Chapter 6 Moore’s Paradox of Self-Knowledge
Do you know what it’s like to be wrong? I’m not asking whether you know what it’s like to have been wrong. If you find yourself recollecting something you were wrong about in the past, what you are effectively doing is thinking about how right you are now: At least you’re not making that mistake anymore! It would be a very different matter to recognize that you are, currently, thinking about something incorrectly.
Error does not tend to survive introspective awareness of itself: when you catch your mind in going astray, at that very moment your mind cleans up its act. As soon as you realize it’s a mistake to think something, you become unable to make the mistake of thinking that. You can still think about it, as, for example, something you used to think, or something someone else might think. You can look at the erroneous thought, but you can no longer look at the world through it. Our errors are blind spots, which is why real open-mindedness is so difficult.
We often speak as though a person could simply choose to be more open-minded than they are, as though each person had, somewhere deep within themselves, a “consider the possibility that you are wrong” lever. We criticize stubbornness and obstinacy in a person as if these conditions were born of unwillingness to simply push the lever.
And it is true that in many contexts I can, in fact, decide to be self-critical: I may believe that I have vetted this paragraph for typos, or that I have added up a series of numbers correctly, but I can always double-check. When I am about to propose a plan to a group of people I might try to anticipate their objections, so as to come prepared with responses. I can consider the possibility that I am wrong, and thus it appears that the lever exists. The problem is that the set of occasions when people most need to pull it—when they are wrong about something fundamentally important, something that approaches the heart of how they live their lives—are also the occasions when the lever seems stuck. And it is precisely on those occasions that we blame people for obstinacy.
When I construct objections to my own plan or check my arithmetic, I suspend my commitment to seeing my plan as the right one, or to arriving at a particular sum. In these cases, I’m able to step back from my conclusion, in order to examine it. But there are other conclusions where this method won’t work, because I can’t step back from them. Considering the possibility that I am wrong about those conclusions would seem to require me to call into question the very belief that I cannot let go of—which is to say, to open myself up to seeing that I’m wrong, while I remain wrong. But is that even possible?
Socrates showed that the answer was yes. Even when open-mindedness is most difficult, it is still possible. How? Not by simply trying harder. Socrates does not believe in such things as “force of will,” and for the same reason he does not think that what obstinate people are missing is any kind of willingness. Socrates grants that there are circumstances when the lever really will not move under your own efforts. What he discovered was how another person could help you push it.
I. Do You Want to Rule the World?
There was once a young man who wanted to rule the world, but never dared give voice to that wish. He couldn’t admit to the people around him that he hungered to rule over them; he couldn’t even admit it to himself. All of that changed on the day he met Socrates, and Socrates said to him:
Suppose one of the gods asked you, “Alcibiades, would you rather live with what you now have, or would you rather die on the spot if you weren’t permitted to acquire anything greater?” I think you’d choose to die. What then is your real ambition in life? I’ll tell you. You think that as soon as you present yourself before the Athenian people—as indeed you expect to in a very few days—by presenting yourself you’ll show them that you deserve to be honored more than Pericles or anyone else who ever was. Having shown that, you’ll be the most influential man in the city, and if you’re the greatest here, you’ll be the greatest in the rest of Greece, and not only in Greece, but also among the foreigners who live on the same continent as we do.
And if that same god were then to tell you that you should have absolute power in Europe, but that you weren’t permitted to cross over into Asia or get mixed up with affairs over there, I think you’d rather not live with only that to look forward to; you want your reputation and your influence to saturate all mankind, so to speak. I don’t think you regard anybody as ever having been much to speak of, except perhaps Cyrus and Xerxes. I’m not guessing that this is your ambition—I’m sure of it.1
Alcibiades wants to rule, and he wants to rule over everyone. If he were consigned to rule only over the Athenians, or only over the Greeks, that achievement would be so puny as to merit suicide. Alcibiades does not think life is worth living unless he is the greatest and most influential man in the entire world. Socrates expects to shock Alcibiades with this description of his own lust for power, and he does. A desire to rule the world was no more socially acceptable in Socrates’ time than it is in ours, but Socrates is not interested in criticizing Alcibiades. Rather, Socrates wants Alcibiades to confront who he really is and what he really wants, without a care for seeming sane and palatable and normal. Socrates wants Alcibiades to take himself seriously, so that he can examine himself with an open mind.
By the end of the dialogue, things have gotten serious indeed. Alcibiades has come to accept a new truth about himself—not that it is wrong to want to rule the world, but that he in particular is unfit to do so. And Socrates pushes him a step further, convincing Alcibiades that “before one acquires virtue it’s better to be ruled by somebody superior than to rule,” and that “it’s appropriate for a bad man to be a slave.”2 When these principles are taken together with what the two of them have learned, over the course of the dialogue, about Alcibiades—how ignorant he is when it comes to justice—the implication is so shocking that Socrates doesn’t quite want to spell it out for Alcibiades:
Socrates: Can you see what condition you’re now in? Is it appropriate for a free man or not?
Alcibiades: I think I see only too clearly.3
Without saying it directly, Socrates has landed on the most offensive and outrageous insult you could make to an aristocratic Greek: “You deserve to be a slave.” But Alcibiades, who wants to rule the world, accepts that this insult is true. There is a sharp divergence between the reality of Alcibiades’ condition and the self-image with which he identifies—and in the dialogue we see this divergence gradually become available to Alcibiades himself. The contrast is all the more salient if we consider it against the sweep of Alcibiades’ life.
Alcibiades’ ambitions were, if oversized, not entirely irrational: he was stunningly handsome, well educated, and descended on both sides from powerful noble families. He was a persuasive and charismatic speaker, and “an ancient equivalent of a record-breaking sports superstar” owing to his Olympic victories in chariot races.4 The encounter with Socrates that is related in the Alcibiades occurs on the eve of Alcibiades’ entry into public life, at the age of about eighteen. This is a moment when all believed him to be destined for political and military greatness—especially Alcibiades himself, who, in the historian Thucydides’ (460–400 BCE) account, seeks election as a general by announcing “Athenians, I more than others am entitled to command . . . and I consider myself deserving as well.”5
Once he has the ear of the Athenians, Alcibiades pushes them to embark on an ambitious military conquest of Sicily. It ends disastrously. During the expedition, Alcibiades was recalled by Athens to stand trial on charges of impiety—he was thought to be involved in a religious scandal—and he chose, instead, to flee, to defect to the side of the enemy (Sparta), and to advise them on how to defeat the Athenians. Later, Alcibiades would fall out of favor with the Spartans and shift his allegiance to the Persians, playing the Athenians and the Spartans against each other.
