Tuesday, December 02, 2014

"I Think There's About a 99.8% Chance That You Exist" Said the Skeptic

Alone in my office, it can seem reasonable to me to have only about a 99% to 99.9% credence that the world is more or less how I think it is, while reserving the remaining 0.1% to 1% credence for the possibility that some radically skeptical scenario obtains (such as that this is a dream or that I'm in a short term sim).

But in public... hm. It seems an odd thing to say aloud to someone else! The question rises acutely as I prepare to give a talk on 1% Skepticism at University of Miami this Friday. Can I face an audience and say, "Well, I think there's a small chance that I'm dreaming right now"? Such an utterance seems even stranger than the run-of-the-mill strangeness of dream skepticism in solitary moments.

I've tried it on my teenage son. He knows my arguments for 1% skepticism. One day, driving him to school, a propos of nothing, I said, "I'm almost certain that you exist." A joke, of course. How could he have heard it, or how could I have meant it, in any other way?

One possible source of strangeness is this: My audience knows that they are not just my dream-figures. So it's tempting to say that in some sense they know that my doubts are misplaced.

But in non-skeptical cases, we can view people as reasonable in having non-zero credence in propositions we know to be false, if we recognize an informational asymmetry. The blackjack dealer who knows she has a 20 doesn't think the player a fool for standing on a 19. Even if the dealer sincerely tells the player she has a 20, she might think the player reasonable to say he has some doubt about the truth of the dealer's testimony. So why do radically skeptical cases seem different?

One possible clue is this: It doesn't seem wrong in quite the same way to say "I think that we might all be part of a short-term sim". Being together in skeptical doubt seems fine -- in the right context, it might even be kind of friendly, kind of fun.

Maybe, then, the issue is a matter of respect -- a matter of treating one's interlocutor as an equal partner, metaphysically and epistemically? There's something offensive, perhaps, or inegalitarian, or oppressive, or silencing, about saying "I know for sure that I exist, but I have some doubts about whether you do".

I feel the problem most keenly in the presence of the people I love. I can't doubt that we are in this world together. It seems wrong -- merely a pose, possibly an offensive pose -- to say to my seriously ill father, in seeming sincerity at the end of a philosophical discussion about death and God, "I think there's a 99.8% chance that you exist". It throws a wall up between us.

Or can it be done in a different way? Maybe I could say: "Here, you should doubt me. And I too will doubt you, just a tiny bit, so we are doubting together. Very likely, the world exists just as we think it does; or even if it doesn't, even if nothing exists beyond this room, still I am more sure of you than I am of almost anything else."

There is a risk in radical skepticism, a risk that I will doubt others dismissively or disrespectfully, alienating myself from them. But I believe that this risk can be managed, maybe even reversed: In confessing my skepticism to you, I make myself vulnerable. I show you my weird, nerdy doubts, which you might laugh at, or dismiss, or join me in. If you join me, or even just engage me seriously, we will have connected in a way that I treasure.


  1. Well, what do you think the chances are that you're surrounded by disguised gods or angels? Maybe you could lead with that if you're worried about being rude.

    I think that there's a difference between ordinary unlikely events and skeptical hypotheses in that probabilities aren't connected to practice for skeptical hypotheses. Even if you think it's more likely that you're a brain in a vat than it is that Yale will win the NCAA basketball tournament or that an 8.0 earthquake will strike Indiana next year, the second two possibilities are connected to practical plans (betting, insurance) in a way that the first one isn't. Even if you think that there's a 5% chance that you're a brain in a vat, it's a negligible possibility because there's nothing that's worth doing about it.

  2. Isn't there some degree of possibility (from my own point of view!) that I'm a figment of the imagination of a speaker at whose talk I am in attendance? After all, if your disembodied mind is all that exists in the universe, then how do we know what it feels like to be one of the human-like characters in your delusion?

    (I may or may not be kidding)

  3. How would your skeptical scenarios compare with the scenario that only I am conscious and everybody else is a machine? which though seemingly radical is probably fairly common when you superimpose radical subjectivism with metaphysical determinism

  4. 1% or 0.1% because we have 10 fingers? I am stuck on the numbers but get the concept.

  5. I think this is just a matter of delivering appropriate-to-the-audience levels of skepticism.

    Start out with little bread crumbs like being skeptical about whether our politicians accept bribes.

    Then you can go on to mid-range skepticisms like being skeptical that free energy will never be created. Then maybe they will be gapped appropriately to deal with such things as their own inexistence.

    Overall, though, getting someone to think about their own inexistence is not polite in certain company.

    As to skepticism, a garden path has to do some leading or it isn't gardening.

