in) that make it very likely that we will believe in God. Call those factors/faculties/processes X. Smith, like many humans, forms the belief that
    1) God exists
    by way of X. He then reads Plantinga. He reasons that if God exists, then God is probably using X as a means for us to form belief that God exists; in other words, God is responsible for X. If this is so, then belief that God exists has a lot going for it epistemically: it is caused by a reliable, truth-aimed process; the relevant faculties were designed and put in place by God and are performing their function and so are properly functioning; the belief is safe (in Sosa’s sense) and sensitive (in Nozick’s sense). In other words, Smith also sees that
    2) If God exists, then the processes, X, which produced my belief that God exists are reliable (and exhibit a whole other host of epistemic goods).
    Being competent in modus ponens, he concludes,
    3) The processes, X, which produced my belief that God exists are reliable (and exhibit a whole other host of epistemic goods).
    Now, here is the circularity. Smith (initially unconsciously and unknowingly) used X to form belief that God exists, which he then used as a premise in reasoning to come to believe that X is reliable. So, in Bergmann’s words, the belief is “infected with epistemic circularity”. That doesn’t mean it’s not okay; from what I can tell, the belief meets the conditions for what Bergmann calls “benign circularity”.
    (Btw, a really nice place to start on the literature on epistemic circularity is Michael Bergmann’s clearly and insightfully written paper “Epistemic Circularity: Malignant and Benign”, which won the Rutgers Young Epistemologist Prize.)
    I hope that makes clear why I think there is epistemic circularity involved. (It’s like using one’s senses to come to believe that one’s senses are reliable.)

    March 4, 2012 — 10:59
  • Andrew Moon

    Thanks for the comment. I don’t see the analogy (although I see the sort of concern you have). Maybe it’ll help if you see my response to the Nates.

    March 4, 2012 — 11:00
  • I think between them Nate and Nate have tagged it perfectly. It’s not actually circular, although it’s probably not very practically useful. It could be part of a larger circular argument in which the only reason why Smith thinks that God exists is that he thinks this belief is likely warranted. But this is not the argument we actually have. If it really is true that God exists, and if the Plantinga Principle is really true, then it’s really true that Smith’s belief that God exists is likely to be warranted, no circularity about it.
    I think this handles Peter’s worry entirely. If there really are unicorns, and if Plantinga’s Principle for Unicorns is true, then it’s a perfectly good conclusion that belief in unicorns is likely to be warranted, and there’s no circularity. You wouldn’t (necessarily) have begged the question, and you’d be a fool not to draw the conclusion given that the premises are taken to be true. Faced with such an argument, of course, we would want to know whether the reasons for believing the first premise to be true in the first place are good reasons. But that’s just ordinary workaday housekeeping with arguments. I think the oddity, as Alex seems to suggest, is nothing more than that when talking about our own beliefs we usually for practical reasons don’t reach talk about warrant from this direction; where it’s not our own beliefs that we’re talking about, the oddity vanishes. The only way there could be a difference is either background belief differences, which we can stipulate away, or if there were something Moorean about this, and it doesn’t look like there is.

    March 4, 2012 — 11:11
  • Andrew Moon

    See my response to the Nates for why there is epistemic circularity. I don’t think it matters whether we’re talking about ourselves or Smith. The problem is that I used X in order to come to believe that X is likely to be reliable.
    See my response to the Nates. I see now that having my conclusion be that belief in God is warranted does not make the circularity clear. Saying that belief in God is produced by reliable process X, which is what my belief in God depended on in the first place, makes it more clear.
    Thanks, everybody, for the responses!

    March 4, 2012 — 11:16
  • I don’t see exactly what the circularity is supposed to be. The reliability of the process isn’t a premise in the argument for X.

    March 4, 2012 — 14:45
  • Andrew Moon

    You might be thinking of “logical circularity”, which is when your conclusion is included in one of the premises of your argument. I’m talking about epistemic circularity. Here’s a quote from Bergmann: “Suppose I form a belief–either noninferentially or inferentially–in the trustworthiness of one of my belief sources, X. If, in the formation of that belief, I depend on X, then that belief is infected with epistemic circularity.” (PPR 2004, “Epistemic Circularity: Malignant and Benign”) (The definition of logical circularity is drawn from footnote 3 of that paper.)
    It’s seemed to many that a belief infected with epistemic circularity is in some way bad or unjustified. For example, using your perceptual senses to come to believe in the reliability of your perceptual senses seems bad (this would be epistemic circularity). But Bergmann (and other epistemologists, like Pryor) have been arguing that not all epistemic circularity is bad.

