Descartes seeks certainty.  He wants to know, of what can he be absolutely certain?  To answer that question, he confronts three successive skeptical arguments, in order to see whether any of his beliefs—everyday or scientific--can resist skeptical doubt.

The Argument from Sense Perception

    The skeptical interlocutor points out that our senses sometimes deceive us, and that we should never entirely trust a source of information that has even once proved unreliable.

    Descartes replies that, so far as has been shown, we get deceived by our senses only when viewing conditions are poor, or our sense organs are impaired, or we are careless.  So long as we’re careful and make sure to trust our senses only under optimal conditions, we can still have certainty in empirical matters.

The Dream Argument

The skeptic offers a more comprehensive reason for doubting the deliverances of sense perception:

           1.  Some dream experience is, qualitatively, just like waking experience.

         \  2.  At any given time, I have no test or criterion for telling whether I’m awake or dreaming.  [1]

         \  3.  At a given time, I cannot tell whether I’m awake or dreaming.  [2]

         \  4.  At a given time, it is (for me) possible that I am dreaming.  [3]

            5.  If I am dreaming, then (normally) my specific experiential thoughts and beliefs are mistaken.

         \  6.  At a given time, it is (for me) possible that I am dreaming and my specific experiential thoughts and beliefs are mistaken.  [4,5]

    But
            7.  If I know X, then in believing X I cannot be mistaken.
            __________________________________________________

         \  8.  None of my specific experiential thoughts and beliefs is (an item of) knowledge.   [6,7]

--and so I do not know anything based on my present sensory experience, or for that matter on any of my past experience.  This wipes out all empirical knowledge.

    Several objections were made in class.

    Objection 1:  Premise (1) is false.  Dream experience is qualitatively different from real, waking experience.  For example: In a dream I’m only a spectator, with no control over what happens; in a dream I can’t feel bodily sensations, such as pain or pressure on my skin; or, I dream only in black and white.  Reply:  Sure, that’s a difference, but how do you know it isn’t just a difference between two kinds of dreams?

    Objection 2:  Step (3) doesn’t follow from (2).  It’s possible that we can tell something without there being any particular test or criterion that we use to tell it.  Reply:  Yes, I suppose.  But we’d have to hear more; the objector can’t just insist that he damn well does know he is awake—because he has, after all, sometimes been sure he was awake when he was not.

    Objection 3:  Premise (7) is ambiguous; there are two different things it could mean.  So we can’t say whether it’s true until we determine which of them it is supposed to mean.
    (7) asserts a connection between knowing and the impossibility of error.  But what connection?
    We can’t know what isn’t so.  So one thing that might be meant by (7) is just that:

                    (7A)  Not possible: [I know X and I am mistaken in believing X].

But the grammar of (7) suggests a different reading:

                    (7B)  If I know X, then: [Not possible: I am mistaken in believing X].

To see the difference, consider a parallel case.  The sentence, “If I am a vegetarian, then I cannot eat meat” shows the same ambiguity:

                    (A*)  Not possible: [I am a vegetarian and I eat meat].

                    (B*)  If I am a vegetarian, then: [Not possible: I eat meat].

