The non-actuality of sceptical scenarios
August 11, 2009 — 9:56

Author: Alexander Pruss  Category: Existence of God  Tags: ,   Comments: 16

(Cross-posted to my blog.)

Sceptical scenarios are usually taken to raise “A how do we know that not p?” question. But let’s ignore that question. Of course, we are not brains in vats, there is no evil demon deceiving us about everything, most of our perceptual states have causes, and the world is more than five minutes old. The question how we know, or at least are justified in believing in, these facts is for the epistemologists to scratch their heads over, but we metaphysicians and natural theologians can take them for granted just as dentists and archaeologists do.

Nonetheless, there are genuinely metaphysical questions in the vicinity. Given a sceptical scenario p, we can ask: “Why is it not the case that p?” Why do we have bodies rather than just being brains, why are there no evil demons deceiving us about everything, why do at least most of our perceptual states have causes, and why did the world come into existence in inchoate form with a big bang, rather than fully-formed the way it was five minutes ago?

We can also ask the more general question: Why are all sceptical scenarios non-actual?

A theist has a fairly easy answer to the general question (essentially Descartes’ answer): God is unlikely to permit persons to be generally deceived in ways that they cannot reasonably get out of no matter how hard they try. And this answer also works for the specific questions. An anti-realist has a way of getting out of the question by arguing that no distinction can be sensibly made between p and not-p.

The realist non-theist’s best bet is to try to answer the specific questions one-by-one, because the prospects for a single answer to them all are poor. Thus, maybe, there are no evil demons because materialism is true. Our perceptual states have causes because a Causal Principle holds, either necessarily or contingently, and perhaps restricted in some way (I don’t know that there is any good way to restrict it in a way that gives causes to most of our perceptual states but does not ground a cosmological argument, but that’s a different debate). It’s harder to explain why there are so few brains in a vat. Maybe one can talk about how brain-in-a-vat scenarios are unlikely to arise apart from certain kinds of agency (people vatting other people), and how these kinds of agency are not so likely to be exercised on a vast scale (or maybe that civilizations are likely to die out before they achieve the technology to turn people into brains in a vat), so that, most likely, most people are not brains in a vat.

The five minute hypothesis presents its special difficulties. Because the entropy of the universe five minutes ago is higher than the initial entropy of the universe, there is an intuitive argument that if the universe came into existence by chance, it is more likely to have come into existence much as it was five minutes ago than much as it was at the Big Bang. That’s a tough one. Evolutionary theorists of mind can argue, however, that consciousness requires an evolutionary history, and hence it is impossible for minded beings to arise in accord with the five minute hypothesis. That probably won’t help much with the fifty million or at least one billion year hypotheses. Or one might find some good physical reason why universes need to start in a lower entropy state than the that obtained five minutes ago. But it’s a hard problem.

Can something like this be said for every sceptical scenario? Maybe, and maybe not. But notice that for a lot of these scenarios, what we will have said is something probabilistic. And unless the probabilities of non-actuality that these considerations yield for the individual scenarios are extremely high, that still leaves the question why no sceptical scenario (to do this argument in earnest, we’d need to explain this notion somewhat!) is actual. And whatever the probabilities are, the theist has a unified explanation of the apparent coincidence that for no sceptical scenario p, is it the case that p is true.

Comments:
  • Insanity is a bit of a problem for Descartes’ principle as I’ve formulated it. Probably the solution is to replace “no matter how hard they try” with “no matter how rationally they try”.

    August 11, 2009 — 10:33
  • Christian

    A theist has a fairly easy answer to the general question (essentially Descartes’ answer): God is unlikely to permit persons to be generally deceived in ways that they cannot reasonably get out of no matter how hard they try.
    Is this right? Suppose it is. I would think that if it is right, the following seems just as reasonable:
    God is unlikely to permit persons to be deceived in ways that they can reasonably get out of, if they just try hard enough.
    In general, if God exists, God is unlikely to permit persons to be deceived. But persons are constantly deceived. We are subject to deception from the senses and in reasoning. Sensory illusions are common, cognitive biases are common, and people constantly deceive each other. So, these common facts would, if we accept the claim about God above, count as reasons to think there is no such being.