    March 4, 2012 — 15:32
  • Nate Shannon

    Remember that Plantinga’s project in WCB is exclusively de jure–he is not interested in de facto, in whether Christian belief is true. So WCB is essentially not an apologetic. He makes this clear in the introduction. So even though Plantinga’s entire proposal is conditional–‘if’ God exists–it does not beg the question since it does not claim to conclude anything metaphysical or theological, but only epistemological.
    That means that overall the text depends on the fact that the question of theism remains–at least in the context in which Plantinga is writing–undecided. WCB uses that as leverage and says, in essence, ‘until you prove God doesn’t exist, the charge of irrationality slips off the back of my theism like water off a duck.’
    For this reasons evangelical philosophers sometimes appear less than excited about the project, since its apologetic value is dubious: it says nothing about the existence of God, and even though it seems to carve out a slice of rational respectability for theistic belief, it carves out an equal space for any theism.
    I was just thinking that this might be what’s irking you about the line of reasoning you’re proposing. It isn’t circular because it never asserts the metaphysical premise you’ve included, “God exists,” but only says that “IF God exists, then I can reason THUS, and I am rational in believing that he does.” Then it adds, “and I believe that God exists.” No circularity, no begging the question.

    March 4, 2012 — 16:45
  • Dianelos Georgoudis

    I’d like to comment on your suggestion that Plantinga’s result about warranted theistic belief is not impressive or useful. I think it is both, for as long as theism is not conclusively falsified his result defeats (removes warrant from) the very common atheistic claim that belief in God is unwarranted. Also, please observe that no naturalist philosopher has ever derived a similar result in favor of naturalism, namely: “If naturalism is true then belief in naturalism is likely to be warranted”.
    Further, if Plantinga’s EAAN is successful then it falsifies that claim in favor of naturalism, thus establishing a concrete epistemic advantage for theism in comparison to naturalism: Whereas belief in theism is warranted if theism is true, belief in naturalism is not warranted even if naturalism is true.

    March 4, 2012 — 16:57
  • Andrew Moon

    Hi Nate,
    I never said that the line of reasoning irks me; I’m quite interested if anybody sees anything objectionable about it.
    The belief formed at the end of the reasoning is epistemically circular. See the response I just made to Alexander Pruss.

    March 4, 2012 — 18:25
  • Nate Shannon

    Thanks for your comments. I understand the overall impact of WCB just as you do, and I didn’t mean to disparage it wholesale. But from the point of view of the Christian apologist, I think there are some shortcomings. For example, do you think the Christian apologist, who is convinced of and committed to the world-defining truth of triune, biblical theism, the God-breathed nature of canonical Scripture, and the irreducibly Christ-redemptive nature of history, do you think he will find a complex “epistemic advantage” for bare “theism” a significant victory? He would say, “no one comes to the father but through me [Christ].” I think we owe it to the Christian apologist to distinguish him from defenders of an epistemic advantage for bare theism. These are different beasts fundamentally. It would become clear just how different if you ask Plantinga how the Trinity functions in his argument or his epistemology, or how his approach is according to Christ. That’s all I meant to say. I am greatly appreciative of Plantinga’s work, and, as you say, coupled with his critiques of naturalism, I’m all for it! But I think the distinction is too often overlooked by philosophers (and perhaps often overstated by theologians).

