Let me begin with a friend of mine — I'll call her Naomi. About a year ago she came out of a long marriage, and came out of it badly. A few months later she met someone, quickly. I could see what it was. He was charming and unreliable and quite wrong for a person who was still recovering, and she had thrown herself into it because it was new and it let her not think about the other thing. Anyone who cared about her could see it. And one evening, over coffee, she put her cup down and asked me directly: what do you think of him, honestly?
I said he seemed very good. I said it was good to see her happy again, and that she deserved something easy after the year she'd had. She was pleased, and we moved on, and I drove home.
Here is the first thing I noticed, because by then I had learned to watch for it. What I felt was relief — not gladness on her behalf, but relief, the feeling you have when a difficult thing is over and you have got through it. And I knew what that feeling indicated, because we had discussed it on this programme: you do not feel relief after an act of kindness. A real kindness usually costs you something. Relief is what you feel after an escape.
There was a second thing, and it was worse, and it took me longer to see. The sentence I had used to excuse myself — the one I would have given if asked why I had lied to her — was: she couldn't handle it right now; she's been through enough. Consider what is in that sentence. To say that she could not handle the truth is to make a judgement about Naomi. It is to look at a grown woman who has just asked me directly for the truth and to decide, quietly, that she is too fragile to be given it — that she is a person to be managed rather than addressed. I had presented a verdict about her weakness as tenderness toward it. And the convenient part is this: judging her too fragile to hear the truth was also what spared me the difficult conversation I did not want to have.
I told Naomi a kind thing. Underneath it I told myself two things — that I was sparing her, and that she could not have taken it — and I am no longer confident that either was true, or was really about her at all.
Last week we examined the first of two flattering accounts we give of our own motives — the brave one, I'm just being honest. We found the real courage it imitates; we saw how it can turn into a way of wounding people; and we saw, with Sartre, that this kind of self-justification is not an occasional slip but is built into the sort of thing a self is. We left one question open, the one Sartre raised and then set aside: all right, then — so check. Examine the motive underneath. Can you?
This is Philosophy for Us — philosophy for everyone, no degree required. This is the seventh of eight episodes on the most ordinary difficult thing there is: telling the truth, or not, to the people we love.
Tonight, two things. First, the other flattering account — the kind one, the one I gave myself with Naomi: I didn't want to hurt them. And then that open question, taken up in full. A number of serious philosophers have spent their careers on this — on how it is even possible to lie to oneself — and they disagree with one another about almost everything. Tonight we will find the one thing they do not disagree about. It is not a comfortable conclusion.
The same rule applies as last week. If I set up a weak version of this account of our motives, no one will ever recognise it in themselves; we will all nod along about other people and leave with our consciences intact. So let me first give the kind lie its due, because it has a much better case than the brave one, and it is that better case which makes it so hard to detect.
Here it is. Most of the time, the soft answer is the right answer — not the cowardly answer, the correct one. The world is full of true things that no one needs to hear and that would only wound: a friend's haircut, two days before it grows out; a eulogy that ran a little long; the fact that you never much liked the husband who has mercifully become the ex-husband. There is no virtue in saying any of it. We looked at the philosophy of this at the start of the series — the brutal truth told to a grieving sister, which changes nothing and costs an hour of pain. Sometimes a kind silence is not a failure of honesty. It is a small competence at being a person. Knowing which truths to withhold is most of what we mean by tact; tact is a real virtue, and the person who has none of it is not a hero, merely exhausting.
So when I sat across from Naomi and decided not to tell her what I saw, I was doing something that is usually right. That is the difficulty. This kind of self-justification does not hide inside some obviously bad thing. It hides inside a genuinely good one — just as the brave version last week hid inside courage. Last week the counterfeit hid inside strength, in the pleasure of being the perceptive one. This week it hides inside gentleness, in the pleasure — or really the relief — of being the one who spares people. And gentleness is even harder to see through than nerve, because no one suspects the soft option of being self-serving. It looks like the costly choice. It looks like the thing you are giving something up to do.
So how would we ever tell the two apart? When is the soft answer a real kindness, and when is it self-protection? We cannot read it off the answer itself, because the kind truth-telling and the cowardly silence come out of our mouths as the same words: he seems great, I'm happy for you. We have to examine, as we have been learning to examine, what lies underneath. And this kind of self-justification leaves two signs. I felt both of them driving home from Naomi, and I want to name them carefully, because they are where the episode begins and they are what the rest of it will take away.
The first sign is the one we met some weeks ago, with my father: relief. When you have performed a real kindness, what you feel afterward is usually some form of the cost. You feel a little worn out, a little sad for the person, a little tender about the difficult thing you have just carried for them. What you do not feel, after a real sacrifice, is that light, clean sense of a thing being over — that's done, I got through it. That is not the feeling of having given something. It is the feeling of having escaped something. And if, after a kind silence, what arrives is relief, then somewhere in the exchange the thing you were protecting may not have been the other person. It may have been the difficult ten minutes you did not have to spend.
The second sign is the new one, and it is worse, and it is specific to the kind lie in a way the relief is not. Consider again the sentence I used to clear myself: she couldn't handle it; she's been through enough. Hear the verdict inside it. To say that someone could not handle the truth is to have quietly decided something about them — that they are fragile, that they are a person for whom reality must be curated, that they are not, at present, a full adult one can be straight with. It is a judgement of their weakness wearing the appearance of concern for it. And it is remarkably convenient, because the very same judgement that flatters me as the strong, protective one — the adult managing the situation — is also the one that gets me out of the conversation I did not want. Contempt and self-interest, folded together so closely that from the inside they feel exactly like love. I was protecting her. From what? From a true thing she had just asked me for, to my face. The most effective feature of this self-justification is that it lets you look down on someone and feel, in the same moment, that you are looking after them.