(A*) is an obvious triviality, true by definition.  Not so (B*); (B*) says that if I happen to be a vegetarian, then I could not have eaten meat—there is not even a possible scenario in which I eat meat.  And that’s false.  Even if I am in fact a vegetarian, it might instead have been the case that I was not a vegetarian and ate meat; I could have eaten meat even though I don’t.  (B*) is, then, very different from (A*).
    Similarly, (7A) is an obvious triviality, true by definition.  But (7B) is a much more substantive and ambitious assertion.  (7B) says that if I happen to know X, then I could not have been mistaken in believing X—there is not even a possible scenario in which I believe X but am mistaken.  (7B) differs from (7A) in the same way (B*) differs from (A*).
    Which is needed for the argument?  If we add (7A) to (6), (8) does not follow.  For (6) says only that it’s possible I am dreaming and mistaken, not that I actually am mistaken, which is what we would need to get (8) from (7A).  On this interpretation, the Dream Argument is invalid.  (7B), on the other hand, makes the inference straightforwardly valid.
    But is (7B) true?  Not at all obviously so.  Virtually none of the things I think I know—my name, my daughter’s name, the back of my hand, that lecture is on Mondays and Wednesdays, etc.—satisfies the strong condition laid down in (7B).  I could, just imaginably, be wrong about any of them, even though I have no reason in the world to suppose that I am wrong.  (You could write a short story in which a series of weird circumstances resulted in my being mistaken about my own child’s name.)  So (7B) seems false.
    The Dream skeptic might insist that (7B) damn well is true and therefore I really don’t know all those everyday things.  But notice two points:  (i) If (7B) is true, contrary to appearances, we don’t need the rest of the argument, or any reference to dreaming at all.  (7B) alone would prove that I have no empirical knowledge.  We already know that nearly any empirical belief is such that I might be mistaken in holding it (exceptions are Descartes’ “I exist,” “I am thinking,” “I have at least one belief,” etc.).  (ii) No one who needed convincing of (8) would grant (7B) without some further defense.  That is, anyone who initially thinks that I do know my own name, the back of my hand etc. will reject (7B) or at least demand some further argument for it.  So even if (7B) is (in reality) true, the argument will still not convince anyone who doesn’t already accept the conclusion.

    When an argument has these characteristics ((i) and (ii)), we say it begs the question.  That is, it simply assumes something that is as near as matters equivalent to the conclusion itself.

The Evil Genius

    The hypothesis that you are really only dreaming is what philosophers call a “skeptical possibility”; the philosopher challenges you to rule it out somehow or otherwise justify your contrary belief.  The Evil Genius hypothesis is another such skeptical possibility, as are the brain-in-vat hypothesis and various virtual-reality hypotheses.  The difference between the Dream hypothesis and the Evil Genius hypothesis is only that the latter cuts deeper, threatening your most basic concepts and even your deductive reasoning ability.  The Evil Genius could not only feed you spurious experiences but mess directly with your cognitive powers, making an inference seem valid to you when in fact it’s not.

    The Evil Genius Argument (which I won’t spell out here) goes much as the Dream Argument did, and appeals to the same principle (7) connecting knowing with possibility.  But its conclusion is more ambitious: that I have no knowledge whatever, let alone empirical knowledge.

    Of course, I would make the same objection to (7) as I did in the preceding section.

The upshot

    If the Dream Argument and the Evil Genius Argument are intended as proofs of skepticism, they fail, because each of them begs the question.  But they have succeeded at a less ambitious task:  By reminding us that our empirical beliefs are not absolutely guaranteed of truth, the arguments raise the question that concerned Descartes himself, viz., “If we do know ordinary empirical things even though the testimony of our senses does not absolutely, logically guarantee the truth of our empirical beliefs, how do we do that?”  Notice that that is a challenging question even if we are firmly convinced and never doubt that we do know the ordinary empirical things.  Descartes’ own answer to it is elaborate, involving the cogito, “clear and distinct ideas,” proofs of the existence of God, a bit of theology, and more.  I don’t think you’d buy it; no one else ever has.  But what’s your answer?

    There is also the deeper question I voiced in class:  Never mind absolute guarantees, certainty and such; given the skeptical hypotheses, how are you at all justified in holding the beliefs you do?   Your experiences may be produced in you in the way you think they are, and your reasoning may be sound; but your experiences may also be being fed you by an Evil Genius, and your reasoning may be being messed up by same.  It seems 50-50.  So why do you have reason to believe anything at all?

Peirce

    I said that the Dream and Evil Genius Arguments fail because (7B) begs the question.  Yet we should not simply write off those arguments, because people feel the attraction of (7B), even if they then do reject it as question-begging.  That attraction still needs to be explained.  That’s where Peirce comes in.  He in effect admits that (7B) contains some truth, indeed important truth; it’s just that that truth has been exaggerated.