The Athenians’ love-hate affair with Alcibiades did not end with his initial act of betrayal: four years after the Sicilian Expedition they recalled him to command a naval fleet, the successes of which eventually led them to reverse his condemnation for impiety—only to go on to dismiss him again shortly thereafter, due to a military defeat. After this, Alcibiades retired, though there were subsequent calls for his return; his final attempt to offer advice to the Athenians met with rejection, and then the next year he was assassinated, probably by Spartans, at the age of forty-six. We have no contemporaneous accounts of Alcibiades’ death, and although the details in Plutarch’s account—written six hundred years later—are almost certainly inaccurate, they convey his iconic status:
The party sent to kill him did not dare to enter his house, but surrounded it and set it on fire. When Alcibiades was aware of this, he gathered together most of the garments and bedding in the house and cast them on the fire. Then, wrapping his cloak about his left arm, and drawing his sword with his right, he dashed out, unscathed by the fire, before the garments were in flames, and scattered the Barbarians, who ran at the mere sight of him. Not a man stood ground against him, or came to close quarters with him, but all held aloof and shot him with javelins and arrows. Thus he fell.6
The nineteenth-century historian and lexicographer William Smith summed up his death and life in these terms:
Thus perished miserably, in the vigour of his age, one of the most remarkable, but not one of the greatest, characters in Grecian history. With qualities which, properly applied, might have rendered him the greatest benefactor of Athens, he contrived to attain the inhumane distinction of being that citizen who had inflicted upon her the most signal amount of damage.7
Alcibiades was not, in fact, well suited to rule the Athenians, the Spartans, or the Persians, let alone all of them taken together. We can see this, and no doubt there were some people during Alcibiades’ lifetime who could see it—but how was Socrates able to get Alcibiades himself to see it? The dialogue presents a stunning reversal, in which a man who thinks he should rule the world, a man given to statements such as “I, more than others, am entitled to command,” instead erupts, repeatedly, in moments of pained self-awareness:
Well, Socrates, I swear by the gods that I don’t even know what I mean. I think I must have been in an appalling state for a long time, without being aware of it.8
Plato’s Alcibiades tells the story of Alcibiades becoming aware that he is making a mistake, even as he continues to make it—moreover, the reader knows he will continue to make it, all the way up to his death.
II. Moore’s Paradox
Honey never spoils. This is true, though many people are unaware of it. Consider the sentence, “Jones believes that honey spoils, though in fact he’s wrong about that: it doesn’t spoil.” This sentence might be true, and I might know that it’s true, and you might know that it’s true, and everyone in Jones’ family might know that it’s true, but there is one person who will struggle mightily with such a sentence, and that person is Jones. Imagine Jones trying to agree with us: “Yes, all of you people are right: I believe that honey spoils, though in fact it doesn’t spoil.” We would feel unsure how to interpret his words. We might doubt his sincerity. Sincere assertions license belief ascription: if Jones tells you that honey never spoils, and he is being sincere, then you can attribute to him the belief that honey never spoils. So, what can he mean by going on to express the belief that it does? And yet notice that the state of affairs he is struggling to tell you about—where the world is one way, but his mind represents it as being a different way—is one that can certainly occur. Moreover Jones, unless he has delusions of epistemic grandeur, knows that it can occur.
Sentences that fit the pattern, “p is the case, but I believe it isn’t”—or its subtly different variant “p is the case, but I don’t believe it is”—are sometimes called “Moore sentences” after the philosopher G. E. Moore (1873–1958), who first singled them out for philosophical attention. It is important to notice that such sentences are not logical contradictions, as is evident from the fact that they can be true. Moreover, the person herself can believe, assert, and know variants of these sentences, for instance ones expressed in the past or future tense. Jones can readily assert “I used to believe that honey spoils, though it doesn’t” and he can also, if he knows he is a forgetful person, tell us that “Honey never spoils, but I, Jones, won’t believe that in the future.” Jones can also be generically aware of his own fallibility: “I am sometimes wrong about food safety.” What he can’t do is believe that this is one of those times. The Moore sentence is not a contradiction; it is a blind spot. A person can be aware that others’ minds diverge from reality, and she can be aware that her own mind can, or did, or will diverge from reality; what she cannot be aware of is the specific way in which it is currently doing so. So goes the philosophical conventional wisdom since Moore.
That conventional wisdom says it’s just straightforwardly impossible to believe or sincerely assert a Moore sentence. No one can do it. It cannot be done. Jones can speak the sentence “I believe honey spoils, but it doesn’t spoil,” he can even shout it, but the one thing he can’t do is mean what he’s saying. Jones can “think” those words in the sense that he can, for example, imagine them scrolling on a screen before his mind’s eye, but he cannot believe the corresponding sentence. He can’t be aware of (exactly) what he’s wrong about.
This bit of conventional wisdom generates a paradox—“Moore’s paradox”—because it is puzzling that such truths would be inaccessible. Consider some other examples of inaccessible truths: many scientific truths are inaccessible to me because I haven’t paid the cost of entry—namely, years of study—whereas others would be inaccessible even if I had, because no one has discovered them yet. I can’t access truths about the past that went unrecorded, or truths about your emotional life that you refuse to share with me. Moore-paradoxical sentences are unlike any of these cases. The facts that correspond to those sentences, and make them true, are inaccessible to us not because they are too far away from us, but because they are, somehow, too close.
Philosophers find this deeply puzzling: How can proximity generate difficulties of access? Nonphilosophers might be just as puzzled that philosophers care about the paradox. Why does it matter whether one can say, “Honey never spoils but I don’t believe that”?