    August 11, 2009 — 13:51
  • Fair enough. I do actually think that formulation would be too strong. It’s too easy to come with theodicies for minor mistakes. The claim should, rather, talk about the unlikelihood of God wanting us to be generally deceived in ways we can’t rationally dig ourselves out of.
    Actually, for my explanatory argument, I don’t even need it to be unlikely. I just need this: God has, and knows he has, a good reason to ensure we are not generally deceived in ways we can’t rationally dig ourselves out of. That’s all I need. I don’t need this reason to be conclusive.

    August 11, 2009 — 14:32
  • Gordon Knight

    why is the non-actuality of skeptical scenerios any different from the non-actuality of anything else. There has to be one way the world is (assuming realism), and this way prevents all these other ways from being real.
    Maybe its an accident that its this way, but its no more accidental with respect to brains in vats than with respec to Pluto being larger than jupiter or any other contingent state of affairs.

    August 11, 2009 — 15:12
  • Mike Almeida

    . . . there are genuinely metaphysical questions in the vicinity. Given a sceptical scenario p, we can ask: “Why is it not the case that p?” Why do we have bodies rather than just being brains, why are there no evil demons deceiving us about everything . . .
    I’m probably misreading, but it sounds a little like you’re saying (or suggesting) that the falsity of skepticism entails that we’re right about lots of things. But the fact that skepticism is false does not entail that we are not wildly mistaken in our science, metaphysics, etc. We might be (and of course have been) wildly mistaken in spite of skepticism being false. It is perfectly compatible with skepticism being false, for instance, that most of commonsense epistemology is misleading. I think it is perfectly compatible with skepticism being false that we are brains-in-vats. All that the falsity of skepticism entails, I think, is that we are not insurmountably benighted (but I’m not even sure about that, since it seems reasonable to suppose that there is a vast amount God knows that we won’t ever know). The falsity of skepticism does not even entail that we know more than someone who is a victim of skepticism. Take an envatted person who has worked out so many theorems that there are more true propositions that he knows than true propositions that we do. It does not even entail that we are right about many things (consider a non-skeptical world that includes one incurious, not very clever but nonetheless intelligent being who doesn’t know much). Perhaps there exist clues to the fact that we are envatted that we haven’t noticed or have been too epistemologically complacent to notice. Theists might note that there are non-skeptics, embedded in a theistic world, who miss altogether evidence about God’s work in the world and whose metaphysical views are radically mistaken as a result.

    August 11, 2009 — 16:16
  • Christian

    It’s too easy to come with theodicies for minor mistakes.
    But you don’t think, I take it, that it’s too easy to come up with naturalistic theodicies for why the various skeptical scenarios don’t obtain. Why is this?
    At any rate, you seem to think it likely that God would not allow people to be massively deceived, but that it is likely that he would allow people to be deceived in various ways. Where does this probability judgment come from? On what does it rest? I ask since I don’t share the assessment. It might be false as far as I can tell.
    You think that God has a good reason to ensure we are not generally deceived, in such a way that we cannot dig ourselves out of it. In a world in which “we” are brains in vats, is there are all the suffering that appears to be common? If no, that would be one reason to deceive us. Or perhaps the Jesus narrative is false, and God has a reason to make it seem true. Perhaps this involves creating a picture that seems believable so as to get persons close to him. I don’t know.
    I think it’s hard to tell a priori what God would want vis a vis the extent to which we are deceived. Here the probabilities strike me as inscrutable.
    And I agree with Mike’s comments above…

    August 11, 2009 — 20:11
  • John Turri

    Alex,
    To me it seems overwhelmingly implausible that we should expect a single, unifying explanation of why all skeptical scenarios are false. (That is, aside from the pretty straightforward sort that Gordon Knight already suggested, though that doesn’t seem to be a very deep explanation.) So I don’t think the non-theist is sacrificing much of anything here by simply conceding that she doesn’t have a unified explanation.
    I said ‘radical’ to distinguish the ones you mentioned from the numerous actual skeptical scenarios throughout history. People were deceived for ages about the relative motions of Earth and Sun, the shape of the earth, the relation of human to non-human animal life, etc. And, for the most part, no matter how hard they tried, they wouldn’t have been able to learn the truth about these things.
    Setting all that aside, I think your proposed unified explanation isn’t perfectly general. You say, “God is unlikely to permit persons to be generally deceived in ways that they cannot reasonably get out of no matter how hard they try.” Here’s a skeptical question:

    How do you know that this isn’t one of those times when God has good reason to let you be deceived in ways that you cannot get out of, no matter how (rationally) hard you try? God had good reason to allow such things as WWII, cancer, etc. Why think that he doesn’t equally well have good reason to deceive you?