    March 4, 2012 — 20:48
  • C

    Here are two cases in which I think a similar argument is neither epistemically circular nor useless.
    1) Blaise is entirely agnostic – he has no position on the existence of God. However, he accepts Plantinga’s argument that if God exists, then belief that God exists is likely to be warranted. For other reasons, he thinks that if God does not exist, then neither belief that God exists nor belief that God doesn’t are likely to be warranted. Finally, he holds that it is better to have as many warranted beliefs as possible, and that there will be no opportunity cost in coming to believe that God exists. On this basis, he attempts to come to believe that God exists (but not in a way that would obviously exclude warrant, like having a mad scientist just insert beliefs into his brain). Is it really possible to hold that Blaise isn’t doing the rational thing without attacking one of his premises, i.e. that given his premises he’s rational to act as he does?
    I think that in this case there won’t be any circularity because the reasoning is properly hypothetical – ‘if I were to believe that God exists and he actually does, that belief would be warranted, so I’d have gained something; there are no plausible circumstances in which if I were to believe that, I’d have lost something; so …’
    2) Alvin is watching the news. He sees a bunch of stories about apparent miracles in Albania, which he takes as strong evidence for the existence of God. However, he also sees stories that in his location there is a contagious disease going around that causes unwarranted belief in the existence of God. He also accepts Plantinga’s argument that if God exists, then belief that this is so is likely warranted. Following the argument above, Alvin can’t rationally come to believe both that it is likely that God exists and that it is likely that he has the disease.
    In this case, there would be a sort of modus tollens argument from the high likelihood of unwarranted belief to the non-existence of God. It would be a lot like an argument from evil – if the world is like that, it can’t have been created by God.

    March 5, 2012 — 3:40
  • C

    I should have been more careful in that last comment. If the disease in case 2 is not very widespread, then it is not in general unlikely that beliefs that God exists will be unwarranted. (The argument in the original post – in which it is specifically my beliefs that are concluded to be likely to be warranted – should probably have a ‘so long as specific personal circumstances do not make my beliefs less than likely to be warranted’ clause.) The case can be amended to make the disease more widespread, though. Maybe scientists have discovered that almost all present beliefs are probably induced by the disease, and its carriers have been found in ancient bones all over the world, etc..

    March 5, 2012 — 5:56
  • C

    Lastly, I wonder if it might matter that Smith doesn’t know what X is. He only knows that whatever led him to his belief is enough to provide warrant, not that X is enough.
    Imagine that Jones is excellent at judging people’s standing heights using her vision – so excellent that any of her beliefs about heights produced by vision in normal circumstances amount to knowledge. One day, she hits her head and loses a lot of memories, so that she often finds afterwards that she has beliefs whose provenance she can’t remember. She sees a whole bunch of people sitting down, and finds that she has some beliefs about their standing heights – that they are all exactly 6′, for instance – but doesn’t remember them or how she came to know their heights. When she sees them stand up, she judges that they are all 6′ just by looking, and so comes to know that they are this height. She thinks that whatever process she used to first learn (before her memory loss) about their heights must have been very reliable, since it achieved very many very precise correct answers. She doesn’t know what it was, though – yes, maybe it was by looking, but perhaps she saw them all at the exclusive exactly-6-foot club, or she measured them with a tape at one of those measure-each-others’-heights parties she used to go to. Who knows. But she does come to know (or at least have a fairly well-justified belief) that, whatever the method was, it was reliable.
    I think that even if that original method turns out to have been the use of her vision, there is no epistemic circularity in her reasoning. She won’t have learned much of use, but that’s beside the point. I can see that this is not entirely analogous to the case of Smith – for instance, Jones’ reasoning isn’t deductive. However, the point is that if Smith doesn’t come to the de dicto belief that X is reliable from use of X, but instead just comes to the belief that the process of his coming to belief in God, which turns out but isn’t known by him to be X, is reliable from use of X, then it may be that there isn’t any epistemic circularity involved.

    March 5, 2012 — 6:44
  • Andrew:
    I grant you that by that definition of epistemic circularity it’s epistemically circular. That’s a tautology. But I have yet to be convinced that such “epistemic circularity” is actually a circularity.
    We can know all sorts of things about a faculty simply by using the faculty. For instance, I look at my eyes (with the help of a mirror) and notice that they have pupils. Here I used a faculty X to arrive at a piece of information about faculty X, without there being any circularity. So the mere fact that I used a faculty to arrive at information about that faculty does not imply circularity of any sort.
    Why is the case where the information I arrive at is that the faculty is reliable different in that respect? Well, if one thinks that the reliability of a faculty is an implicit premise in the argument, then that’s obvious. But we’re assuming in this discussion that it’s not. (Actually, on reflection I am kind of open to the idea that it implicitly is, and in that case we do have logical circularity.)
    It’s tempting to say something like: “Well, the argument has this property. If its conclusion were false, at least one of the premises would be false or unwarranted. And any argument that has this property is circular.” But that’s false.
    I don’t want to say that there isn’t some sort of circularity. Maybe there is. But it’s not clear to me.