I am going to argue that it is, in fact, possible to sincerely assert a Moore sentence, and that it is important that this is possible: there exist Moore sentences whose inaccessibility would be a moral and intellectual disaster for us.
III. Aporia
If two philosophers meet, and they meet as philosophers, then it is likely that before long one of them will tell the other why they are wrong about something. Refutation is the fundamental form of philosophical interaction; even if most of us are not Socratics, we all share in the Socratic patrimony. Socrates understood refutation (in Greek, the word is elenchos) as the hallmark of his philosophical activity,9 and the Socratic dialogues even have a name for the experience in which refutation culminates: aporia. One of Socrates’ interlocutors describes aporia as a feeling of perplexity that is akin to being “stung” by a torpedo fish: “My mind and my tongue are numb, and I have no answer to give you.”10 Another describes it as watching his own sentences move before his very eyes: “I have no way of telling you what I have in mind, for whatever proposition we put forward goes around and refuses to stay put where we establish it.”11 What happens to you when you are refuted? What is it to be in a state of aporia?
Let’s start with the notion that aporia—being refuted—is something that is experienced and felt by the person who undergoes it. Suppose that person A shows person B, in the absence of person C, that C’s views are incorrect. Has A refuted C? Not as far as Socrates is concerned: so long as C is unaware of what has transpired, there is no state of aporia, and thus no refutation. The phrase “He did not realize that she had refuted him” is an acceptable use of the word “refute” in ordinary English, but Socratic refutation includes the proviso that no one is unwittingly refuted.12 What would Socrates say to those who insist that a view has been refuted if there’s an available proof as to its falsity, irrespective of whether anyone holds the view (or knows the proof)? He would say that one can use “refutation” to describe such a case if one likes, much in the way that one can, pointing to a painting of a dog, say “that’s a dog.” But of course it also makes perfectly good sense to say, of the dog in the painting, “That’s not a real dog.” We will soon see why, for Socrates, real refutation requires recognition by the person being refuted.
So, what is the experience of being (really) refuted? Is it the same as changing one’s mind, or suspending judgment on some question one used to have a firm belief on? No. First, there are non-refutational routes to changing one’s mind or suspending judgment: I might simply forget my old views, and come to adopt a new one or no view at all, without noticing what I have given up. But even in the case where I suspend judgment because I have been refuted, those two events are not identical. The same point holds for change of mind: it is not the same thing as refutation, but rather is an effect of refutation. Refutation may cause a change of mind or suspension of judgment, but something is not identical to what it causes. Moreover, the qualitative character of refutation is quite different from either change of mind or suspension of judgment. Being refuted feels like ignorance, confusion, perplexity, whereas once you have changed my mind the perplexity is over and I think I am now in the right. And if I have suspended judgment I at least know I am not wrong, so that is a kind of safety as well.
Refutation is a (possible) reason for the suspension of judgment, or change of mind; but refutation should not be equated with the effects of refutation. A change of mind or suspension of judgment is sometimes undertaken in response to a predicament that necessitates it, and it is that predicament that we want to describe. The predicament is an experience to the effect that, until I change my mind or suspend judgment, I am in the wrong, I am making a mistake. One needn’t voice this predicament out loud, but in a philosophical context, it is polite to do so. Instead of skipping directly from asserting p to denying it, or suspending judgment about it, we mark the transition by saying, “You got me” or “You’re right” or “I’m wrong” or “I see now” or “Okay”—or just by pausing for a moment. These phrases (or silences), taken together with their corresponding facial expressions, serve to distinguish cases where one changes one’s mind as a result of being in the predicament of being refuted from cases of simply changing one’s mind. But what is that predicament?
One possible way to describe what you are doing when you are refuting someone is that you are getting them to contradict themselves. Should we then say that being refuted amounts to asserting a contradiction? That cannot be right, either. Most philosophers, as well as most nonphilosophers, believe that no claim of the form “p and not p” could possibly be true, and that it is impossible to sincerely assert one—and yet these people can still be refuted!
There is a group of philosophers who allow for the possibility of true contradictions—they are called “dialetheists”—and it is helpful to distinguish their view from anything that happens in a refutation. A dialetheist takes the contradiction he asserts to be an accurate representation of the way the world is. The contradictoriness of his sentence simply mirrors the contradictoriness he thinks he sees in the things themselves. To judge that “p and not p,” is to judge that in some way or other the world is contradictory. But that is not the kind of thought we are trying to produce in refuting someone. We are trying to tell him: the world is not the problem, what is broken is your thinking about it. We insist that there is a way of thinking properly, and it is not his, that he is in the wrong, that his mind is—not was—in some kind of defective condition. We want him to judge that the world does not really work the way he thinks it does. This is just to say, we want to get him in a position to sincerely assert a Moore sentence. The sentence that expresses the content of a refutation is not “p and not p” but “p, but I don’t believe it” (or “not p, but I believe it”). Thus, the possibility of asserting Moore sentences and the possibility of refutation are one and the same. What is at stake in Moore’s paradox is nothing less than the practice of philosophy itself.
IV. A Moore Sentence, Asserted
Often, the best kind of proof that something is possible is one that shows that it is already actual. I thus propose to prove that Moore sentences can be sincerely asserted by pointing you to a place in the Socratic dialogues where Socrates gets Alcibiades to assert one.
Socrates: Alcibiades, the handsome son of Clinias, doesn’t understand justice and injustice—though he thinks he does.
Alcibiades: Apparently.13
(This is a slightly edited version of the exchange. I’ll quote the full version below.)
“P isn’t the case, though I believe it is” is a formula for a Moore sentence. Alcibiades’ “apparently” constitutes agreement to such a sentence, with “Alcibiades understands justice,” standing in for p. Notice that both of Socrates’ verbs are in the present tense. What Alcibiades agrees to—and thus asserts—is that now, at the moment when he says “apparently,” he continues to believe he understands justice and injustice, while at the same time continuing to assert “I do not.” Socrates has managed to orchestrate a situation in which Alcibiades sincerely asserts what he doesn’t believe: I think I understand justice, though I don’t in fact understand it. It is important to distinguish what Alcibiades agrees to from sentences such as “I think I understand justice, and I think I don’t” and “I understand justice, and I don’t.” Those sentences are contradictions, whereas the one Alcibiades effectively asserts is in fact a correct description of the way the world is.