    August 11, 2009 — 21:01
  • John Turri

    Oh, sorry, I guess I forgot to say ‘radical’ in the first paragraph of my previous post. I meant to say, “To me it seems overwhelmingly implausible that we should expect a single, unifying explanation of why all radical skeptical scenarios are false.”

    August 11, 2009 — 21:03
  • Mike:
    Well, it’s still, surely, true that we’re not brains in vats, that evil demons aren’t deceiving us about everything, etc. And if these things are true, it seems legitimate to ask for an explanation. I do want to think some more, however, about how exactly to characterize a sceptical scenario.
    Christian:
    To give an agential explanation for p, all I need to do is to give a subjectively (i.e., subjectively to that agent) strong reason for the agent to bring it about that p. I do not need to show that the probability of the agent bringing it about that p is high. God has a strong reason not to have me be deceived in a radical way, because then my free agency wouldn’t have the meaning it does.
    John:
    God has a strong reason not to deceive me. That explains why he isn’t in fact deceiving me. Of course, if he were deceiving me (actually, I am strongly inclined to think that God cannot deceive; but my argument doesn’t depend on this), he would have a strong reason to deceive me. But he would still have a strong reason not to deceive me, just not a conclusive reason not to deceive me.
    I am assuming, of course, that God isn’t deceiving me. I am not solving an epistemological problem, but a metaphysical one. God, in fact, isn’t deceiving me radically (why? well, I have two hands, etc.) It is legitimate to ask for an explanation of that fact.
    “So I don’t think the non-theist is sacrificing much of anything here by simply conceding that she doesn’t have a unified explanation.” When an otherwise not unreasonable competing view does have a unified explanation of a set of phenomena, one is sacrificing something by not having such an explanation oneself. For one could have opted for the competing view.

    August 12, 2009 — 7:37
  • John Turri

    Alex,
    What I meant was that your account, “God is unlikely to permit persons to be generally deceived in ways that they cannot reasonably get out of no matter how hard they try,” cannot be an explanation of why it’s not the case that God has all-things-considered conclusive reason to let you be deceived in ways that you cannot get out of, no matter how hard you try. The fact that he’s unlikely to can’t be the explanation for that.
    If you do think it counts as an explanation, then I’d ask why the non-theist can’t give an equally unified explanation along these lines:
    Given the way the world has turned out, it is extremely unlikely that such a skeptical scenario would obtain.

    August 12, 2009 — 8:21
  • Mike Almeida

    Well, it’s still, surely, true that we’re not brains in vats, that evil demons aren’t deceiving us about everything, etc.
    Alex,
    My question was how you got from the denial of skepticism to the assertion of those two claims. You’re trying to get to the conclusion that (ii) these claims are true and (iii) they need explanation from (i) the denial of skepticism. I think that’s not a good inference. To be clear, I agree with (i) and (ii), but we do not know that (ii) is true from (i). This is important for your argument, I think, since the auxilliary premises needed to get you to (ii) from (i) might be ones that non-theists reasonably deny. So, they won’t have to explain the same things you need to explain.

    August 12, 2009 — 8:32
  • I am not sure they need explanation from the denial of scepticism. For instance, I can explain them as follows:
    1. It is good that our faculties be by and large reliable.
    2. God knows (1), because (1) is true.
    3. Because of (2), God brings it about that we are not brains in vats, etc.
    I don’t need to mention scepticism at all, though of course I did in the original presentation and even in its title. Maybe it was a mistake to mention scepticism, actually.
    Another mistake I made is was to include the item about our perceptions having causes. For the explanation of that is not in God’s bringing it about that it is so, but in the Causal Principle that makes it impossible to have uncaused contingent events. This limits the scope of the unified explanation–I am only using God’s will to explain the falsity of those sceptical hypotheses that are possibly true.

    August 12, 2009 — 21:34
  • Mike Almeida

    1. It is good that our faculties be by and large reliable.
    2. God knows (1), because (1) is true.
    3. Because of (2), God brings it about that we are not brains in vats, etc.
    I don’t need to mention scepticism at all, though of course I did in the original presentation and even in its title.