  6. Would you have any ethical obligations to dream phantasms? To simulations? Might it be natural to then hear/fear the claim as saying: "I think there's about a 99.8% chance that I have ethical obligations to you." This seems a bit offensive - I might wonder why you're doubting whether you have to treat me morally. This isn't the letter of what you've expressed, but if I'm suspicious about why you're wondering whether I exist, etc... This might be one way to explain the strangeness or worries about it being somehow disrespectful or offensive to express this doubt rather than a global one (the global seems to fit more neatly and squarely in demonstrating that you've got something in the domain of epistemological skeptical doubts on the mind.)

  7. If you join me, or even just engage me seriously, we will have connected in a way that I treasure.

    I am 99.8% sure you would treasure it, Eric......

    So I guess it comes down to commitments. Where you simply commit to timeline X, instead of thinking it's a 99.8% chance timeline X is the case rather than timeline Y (or Z, or A...). Otherwise, as you say, you throw up walls. Or cast down mutual commitment, anyway.

    What is 1% skepticism, anyway - if you walk past a wishing well, roughly every one hundreth passing will you throw in a coin? As representing your 1% skepticism about it being useless to do so? If you don't, are you actually commited? Or you never got to your thousandth passing, or millionth? Whatever it takes for the coin to pass to water.

    On the flip side, it kind of seems like there's a 1% chance of simply recinding commitment - much like Anon says above, as in not having to treat in a moral way (some sort of moral way). Like as if we were all just suddenly animatronics. So the 1% could be of a massive change in commitment for the rest of your life. That seems bigger than 1%! 1% seems dangerous, that way.

  8. Thanks for all the interesting comments, folks!

    Mike J.: I'm trying to get away from that general approach to skepticism in the larger essay I'm writing on this topic. Unlike brain-in-vat skepticism or I'm-surrounded-by-gods skepticism, I think there *are* grounds for doubt and for very rough probability assessments for dream skepticism, simulation skepticism, and cosmological skepticism. I also think that there are practical consequences of taking these skepticisms seriously at about a 0.1%-1% credence level. Of course, none of that is in this post!

  9. Eric: I might agree, but my credence level in that is much lower than my credence in dream skepticism or sim skepticism, since there aren't really much positive grounds for thinking it.

  10. Howard: As in my responses to Mike and Eric, the issue here -- in my mind -- is that that sort of possibility doesn't have much evidence in its favor. The dream possibility and the sim possibility, in contrast, we do seem to have some evidence that provides non-trivial grounds for doubt.

  11. Anon 10:11: Well, I wasn't trying to address that here! See the linked paper or this post:

  12. paloalto: Based on your comment and others', I fear that I didn't set up this post very well. I wasn't trying to convince the reader to accept 1% skepticism. (I address that in the linked posts and papers.) I do try a bit of a garden path argument, though it's set up pretty differently, in terms of considering the fallibility of the arguments *against* dream and sim and cosmological skepticisms.

  13. Anon 02:00: Very interesting thought! Yes, I think that might be part of what's going on, and maybe too part of why the way of approaching things that I recommend at the end works as an alternative, if it works.

  14. Callan: I'm not sure about the relationship between belief and commitment -- either commitment to action or intellectual commitment. That's a pretty tricky issue! As you suggest at the end, even commitment is rarely entirely-unrescindable-come-whatever-may commitment; so maybe commitment, really, turns out to be scalar, too.

  15. I guess I see the practical consequences of ordinary dream skepticism. You could do those tests that they talk about in Waking Life such as trying to read small print or seeing if light switches work as steps towards lucid dreaming. I don't see practical consequences to the possibility of an indiscernible dream or, so far as I can tell, to simulation skepticism or cosmological skepticism. Are you thinking that you'll go looking for red pills or something like that?

    It's not very rude to tell people that you assign a .2% chance to the possibility that they don't exist. Non-philosophers will just be confused and philosophers will understand. It would be ruder to tell people the odds that you think that they've committed a felony or the odds that they aren't competent at their jobs. (Maybe that's interesting, too.)

  16. People hear a confession of unreliability as well as witness a propensity to use tools for jobs they simply aren't designed to solve. This is the impression I get from my family at least!

    Why should good skepticism be bad PR unless doubts, like beliefs, are not so much things we 'have' as tools we use to communicate?

  17. Mike J.: Well maybe not. My sense is that there's a lot of variability here.

    On sim skepticism, one practical consequence I see is this. If this is a sim, there's a reasonable chance that it will be over soon. Sims *might* almost always be very stable; but I think it's also quite possible that many sims are small or short-term. And that chance should very slightly tilt me toward short-term gains over long-term ones.

  18. Scott: "a propensity to use tools for jobs they simply aren't designed to solve" -- nicely put. Having this propensity might be a necessary condition of being a philosopher!