“But Moore sentences are unassertable!” Given how entrenched this piece of conventional wisdom has become in the discipline of philosophy, I expect that some readers may be inclined to resist granting that this exchange qualifies as a full-fledged Moore sentence. First, how sincere is Alcibiades being? “Apparently” sounds grudging, and further examination of the context will support this impression. Moreover, the person who utters the words “Alcibiades doesn’t understand justice and injustice—though he thinks he does” is not Alcibiades, but Socrates. For both of these reasons, one might suspect that it is only Socrates who recognizes Alcibiades’ lack of understanding, not Alcibiades. And, of course, “Alcibiades doesn’t understand justice, though he thinks he does” fails to be Moore-paradoxical when said by Socrates. If Socrates is the one uttering the relevant words, doesn’t it follow that Socrates is the one talking, and that Socrates is expressing the views of Socrates? No.
All of these worries about whether what I have claimed to be a Moore sentence really counts as a Moore sentence are, in fact, addressed in the conversation that leads up to it. Let me now quote the context at some length. (I’ve italicized where the Moore sentence appears.)
Socrates: Well then, given that your opinion wavers so much, and given that you obviously neither found it out yourself nor learned it from anyone else, how likely is it that you know about justice and injustice?
Alcibiades: From what you say anyway, it’s not very likely.
Socrates: See, there you go again, Alcibiades, that’s not well said!
Alcibiades: What do you mean?
Socrates: You say that I say these things.
Alcibiades: What? Aren’t you saying that I don’t understand justice and injustice?
Socrates: No, not at all.
Alcibiades: Well, am I?
Socrates: Yes.
Alcibiades: How?
Socrates: Here’s how. If I asked you which is more, one or two, would you say two?
Alcibiades: I would.
Socrates: By how much?
Alcibiades: By one.
Socrates: Then which of us is saying that two is one more than one?
Alcibiades: I am.
Socrates: Wasn’t I asking and weren’t you answering?
Alcibiades: Yes.
Socrates: Who do you think is saying these things—me, the questioner, or you, the answerer?
Alcibiades: I am.
Socrates: And what if I asked you how to spell “Socrates,” and you told me? Which of us would be saying it?
Alcibiades: I would.
Socrates: Come then, give me the general principle. When there’s a question and an answer, who is the one saying things—the questioner or the answerer?
Alcibiades: The answerer, I think, Socrates.
Socrates: Wasn’t I the questioner in everything just now?
Alcibiades: Yes.
Socrates: And weren’t you the answerer?
Alcibiades: I certainly was.
Socrates: Well then, which of us said what was said?
Alcibiades: From what we’ve agreed, Socrates, it seems that I did.
Socrates: And what was said was that Alcibiades, the handsome son of Clinias, doesn’t understand justice and injustice—though he thinks he does—and that he is about to go to the Assembly to advise the Athenians on what he doesn’t know anything about. Wasn’t that it?
Alcibiades: Apparently.
Socrates: Then it’s just like in Euripides, Alcibiades; “you heard it from yourself, not from me.” I’m not the one who says these things—you are—don’t try to blame me. And furthermore, you’re quite right to say so. This scheme you have in mind—teaching what you don’t know and haven’t bothered to learn—your scheme, my good fellow, is crazy.14
Plato died thousands of years before G. E. Moore was born, so he cannot have been worried about whether readers would take the italicized language above as a “real” Moore sentence. And yet it is remarkable how careful Socrates is to distinguish the situation they are in from one in which it is Socrates who is asserting the relevant sentence. Plato has Alcibiades initially interpret Socrates like this: “Aren’t you saying that I don’t understand justice and injustice?” And then Plato has Socrates correct him and point out that Alcibiades is the one making the relevant assertion: “You heard it from yourself, not from me.” Socrates is not telling Alcibiades that he is wrong. Socrates is getting Alcibiades to say, “I’m wrong” and then he’s making sure that Alcibiades sees that those were his own words.
The Moorean drama does not end here. The argument that follows the passage quoted above also arrives at a Moore sentence—and this time Socrates announces where he is headed in advance. After Alcibiades insists that just things needn’t be advantageous, Socrates tells him that soon the opposite words will be coming out of his mouth:
Socrates: Try to prove that what is just is sometimes not advantageous.
Alcibiades: Stop pushing me around, Socrates!
Socrates: No, in fact I’m going to push you around and persuade you of the opposite of what you’re not willing to show me.
Alcibiades: Just try it!
Socrates: Just answer my questions.
Alcibiades: No, you do the talking yourself.
Socrates: What?! Don’t you want to be completely convinced?
Alcibiades: Absolutely, I’m sure.
Socrates: Wouldn’t you be completely convinced if you yourself said, “Yes, that’s how it is”?
Alcibiades: Yes, I think so.
Socrates: Then answer my questions. And if you don’t hear yourself say that just things are also advantageous, then don’t believe anything else I say.15
Even though Socrates has announced where he is headed in advance, Alcibiades is unable to divert the argumentative train to a different conclusion. Socrates’ prediction comes true—see chapter 4, here, for a discussion of the argument Socrates uses to make the prediction come true—and Alcibiades ends up assenting to another Moore sentence:
Socrates: It’s obvious from what we’ve said that not only are you ignorant about the most important things, but you also think you know what you don’t know.
Alcibiades: I guess that’s right.16
The fact that Alcibiades’ assent is grudging, that he is driven to make admissions he was not initially inclined to make, is no reason to doubt the sincerity of his responses. When a witness is brought, by the pressure of precise cross-examination, to grudgingly admit corrections to an earlier version of his testimony, we don’t doubt the sincerity of those admissions. Instead, we think, “It is a good thing that that lawyer was there to question him.” Indeed, grudging admissions are generally more sincere than non-grudging ones, not less; if someone is happy to admit something, his motives are potentially a mix of truth and pleasure. In grudging admissions, truth pulls against pleasure. Socrates’ questioning makes it possible for Alcibiades to say, and therefore to think, a thought that would otherwise have lain in his blind spot. To understand how this is possible, we need to again consider the problem in the abstract.