    I think you’re right that you don’t have to mention skepticism. But it is difficult to compare what the theist can explain with what the non-theist can explain, since being a theist puts you in an epistemic position to explain more. Where a theist might need (and have) an explanation for why he is not a BIV–this follows from your (1) and (2) above–a non-theist might not need or have such an explanation. Is the theist’s epistemic position better? It’s hard to say, since there don’t seem to be any independent states of affairs that theism explains and non-theism does not. I think there are states of affairs that theism explains (and non-theism does not) but these are states of affairs that the non-theist (in virtue of being a non-theist) has no (or less) reason to believe obtain.

    August 13, 2009 — 8:04
  • Mike:
    This is an IBE argument. An IBE argument starts with a fact, and offers an explanation. Following out your suggestion, the atheist could get out of the argument by saying: “Yes, am not a BIV. But I have no good reason for believing this. And I only need explanations for facts that I have good reason for believing, not the fact that I believe with no good reason.”
    But that surely is most unsatisfactory for the atheist. It is already unsatisfactory to have to say: “p, but I have no good reason for believing that p.” The cost of a position that forces one to say this is high. Let’s put that aside.
    Next, note that an IBE-based argument like mine starts with a fact or a set of facts, and posits an explanation, and is based on some principle like:
    (*) If T1 explains a set F of truths better than T2, then, ceteris paribus, T1 is more likely to be true than T2.
    Now, one way to take your suggestion is to amend (*) to:
    (**) If T1 explains a set F of truths better than T2, and if I am justified in believing in every truth in F, then, ceteris paribus, T1 is more likely to be true than T2.
    But this is a weird principle. It is akin to amending the claim:
    (***) If all ravens hitherto observed are black, likely the next raven observed will be black
    to:
    (****) If all ravens hitherto observed are black, and I have good reason to believe that all ravens hitherto observed are black, likely the next raven observed will be black.
    Whether induction works well should not depend on whether I am justified in believing in the premises of the induction, but only on whether the premises are true. And ditto for IBE.
    Moreover, consider this. Suppose I have good reason to believe p, and it’s obviously true that p entails q. Suppose further that T1 better explains q than T2 does. Then surely these facts favor T1 over T2 in some important way. But of course we have good reason to believe all sorts of things that entail we’re not BIVs.
    Compare someone who responds to the First Mover argument as follows: “I accept that there is motion, and I accept the validity of the inference from motion, together with uncontroversial premises, to the existence of a First Mover. But I reject the existence of a First Mover. How do I manage this? Well, while I accept that there is motion, I do not believe myself to be justified in this acceptance.” That’s weird!

    August 13, 2009 — 8:31
  • Mike Almeida

    I understand that it is an IBE argument. My point was that for such an argument to be successful there has to be some agreement about what it is to be explained. What seems false is that there is any set of independent facts that the theist and non-theist agree need explanation.
    Now, at first I thought you were saying that there is a set of independent facts that need explanation. And those facts follow from the denial of skepticism. Since both the theist and non-theist deny skepticism, they have the same set of facts to explain. But I noted above that no such set of facts follows from the denial of skepticism, and you seemed to agree with that.
    In your penultimate post, you seemed to be saying that the set of facts that the theist needs to explain–e.g., the fact that he is not a BIV–actually depend on his acceptance of theism. So, his acceptance of theism does two things with regard to the fact that he is not a BIV. First, it assures him that he is not a BIV, and second it explains why he is not a BIV.
    Now a non-theist could respond that he does not have the same sort of assurance that he is not a BIV, and so he does not have the same set of facts to explain. So the set of facts to be explained do not seem entirely independent of one’s views on theism. Theism has epistemological implications that non-theism does not.

    August 13, 2009 — 8:54
  • I actually think it is less controversial that we’re not BIVs than that scepticism is false. And the belief that we’re not BIVs does not need to be based on the premise
    (*) If we’re BIVs, then scepticism is true.
    It could be based on the Moorean claims that we have hands, and BIVs don’t have hands.
    Do you really think there are significantly many atheists who do not agree that they are not BIVs? But if they agree, then we have an explanatory question to be asked. Ditto for the other claims. We can just take them as isolated claims, not as members of the species “sceptical hypothesis”.

    August 13, 2009 — 13:50