    March 5, 2012 — 15:55
  • C:
    What kind of evidence does Alvin, or the experts he is relying on, have that the belief in God coming from the disease is unwarranted?
    There are two hypotheses that would need to be ruled out:
    a. Belief in God coming from the disease is overdetermined by the disease and by some properly functioning belief-formation process.
    b. God intentionally uses the disease as his vehicle for inducing in us the belief that he exists, and this is a non-aberrant truth-directed belief-formation process.
    In defense of b, I imagine a society that directly imprints a full body of mathematical and scientific truth in the brain of every twelve-year-old, in order to save the expense of schooling. I am inclined to think the twelve-year-old knows the mathematical and scientific truths if the process can be tracked back to someone who knew them in some more fundamental way.

    March 5, 2012 — 16:03
  • Andrew Moon

    Well, Bergmann’s definition is at least pretty close to what most epistemologists accept. I care less about how we’re using ‘circular’ and more about whether beliefs that are epistemically circular are in some way unjustified in virtue of being epistemically circular (as epistemologists define the expression). And it looks like you don’t think a belief’s any worse for being epistemically circular. I and Bergmann and others agree.
    But there is still a felt intuition by many that something is wrong. How do you know that Fred is a reliable testifier? Because Fred told me so! How do you know your sense perception is reliable? Because the deliverances of my sense perception lead me to think that it is reliable. Maybe these cases of epistemically circular reasoning doesn’t seem bad to you.
    There are obviously bad cases. You are wondering whether Fred is an honest person. You go ask Fred, “Hey, are you an honest person?” He says, “Yes”. “Ah, thanks for allaying my worries!”
    Bergmann’s strategy is to distinguish benign vs. malignant circularity and then to say that malignant cases are when you try to use a source X to justify the reliability of your source X when you are already doubtful (or ought to be doubtful) about the reliability of X. (They’re benign otherwise.)

    March 5, 2012 — 20:27
  • C

    I accept that, as the case is described, Alvin would need to rule out (a) and (b), though I think I could just as well add to the description that Alvin believes, for whatever reason, that beliefs caused by the disease are unwarranted (or that they are sufficiently likely to be unwarranted). I was just looking for a case where the argument from the original post actually does some work for someone. How Alvin comes by his premises isn’t really important.
    On (a), I’m tempted to think that e.g. a paranoid person can never be warranted in believing that someone is out to get them, no matter how much evidence they collect, if they’d believe that, and believe they had evidence for that, whatever their actual evidence amounted to. But I acknowledge that the disease doesn’t make warranted belief in God impossible merely by virtue of being a disease.
    On (b), this just sounds a bit too sneaky, though I suppose it’s not really my place to judge. As for the 12-year-olds – I wouldn’t like to attribute them knowledge unless they can rationally re-assess their beliefs (if they can, I can see them having knowledge of all the mathematics, and maybe the rest too if they either know the provenance of their beliefs in trustworthy adults or reason from the fact that they have so much innate mathematical knowledge to the likelihood of the truth of their other innate beliefs). In the belief-disease case, if it produces paranoia-like delusions rather than ordinary beliefs, then it plausibly doesn’t allow for reflection on the belief produced and so it can’t produce knowledge.
    If it doesn’t produce delusions and allows for reflection, I don’t really see that people who catch the disease can gain warrant over those who don’t. On reflection, they won’t have any new information. I think it’s difficult to have gained warrant just by coming to believe through a process (like catching the disease, but unlike e.g. using the senses) that appears to you to have no epistemic effect – no introduction of new evidence or connection to the facts – other than producing that belief. That goes for the 12-year-olds too, if they don’t know where the beliefs come from and why, except for the mathematical beliefs which are probably warranted if and as soon as they are understood.

    March 5, 2012 — 22:08
  • Dianelos Georgoudis

    If you want to help others then convincing them of the truth of theism will give them 90% of the benefit anyway. After all Christ is not distinct from God, so one who believes in God also believes in Christ, and one who walks towards God is following Christ’s path. Let’s not confuse the external signs with the internal repentance. “Bare theism”, as you put it, is the great treasure.
    My main argument though is methodological. Trying to convince an atheist of the truth of Christianity is like trying to explain to somebody who does not believe that cars exist why it is that one is driving a Toyota. You must first convince the atheist that theism makes more sense than naturalism, and then perhaps draw her towards Christianity. And to draw somebody towards Christianity is not I think a matter of argumentation, but a matter of becoming oneself a testament of Christ. It’s not really arguments, nor indeed the Bible or the church, which draws people towards Christ, but Christ Himself. So if you want to draw people towards Christ then you must become like Christ.