V. Normative Self-Blindness
Consider a third personal, and therefore non-paradoxical, variant of a Moore sentence. If I say “p, but Jones doesn’t believe it,” I am expressing a critical evaluation of Jones’ thought. I am holding Jones’ thinking up to a normative standard—that of the truth—and saying that he is missing something. The thought “p but Jones doesn’t believe it” claims that there is a truth Jones fails to have, that his mind is missing something. Likewise, if I say, “Jones believes p, but it’s not the case that p,” I am saying that Jones’ thinking fails: he is wrong. Moore sentences are, first and foremost, evaluative judgments about some belief. The problem with first personally asserting a Moore sentence is that it would require me to take a critical attitude toward my own thoughts, and I do not seem able to do this. When I assess what you think, I compare it to the way the world really is. But looking at how the world really is is how I figure out what I think in the first place. When I ask myself what I think about something, I am already asking what is true about that thing. There is no room for a separate assessment step, which means that my thoughts are evaluatively inaccessible to me.
Normative self-blindness is the phenomenon that underlies Moore’s paradox. When I make an assertion about the way the world is, you can also ascribe to me the corresponding belief, as though, when I said, “Honey never spoils,” I had said “I believe that honey never spoils.” The reason you can take this liberty is that these two things—p’s being the case, and my believing that p—are, though distinct from your point of view, identical from my own. If I believe that p, I think I am correct in holding that belief, and that p is the case. When I have a belief, I don’t have the ability to separately evaluate the truth of that belief. But that is not the case when I look at your beliefs: the question of what you believe and whether it is true are not the same question.
Moore paradoxical sentences are the clearest, but not the only, examples of normative self-blindness. Let’s consider a few other examples. These examples are imperfect: in each case, questions arise as to whether we are necessarily or fully self-blind about the relevant phenomenon. I won’t extensively engage with those questions; my goal is to highlight the existence of, rather than drawing precise lines around, a certain territory peripheral to our Moore sentences. By considering these related phenomena, we’ll get a better grip on the problem to which Socratic refutation is the solution.
Example 1: Parental Praise
When parents praise a child’s artwork, appearance, or intelligence, the child often doesn’t take that praise seriously. Sometimes, the child may suspect his parents of insincerity, but this needn’t be the case. Even if he believes his parents are saying exactly what they think, he may think that they are incapable of assessing him accurately, because he is too close to them. Because he is their own, their flesh and blood, his parents are inclined to see what he does (or how he looks or thinks), and see it as good, in the same glance. They cannot take a separate step of applying standards to him. This is much the way we are with our beliefs: if I were to start to enumerate some things I think, and pause after each one to decide whether it is true, I would miraculously end up placing a check mark next to each and every one. Belief #1, correct. Belief #2, correct again. Belief #3, also true. If I were enumerating your beliefs, things would be very different. To say I am biased toward myself is a wild understatement: I am not more likely to judge my beliefs true; I am utterly incapable of judging them to be false. I am not evaluating them at all, because they are evaluatively inaccessible to me.
Example 2: The Paradox of Modesty
Consider a puzzle about modesty: if being an unassuming, reserved, humble person—the sort who is averse to self-praise—is a virtue, then it deserves praise. It would follow that the modest person cannot evaluate herself as possessing the virtue of modesty: were she to do so, she would pride herself on her modesty, and thereby lose it.
The philosopher Richard Moran, developing Bernard Williams’ thought that modesty is “self-effacing,” in that “a modest person does not act under the title of modesty,” observes: “It is part of modesty, insofar as it is a virtue, and hence praiseworthy, not to reflect on itself, in particular not to insist on taking credit for its praiseworthy character . . . if the quality of modesty is taken to be incompatible with praise of oneself, this would appear to be a quality that cannot survive reflection on itself.”17
Notice that while Moran’s and Williams’ attention is on whether one can be aware of oneself as being modest, the real difficulty here is not awareness but evaluation. The modest person can notice that she doesn’t speak up as often as others, or that she is not as inclined to praise herself, or that she tends to adopt an unassuming pose, and so on. What she cannot do is evaluate these features of herself positively. The modest person can be perfectly well aware of her modesty so long as she doesn’t see modesty as a virtue, or if she has some special reason for thinking that her own modesty is somehow non-virtuous. When it comes to her own modesty, the modest person must be normatively self-blind.
Example 3: Self-Promising
Try this experiment. Pick something that you usually struggle with—answering emails promptly, staying off your phone, going to bed on time—and promise yourself that you will do better on this front for the next twenty-four hours. Done? Okay, I predict that twenty-four hours from now, you will have trouble answering the question, “Did you keep your promise?” Suppose you picked the bedtime promise, and that you do in fact go to bed on time. Couldn’t that be because I drew your attention to the issue, rather than because you felt bound by the force of the promise? Suppose you stay up late. Why not think that means that you released yourself from a silly promise that was, in the first place, only an example in a philosophy book?
What you will surely know, twenty-four hours from now, is the answer to the factual question as to when you went to bed. It is less clear that you will have an answer to the normative question about whether your bedtime met the standard set by a promise. Can you make promises to yourself? We sometimes say “I promised myself that I would . . . ,” but it is not clear that there is any difference between breaking such a promise and releasing ourselves from it.
Imagine how much easier it would have been for you to separate my two questions—Did you go to bed on time? and Did you keep your promise?—if you hadn’t had to rely only on your own devices. Imagine that you’d promised your spouse you would go to bed on time tonight. It wouldn’t matter whether you did so at my prompting, whether you subsequently changed your mind about the wisdom of doing so, and so on. Like it or not, you’ve promised, and that means there is a standard governing your action: if you don’t go to bed on time, you’ve failed to meet it. It is now up to them to decide whether they want to hold you to your promise; the distinction between breaking and being released from the promise becomes easy to draw.
People who cooperate with us in the right way make possible a form of normative assessment that has no solipsistic counterpart. This happens when we make a promise to someone, and it happens when someone refutes us. I cannot distinguish between “something I believe” and “a truth about the world” introspectively, because I cannot assess my own beliefs for truth while continuing to believe them; but of course my beliefs are assessable for truth, and the practice of assessing them is one I can participate in—with the help of another. Socrates articulates this point by saying that another person can serve as a mirror to me.