    He distinguishes between “real and living doubt” (p. 291) and merely made-up, meretricious “doubt,” that is “idle.”  This corresponds to a distinction between two kinds of possibility, what we may call “real” possibility and merely made-up, fanciful possibility.  A real possibility is one which we have some actual reason to think may well obtain.  As I write this, I believe that my wife is at home, because she usually is at this time of day.  But often she does go out, so it is a real possibility for me that she is out.  It is not a real possibility for me that my wife has declared Afghan citizenship and taken up life as a secret agent against Russia, or that she has flapped her arms and flown to the moon, or that I am being deceived by an Evil Genius into thinking I have a wife at all.

    Getting back to (7B):  What’s true, Peirce grants, is that if I know X, then it’s not (for me) a real possibility that I’m mistaken in believing X.  That is, if I have some actual reason to think I may well be mistaken, I am not certain and do not know.  For example, I do not know that my wife is home now.  She often does go out; it’s not just that I can make up some entirely fanciful scenario about her being spirited away by invisible aliens from the planet Mongo.

    Contrast the case of ordinary empirical knowledge, such as my present knowledge that I am sitting in front of a computer typing words into it.  I haven’t the slightest reason to suspect that I’m only dreaming, or that my experiences are really being produced in me by an Evil Genius.  Those possibilities (while possible) are “idle,” not real possibilities.  So my inability to rule them out is normal, unthreatening, and perfectly consistent with my knowing and being certain that I am sitting at my computer and typing.

    The skeptic will freely grant that, sure, sure, “for practical purposes” I am certain, but point out that what’s good enough for the purposes of the moment may not withstand real scrutiny.  Peirce would respond that certainty for practical purposes is all the certainty there is.  The point and function of a belief, a belief’s whole raison d’etre, is to help you get around the world and fulfill your goals.  That’s all there is to it.  If a particular belief does thus help you, and nothing has caused you any doubt about it, then that belief is certain for you in the only sense of “certain” that is at all useful.  Descartes’ sense, of absolute indubitability, is at best silly and pointless to worry about, possibly even meaningless.  (Descartes should have stayed in the army, or just stuck to practical pursuits like analytical geometry and physics and anatomy.)

    Peirce does not rule out all skeptical hypotheses ever; he only insists that they be motivated in the context.  For example, in Pollock’s story, the narrator Mike actually does have positive reason to think he might be a comprehensively deceived brain in a vat, because of the bizarre circumstances; his doubts count as “real” doubts and are not merely idle or fanciful.  It doesn’t follow that Descartes’ do.

    Objection 1:  Peirce presumes that there is a firm and fixed difference between “real” and made-up doubt.  But surely it’s a matter of degree.  I know for all practical purposes that Davis Library is currently open 24 hours.  But suppose you ask me how much I am willing to bet on that.  $50?  Sure.  $100?  Um, OK.  $5,000?  Well, not really….  How about if I’m holding a gun to your baby daughter’s head?   Reply:  Right, I don’t know for all possible practical purposes—only for all present actual purposes, such as going to Davis at night or betting small amounts.  If you up the ante, of course I’m not certain any more, but that’s just because the context and the practicalities themselves have changed, which is perfectly consistent with Peirce’s thesis.

    Objection 2:  As I write this at 9:50 a.m., I believe that George W. Bush is (physically located) in the White House.  And I have no positive reason to doubt that he is.  My purposes vis-á-vis W. are minimal; for me, nothing much hangs on whether he is in the White House or not.  So by Peirce’s standards, I am certain and I know that W. is in the White House.  But that’s silly; no one would say that I know any such thing.  He could be anywhere.   Reply:  This is like the wife example.  Even though you have no reason to think W. is any place else in particular, you know that U.S. Presidents are often not in the White House, but elsewhere in D.C. or traveling somewhere in or out of the USA.  That general knowledge creates real doubt.

    Objection 3:  Peirce says that circumstances determine what possibilities/doubts are “real” ones.  But that brings up the question of evidence for beliefs.  If I hold a belief ever so firmly and stably, without a smidgen of doubt and without any positive reason to doubt it, Peirce must say that I am certain and I know, even if I have no evidence for the belief.  And that’s crazy.