    March 7, 2012 — 3:18
  • Rizzieri

    The answer to this question might depend on how seriously the subject should take the possibility that God does not exist from a subjective point of view. Consider the following response to Cartesian skepticism:
    (1) The external world exists.
    (2) If the external world exists, I am likely to be warranted in believing that the external world exists.
    (c) I am likely to be warranted in believing that the external world exists.
    We do not typically take the skeptical alternative to be a relevant alternative and hence we are (or at least take ourselves to be) entitled to premise 1. If at least some people are equally entitled to “God exists”, then those people should be able to use the argument that you employ in order to draw the conclusion that you specify. However, I think that the more educated and aware one becomes the harder it is to be entitled to your (1), because it is easy to think of defeaters to theistic belief that one ought to both take seriously and can not rule out. If knowledge (or at least knowledge level internal justification) of a proposition is the norm for employing that proposition in one’s reasoning, then one would have to know (1) in order to give the argument you give.

    March 7, 2012 — 16:25
  • Suppose an oracle that I know for certain to be perfectly reliable reveals to me that all the true beliefs that I will ever have on subject matter S will have some property Q.
    Suppose that I know the conjunction: p is true and I believe p and p is on subject matter S. Then when I reflect on what I also know from the oracle, it seems I am also in a position to know that p has Q.
    For instance, let Q be the property “being known to Smith” and let S be logic. Then when I learn that p is true and I believe p and p is about logic, I also learn that that Smith knows p. There is nothing outlandish or weird about situations like this. We do sometimes say to ourselves things like: “If this is true, Smith will know it.”
    There is no problematic circularity in my argument:
    1. If p is true, believed by me and about logic, the Smith knows p. (By the oracle’s testimony)
    2. p is true
    3. I believe p and p is about logic
    4. So, Smith knows p.
    But now suppose that I find out that Smith is me, that “Smith” is just the oracle’s idiosyncratic name for me. That doesn’t defeat any part of the above reasoning. So I get to conclude:
    5. So, I know p.
    And if I have very good reason to accept an epistemology on which knowledge requires warrant, I also get to conclude:
    6. So, I have warrant for p.
    The argument 1-4 isn’t circular in any problematic way when Smith isn’t. In fact, I don’t see how it is circular at all. And I don’t see how adding that Smith is me makes it problematically circular, or even circular at all.
    Another variant. The oracle reveals to me the somewhat depressing truth that if I believe any truth about logic, then that truth is known to every logician. Then when I know that p is true, believed by me and about logic, I get to conclude that every logician knows p. If I know I am a logician, I should be able to conclude that I know p, and hence that I am warranted.

    March 9, 2012 — 7:55
  • Eric Sotnak

    It is not clear that Smith can reason legitimately as suggested because it is not clear that Smith has good reason to accept premise 2. It may be that Plantinga has provided a successful argument for premise 2, but it isn’t stated that Smith accepts premise 2 on the basis of Plantinga’s argument. Nor, even if he does, is it stipulated that Smith has properly understood Plantinga’s argument. What if Smith is a philosophical novice who uncritically accepts theism and accepts the argument only because he has been told by a Christian apologist that Plantinga has shown the absurdity of atheist charges that theism is irrational?
    By analogy, assume:
    (a1) The Big Bang happened spontaneously without a theistic cause.
    (a2) Stephen Hawking has successfully argued for a model of quantum gravity on which it is not unlikely that the Big Bang happened spontaneously without a theistic cause.
    So now Smith argues:
    (p1) The Big Bang happened spontaneously without a theistic cause.
    (p2) If the Big Bang happened spontaneously without a theistic cause, then it is not unlikely that a scientific model exists on which the Big Bang happens spontaneously without a theistic cause.
    (p3) Therefore, it is not unlikely that a scientific model exists on which the Big Bang happens spontaneously without a theistic cause.
    Before determining whether or not Smith’s argument is ill or well-founded, I would like to know much more about Smith’s reasons for accepting each premise.

    March 9, 2012 — 19:27