VI. Socratic Self-Knowledge
After having been led to assert a Moore sentence two times, Alcibiades is eager for self-improvement:
Well, Socrates, what kind of self-cultivation do I need to practice? Can you show me the way?18
Socrates points out a problem:
I’m afraid we often think we’re cultivating ourselves when we’re not.19
What is it to really cultivate oneself? This question leads Socrates to raise a puzzle about the famous Delphic injunction to “Know thyself.”
Socrates: Is it actually such an easy thing to know oneself? Was it some simpleton who inscribed those words on the temple wall at Delphi? Or is it difficult, and not for everybody?
Alcibiades: Sometimes I think, Socrates, that anyone can do it, but then sometimes I think it’s extremely difficult.20
Is assessing one’s own beliefs for truth easy or difficult? In one way, it is very easy—each one immediately receives a check mark. When I assess the beliefs of others, by contrast, I have to do actual work to check whether each is true. But of course the “ease” of assessing our own beliefs is a sign we are not really assessing them. Socrates returns to this point about the Delphic injunction later in the discussion:
Socrates: I’ll tell you what I suspect that inscription means, and what advice it’s giving us. There may not be many examples of it, except the case of sight.
Alcibiades: What do you mean by that?
Socrates: You think about it, too. If the inscription took our eyes to be people and advised them, “See thyself,” how would we understand such advice? Shouldn’t the eye be looking at something in which it could see itself?
Alcibiades: Obviously.
Socrates: Then let’s think of something that allows us to see both it and ourselves when we look at it.
Alcibiades: Obviously, Socrates, you mean mirrors and that sort of thing.
Socrates: Quite right. And isn’t there something like that in the eye, which we see with?
Alcibiades: Certainly.
Socrates: I’m sure you’ve noticed that when a person looks into an eye his face appears in it, like in a mirror. We call this the “pupil,” for it’s a sort of miniature of the person who’s looking.*
Alcibiades: You’re right.
Socrates: Then an eye will see itself if it observes an eye and looks at the best part of it, the part with which it can see.
Alcibiades: So it seems.
Socrates: But it won’t see itself if it looks at anything else in a person, or anything else at all, unless it’s similar to the eye.
Alcibiades: You’re right.
Socrates: So if an eye is to see itself, it must look at an eye, and at that region of it in which the good activity of an eye actually occurs, and this, I presume, is seeing.
Alcibiades: That’s right.
Socrates: Then if the soul, Alcibiades, is to know itself, it must look at a soul, and especially at that region in which what makes a soul good, wisdom, occurs, and at anything else which is similar to it.21
The question of how Alcibiades is to cultivate himself—which is to say, acquire self-knowledge, as per the Delphic injunction—is here compared to a visual problem: How can someone see herself? When I look out into the world, the one thing I don’t see is me. I’m not in my own field of view. But wait, what about the fact that I can see my arm or leg by turning my eyes downward? Socrates anticipates this objection, and clarifies that he is not referring to that kind of self-seeing: we should imagine that the oracle “took our eyes to be people” and advised these eye-people to try to see themselves.
Indeed, Socrates goes one step further to specify that we should imagine the order “See thyself” being given to a part of the eye. Socrates is taking great care to articulate the problem as one of self-seeing: seeing the part of the eye that is doing the seeing. In principle, if the part of my eye that sees could project itself from the rest of my eye—imagine it protruding on an antenna—my eye could “see itself,” which is to say, see other parts of the eye. But what it could not do is see the very part that is seeing. The activity of seeing is, in some way, self-blind. What does the seeing cannot see itself, just as what does the knowing cannot know itself—at least, not by itself.
The eye can see itself by means of a special sort of intermediary, one that “allows us to see both it and ourselves when we look at it.” Still water, or polished bronze, or any other highly reflective surface, could, in principle, play the role of this intermediary, but Socrates fastens onto one very specific kind of reflective surface—the shiny blackness of another person’s pupil—in order to bring the visual case into line with the intellectual one. If you want a normative grip on yourself, you are going to need the help of another person.
Thinking about one’s own thoughts is a two-person job. That is the immediate point of the analogy. But why is Socrates so intent on making this point with reference to a subset of thoughts, namely the ones that lie in the circumscribed region of the soul in which wisdom occurs? Recall that this whole discussion is meant to elaborate the project of self-cultivation. Socrates thinks knowing yourself is how you cultivate yourself—but only if you focus on the right part of yourself. Not all of your thoughts are equally significant to who you are. Socrates is drawing attention to the importance of evaluating the part of you that is doing the thinking. Those thoughts that are undetachable from the project of thinking—thoughts you are always “using” in order to think anything at all—are what you really need to know if you want to cultivate yourself.
G. E. Moore and the philosophers who inherited his paradox focused on relatively trivial sentences, such as about Jones and honey, and those sentences stand in contrast with the more momentous, life-changing Moore sentences we find in Plato. We can now explain why this difference matters. On most topics, I can alienate myself from what I happen to believe by deciding to suspend judgment. Even though I have a belief, I can step back and reflect on it. I can express such a decision by saying, for example, “I believe honey spoils, but is that right?” This question resembles a Moore sentence, but without the accompanying verbal infelicity and aura of paradox. It conveys my intention to set aside the fact that I already have a belief, one way or the other, about honey, so as to investigate the matter as though I didn’t. My suspension of judgment creates enough distance between me and my belief that I can evaluate it; no paradox arises.
Suspending judgment is the conceptual analog to twisting or turning my body to bring parts of myself I can’t usually see into view. There are parts of my body whose invisibility follows from how I usually position myself in order to look out at the world: my shoulder, the back of my knee, the soles of my feet. But if I set aside the goal of looking out onto the horizon for new information, and the upright posture that optimizes for that aim, I can contort myself to bring these parts into view. If I had the antenna, I could even see the other parts of my eye. Likewise, suspending judgment allows me to investigate claims I already believe. I usually treat such claims as settled background against which I acquire new information, but I can decide to treat them differently. I turn my belief into a nonbelief by detaching myself from it, and thereby put myself into a position to apply evaluative standards to it. Suspension of judgment is a measure we can take to eliminate such blind spots—though not, of course, to eliminate all of them simultaneously. If I am contorting myself to bring the back of my left knee into view, I can’t see my right shoulder.