  19. I guess it's also true that the more likely it is that one of these hypotheses is true, the more rational it is to be selfish.

    So try not to be too persuasive.

  20. Eric, there's possibly an element to this that's like going from an automatic to a manual transmission.

    Like what is the set duration of the commitment? The duration before a reconsideration occurs that may cancel any further commitment on that matter? Well, you can make a call on that duration - days, weeks, years, etc. Wheras just having 1% skepticism - well, that doesn't commit to any time period. So what is that? Atleast what is that in regard to a tribal/group orientated species who rely on co-operation (ie, commitments) to work as a group.

    You could say the problem your original post describes is a bit like failing to use the clutch properly, at the social level. Just not used to a manual! :)

  21. Quick side note: If it's a sim, what gain is there to be had, anyway?

  22. Damn, must have been a simulation after all!

  23. I agree with the ethical worry. But I think there is a genuine problem with the way you've written the numbers. 1% is actually a very substantial chance. If there's a 1% chance of something happening on any given day, then it's going to happen several times a year. But our reality doesn't crash around us several times a year.
    So it's not a 1% chance, nor a 0.1% chance. If anything, it's a 0.00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000001% chance. Now, often when I say things like that to philosophers, they just look at me as if I'm dumb, and say, yeah, small number, so what? But these differences are important because they can determine whether a thing is fundamentally part of human experience or not part of human experience. Even more fundamentally, who knows if there is such a thing as a quantum of probability? If there is, then very small chances (like the Bayesian probability that we're brains in vats) might be below the quantum level, i.e. literally equivalent to zero.

  24. Sorry about the slow reply, folks! Catching up from my Miami trip.

    Callan: Yes, I agree there's a duration issue here, and what durations seem most plausible depend on the details of the skeptical scenarios (Boltzmann brains: short; sims possibly much longer). In a sim there are at least short-term psychological goods like pleasure.

    Chinaphil: Wait, how do you know that there aren't such crashes? I agree that it can make sense to have such tiny credences, but I'd disagree that the skeptical sim and dream credences should be anything like that low. The basis of your high confidence is part of what the sim and dream possibilities put into doubt!

  25. By "crashes" I meant events which would reveal to us that we're not what we think we are and are in fact in a sim/game/vat/whatever.

    If you are suggesting that no such events will occur, and we'll live our entire lives as brains in vats and never know it, i.e. there is literally no empirical difference between the real life with think we have and vat life, then I think the vat hypothesis becomes pointless. In effect, we're already brains in vats: if quantum mechanics is to be believed, we live in a world of probability waves (and maybe multiple universes), but that's not what we perceive. The world *is* radically different from our naive perception of it; but so what? Matrix/vat hypotheses only become interesting at the point where Neo discovers the glitches and breaks out.

    So I'm talking about vat situations which are in principle empirically discoverable, and for those the Bayesian probability is extremely low, I think.

    I like your next post very much!

  26. I agree about the Matrix point, chinaphil (David Chalmers argues similarly in his paper on the Matrix), so the more relevant case is whether the memories are of events that never occurred *either* in base-level or in Matrix-level reality, as in a sims version of Russell's world-created-5-minutes-ago scenario.

  27. I can't quite see how the world being created five minutes ago would be ethically different from it being created before my birth, or before the birth of anyone living, or before humankind evolved, or at the time of the big bang.

    Assuming that the sim is created so perfectly that we can't tell... how could it affect us morally? It would mean that I didn't get married in 2006 - but all the events surrounding the marriage are remembered by me and others exactly as if they'd really happened. Those events obeyed (in our fictional memories) the same physical and ethical rules as exist now. I think that's the point: if the faked memories are good enough that we can't tell, then by definition all ethical rules are the same in our memories as in our reality. So I'm not sure that this possibility manages to leak out of the metaphysical into the moral/ethical sphere.

  28. Hello Eric,

    One of the things I do when I teach skepticism is distinguish two hypotheses. One in which you are a brain in a vat, your experienced controlled by another, alien force or being. The world is an illusion, but epistemically its as real as whatever is happening now. There is a huge contrast between this hypothesis and what goes on in, e.g., the Matrix movies. In the matrix, the physical world (so called) is illusionary, but the people are real and the interaction between them are real.

    As far as skepticism. what matters is not,"Do I have reason to think this chair have absolute, mind independent existence?" What matters is whether there are other minds with which we share a world (mind independent or not)

  29. Hi Gordon -- Sorry for the slow comment approval! I fell behind with the holidays. Anyhow, I agree. But there are skeptical versions of the Matrix scenario in which the world is brief or small. I regard *only* such versions as true skeptical versions. Then the question is conditionally upon Matrix-like cases, what proportion of my credence should be on large stable sims vs small or unstable sims. (See also my post on Godzilla.)