Even if it is true that I cannot get all of the parts of my body into view at once, the difficulty of getting my knee or shoulder into view is of an entirely different kind from the difficulty of getting the seeing part itself into view. That is a persistent kind of blind spot—it moves with me, no matter where I focus my attention. Socrates did not speak of “untimely questions”—that phrase is my own invention—but when he speaks of “that region in which what makes a soul good, wisdom, occurs,” I take him to be referring to the thoughts that, at a fundamental level, guide our lives, and thus determine whether we count as “wise” or “foolish.” These thoughts are the answers to questions on which we cannot suspend judgment, because the answers drive everything we do, including any inquiry we might make into the questions themselves. Untimely questions help us reframe the stakes of Moore’s paradox.
Consider the claim, “No one can ever see an unseen thing.” A book that has just been printed and never opened is made up of “unseen” pages, which become seen as soon as someone opens it. In this formal sense, you can never see the unseen pages, but of course in a substantive sense you can see those pages, but we are going to call them “seen pages” now. The problem posed by trivial Moore sentences, such as the one about honey, is akin to the formal problem of seeing the unseen pages. Socrates’ problem—that of getting the seeing part itself into view—is analogous to the substantive difficulty we might have if some of the pages were glued shut or written in invisible ink. It is Socrates who identifies the philosophically troubling form of self-blindness, namely our stubborn inability to subject certain of our beliefs—the ones whose guidance matters most to us—to evaluative standards. It is our load-bearing beliefs—the ones that answer untimely questions—that we struggle to evaluate.
What might initially have appeared to be a narrow, technical problem of interest only to the subset of epistemologists and philosophers of language who study it turns out to be the Tolstoy problem in disguise. Once we view Moore’s paradox through a Socratic lens, we gain sympathy for Tolstoy’s struggles to confront questions such as “Why should I care about educating my children?” or “Why should I write novels?” We can now see that his problem is a problem of self-knowledge: he cannot evaluate his answers to these questions, because he is using them. That problem is not insoluble, but he cannot solve it on his own. He needs the help of others. Refutation is how they provide it.
VII. How Refutation Works
Socrates noticed a simple difference between intrapersonal wavering and interpersonal disagreement. When you say or do one thing and then, later, say or do something that conflicts with it—recall our various examples of wavering from chapter 1, from weakness of will to Russell’s emotive conjugation to young Hippocrates’ discombobulated scuttling—there is no through-line connecting your earlier thoughts and actions to your later ones, no way for you to hold one of those sets of thoughts to the standard set by the other. When you disagree with yourself, you are simply disjointed. But when you say one thing and I disagree with you, and we conduct that disagreement together, then there can be a coherence to our activity of arguing. When, for example, you seek the truth and I avoid error, we are doing one thing, together—disagreeing—in a way that the various time-slices of you are not doing one thing, together, when you disagree with yourself by wavering.
Return to the case of making a promise to someone. Because there is a promisee, there is space for normative assessment: you can see your own action as a failure insofar as it doesn’t meet someone else’s promise-grounded expectation. The refuter affords a similar experience. Here is how it works. If you are the refuter, first you ask someone a question, then they answer, and then, by way of further interrogation, you show them that you can’t accept their answer. You do this by showing them that it contradicts something else that both of you accept, or that it is internally incoherent, or that it simply doesn’t count as an answer to the question once the question has been clarified. Because you are holding them accountable—reminding them of what they said earlier in the conversation, or of what follows from what they said earlier, or of common sense, or of what they’ve agreed to on other occasions—they can come to see their answer as bad. They see that it would rightly be judged unacceptable by anyone who wasn’t caught up in already thinking it.
But that doesn’t mean that they instantly drop it, either. If the question was untimely, they can’t suspend judgment on it, so they can’t simply “give up” their only answer as soon as they see problems with it. Until they come up with a replacement, they continue to accept it, yet at the same time understand why you don’t. They acknowledge that you are right not to buy what they are selling; because of you, they can see a defect in their answer; you are a normative mirror for their thought. In the Platonic dialogues it doesn’t tend to take very long for someone like Alcibiades to regroup; nonetheless, even if only for the duration of one dialectical exchange, he is afforded the experience of watching his own thinking fall short. He sees the truth about himself reflected in the soul of his interlocutor.
One might ask: If the question is an untimely one, how is even the refuter capable of posing it? One possibility is that the question only counts as untimely for the refutee. Consider a question such as “Do you think that you and your spouse will stay married?” The person who poses this question can approach it with the detached curiosity unavailable to the person to whom it is being posed. We have seen that Socrates often gets his foot in the door by way of such questions—for instance, when he asks Lysis whether his parents love him, Euthyphro whether his prosecution of his father is pious, and Alcibiades whether he has what it takes to rule the world.
Yet Socrates systematically translates such one-way untimely questions into questions that are untimely for both parties: his conversation with Lysis quickly becomes “What is a true friend?” with Alcibiades it turns into “What is justice?” with Laches and Nicias he explores “What is courage?” with Euthyphro “What is piety?” In other words, Socrates guides each conversation toward the special subset of questions that are untimely for everyone, including himself. Just like the rest of us, Socrates needs to believe that he is a good person, which means that he needs to believe he is conducting himself in the manner of a true friend, with justice and courage and piety. Once again we see that Socrates is targeting the most stubborn blind spots: the ones we all share. Which returns us to the question: How are such questions even askable? How can Socrates pose questions on which he himself cannot suspend judgment?
The solution is to recognize that it is possible to direct a question such as “What is X?” at someone else, even if you have your own answer to that question, and even if you do not suspend judgment with reference to that answer. When Socrates asks “What is X?” and his interlocutor offers an answer, Socrates doesn’t reply with a simple “Great, thanks so much.” Instead, he asks a series of pointed questions, each of which leads to an explanation of why he cannot accept the answer. What is the soil from which his probing questions spring? Socrates’ own conception of X, refined over the course of many such conversations.
As we saw in the last chapter, Socrates’ conversations are not merely destructive but make positive progress toward knowledge of the subject matter. That progress is available to be deployed in subsequent conversations—or even later in the same conversation—as a basis for resisting, or demanding elaborations of, his interlocutors’ contrary claims. For example, in the Gorgias, two of Socrates’ interlocutors have claimed that doing injustice is better, for the one doing it, than undergoing injustice, and Socrates has refuted this claim so many times, in so many ways, that he describes the claim that doing injustice is worse than undergoing it as “held down and bound by arguments of iron and adamant.”22 Socrates uses what he has learned about X over past encounters as a standard against which he can compare his new interlocutor’s ideas, and will take what he learns over the course of that conversation into the next. Socrates is, like his interlocutors, both using his answers to an untimely question, and inquiring into that question, at the same time.
Socrates is testing what he has learned from others against what he stands to learn from each new person. He introduces an enlightening image for this process in the Gorgias when he describes one of his most oppositional interlocutors, Callicles, as “one of those stones on which they test gold.”23 The way for two people to inquire into a question each takes to be already answered is by pitting those answers against one another. Each serves, for the other, as a kind of testing stone, or, along the lines of the previous analogy, a mirror in which their own view becomes evaluable. It allows the refutee to ask: Is what I think the sort of thought that someone who doesn’t already think it would accept? How does it look from the outside?
But even if both participants are inquiring by using their own answers, it is worth observing two asymmetries between what Socrates is doing and what his interlocutors are doing. First, there is the fact that puts Socrates at such an advantage: his answers are the products of past investigations, whereas his interlocutors’ are usually not. Second, Socrates is playing Clifford, whereas his interlocutor is playing James. The interlocutor uses his answers to furnish the thesis to be examined, whereas Socrates uses his answers to examine that thesis. It is because the Clifford role is entirely responsive, skeptical, critical, and negative that Socrates denies having said anything—“Well then, which of us said what was said?”24—or rather, Socrates brings Alcibiades to deny that Socrates has said anything!
When Alcibiades speaks, he is saying what Alcibiades thinks. When Socrates questions, he is also—provided Alcibiades assents—saying what Alcibiades thinks. Once the project of saying what Alcibiades thinks is distributed over two people, “p, but I don’t believe it” becomes both sayable and thinkable. Socrates discovered that the space of speech is more capacious than the space of thought: it allows a person to see their own mistakes. Socrates drives Alcibiades to sincerely speak against himself, and then to witness the surprising mismatch between the words coming out of his own mouth and what he sees when he introspects. The blind spots associated with untimely questions cannot be inwardly “seen” by means of introspection, but they can be spoken—outwardly “seen”—by being mirrored in another. We can indeed assert Moore sentences, but that project is a collaborative one. With Socrates’ help, Alcibiades manages to explore the thoroughly familiar territory of his self as though he were in an uncharted land. And Socrates can, in turn, learn from what he has helped Alcibiades to say, and make use of it to sharpen his future refutations.
VIII. The Fate of Alcibiades
But if Socrates is so successful at exposing Alcibiades’ otherwise hidden ignorance, why did Alcibiades’ life turn out so badly? Why did he persist in his self-destructive quest for world domination? I believe that Plato asked himself this question, and that his answer can be found in the speech he puts into Alcibiades’ mouth in the Symposium. Whereas the conversation that we have been examining, in the Alcibiades, marks the beginning of Socrates’ association with Alcibiades, the Symposium marks its end: the Symposium is set seventeen years after the Alcibiades, which is about a year before the disastrous Sicilian expedition. In the Symposium, Alcibiades describes the extraordinary effect of Socrates’ words:
The moment he [Socrates] starts to speak, I am beside myself: my heart starts leaping in my chest, the tears come streaming down my face, even the frenzied Corybantes seem sane compared to me—and, let me tell you, I am not alone. I have heard Pericles and many other great orators, and I have admired their speeches. But nothing like this ever happened to me.
But then Alcibiades confesses that, despite Socrates’ refutations, Alcibiades keeps returning to his usual, politically ambitious ways. Every encounter with Socrates becomes a reminder that this choice is a mistake:
He always traps me, you see, and he makes me admit that my political career is a waste of time, while all that matters is just what I most neglect: my personal shortcomings, which cry out for the closest attention.
Alcibiades reports that over the course of such conversations he would find that “my very own soul started protesting that my life—my life!—was no better than the most miserable slave’s” and that Socrates “makes it seem that my life isn’t worth living!” He runs away from Socrates:
So I refuse to listen to him; I stop my ears and tear myself away from him, for, like the Sirens, he could make me stay by his side till I die. Socrates is the only person in the world who has made me feel shame—ah, you didn’t think I had it in me, did you? Yes, he makes me feel ashamed: I know perfectly well that I can’t prove he’s wrong when he tells me what I should do; yet, the moment I leave his side, I go back to my old ways: I cave in to my desire to please the crowd. My whole life has become one constant effort to escape from him and keep away, but when I see him, I feel deeply ashamed, because I’m doing nothing about my way of life, though I have already agreed with him that I should.25
Plato seems to be saying: Socrates can be a mirror showing you what’s wrong with you, but he cannot prevent you from looking away. Yet why does Alcibiades want to look away? He describes his encounters with Socrates as thrilling, and he clearly finds the mirror to be truthful. Alcibiades’ difficulty seems to be that everything he sees in it is negative: “my shortcomings,” “my life isn’t worth living.” Recall Clitophon’s inability to see the positive aspects of the Socratic project. Alcibiades cannot perceive how the Socratic method will ever afford him the possibility of moving forward in life, of making progress. The life of military conquest and political power promises concrete gains in wealth and status. When Alcibiades compares this life to the philosophical one, he can’t help but finding the latter wanting. Alcibiades grasps only the “refutation” side of the Socratic method, not the “inquiry” side.26
To be fair to Alcibiades, the positive side of the Socratic method is the much more difficult side to bring into view, to the point that an entire philosophical tradition—that of the ancient Skeptics—ended up framing the Socratic project in purely negative terms. Nonetheless, in the next chapter, we will try.