Now, this matters, because I am not here to make you a worse friend: sometimes it is real. Sometimes a person genuinely cannot take one more thing that week, and withholding a truth is the real kindness, and the relief you feel afterward is only the ordinary relief of a hard moment passed. I am not saying that every soft answer is contempt in disguise. I am saying that the soft answer can be that; that when it is, it is impossible to distinguish from the real thing; and that you, of all people — the one giving yourself the account — are the worst placed to tell which it was. Hold on to that. Because this kind of lie has a defender far more intelligent than my self-justification in the car, and before we turn inward I have to let that defender make its case. It is the strongest argument the kind lie has, and we met it on the very first night.
The kind lie's best defender is not a feeling. It is an argument, and a good one, and we built it carefully back in the second episode, so let me set it up again at full strength.
It is the consequentialist case for the white lie. It runs as follows. Stop attending to rules and motives and the purity of your own conscience, and look instead at what actually happens. What happened when I told Naomi he seemed great? She felt supported on a difficult evening, and no one was harmed. What would have happened if I had told the truth — that I thought the man was a rebound and would hurt her? An argument, most likely. Her, hurt and defensive. The relationship not ended a single day sooner, because people do not leave someone because a friend disapproves; they simply stop telling that friend things. So: real pain on one side of the calculation, and on the other side no benefit at all, because the truth changes nothing she is going to do. Add it up. The kind lie produced more good and less harm than the blunt truth would have. It was not a cowardly evasion of my duty; it was the right action. And the preoccupation with but was I honest with myself about my motive is just the kind of self-absorbed hand-wringing that a focus on consequences is meant to cure. Who cares what was in my heart? Look at the outcome. The outcome was good.
I want to be straight with you: that is a serious argument, and I cannot simply set it aside. It sees something true that the whole inward turn of this series is in danger of forgetting — that morality is, in the end, partly about what happens to people, and that there is something self-indulgent about agonising over the private purity of one's motives while a real person sits in front of one needing something practical. If the kind lie genuinely spares someone real pain and costs nothing real, then a certain kind of relentless truth-teller — the one who would wound you for the sake of his own clean conscience — starts to look like the less moral person, not the more. The consequentialist has a point, and I am going to leave it standing for the rest of the evening, because it may be right. It may be that the outcome is the whole of the matter and the rest beside the point.
But the argument quietly requires one thing in order to work. It requires the calculation to be accurate. The whole case rests on those two figures — real pain avoided on one side, no benefit lost on the other. The kind lie is the right act if and only if the numbers are what the consequentialist says they are. And now ask the question this series has taught us to ask: who filled them in?
I did. I am the one who scored it. I am the one who decided the truth would change nothing she was going to do — which is a remarkably confident claim about a future I cannot see, and a convenient one, since if the truth would change nothing then there is no point in my paying for it. I am the one who decided the conversation would be merely an argument that helps no one, rather than the thing a real friend says that she remembers and returns to three weeks later when she needs it. I am the one who entered no harm done on the very evening I drove home relieved at my own escape. Every figure in that calculation was entered by the one person with a powerful stake in how it came out.
And here is the careful part: that does not refute the consequentialist. It would be easy and cheap to say, there you are, consequences are nonsense. They are not. The consequentialist can reply, rightly: my theory was never that whatever you happen to calculate is correct; my theory is that the right act is the one that actually produces the best outcome. If you got the sums wrong in your own favour, that is not a defect in consequentialism. It is a defect in you. And that is a fair reply. An honest, perfectly impartial assessor — someone with no stake, weighing Naomi's real interests dispassionately — might run the numbers and arrive exactly where I arrived, and be right. The white lie might genuinely have been the best act available. The trouble was never the theory.
The trouble is that the assessor was me. The person with the stake assessed his own conduct and cleared himself, and the verdict may even be correct — but you cannot tell whether it is correct by asking the person with the stake, because he is precisely the party whose interest is in question. So the consequentialist case does not fail. It simply hands you, very politely, a task. It says: the kind lie is right if the calculation is honest, so go and confirm that your calculation is honest. Confirm that you scored no harm because there was no harm, and not because you wished there to have been none. Examine your own motive, and come back and tell me the calculation is clean.
Which sounds entirely reasonable. It sounds like just what an honest person does. And it is the sentence to be most careful with, because it assumes the one thing we are about to find we cannot assume: that we can examine our own motives at all.
Here is the puzzle the consequentialist has walked us into — the one Sartre handed us last week and then stepped around. To lie to oneself, one would apparently have to be two people. There has to be a liar, who knows the truth, and someone deceived, who does not; that is simply what a lie is. But in self-deception the liar and the deceived are the same person. You would have to know the thing, in order to hide it, and not know it, in order for the hiding to work. Stated flatly, it sounds impossible. And yet I did precisely this to myself in the car. So it is not impossible. Which means we have something wrong about the self, and the interesting question is where.
I have to tell you that there is no agreed answer. A number of philosophers have given their working lives to this exact problem — careful, undramatic thinkers — and they have arrived at genuinely different pictures of what happens when we deceive ourselves, and they argue about it still. I am going to give you the three that matter, because the disagreement is not a detour; the disagreement is tonight's finding. So hold all three in mind, and watch what they have in common.
Begin with the boldest. Two thinkers — Donald Davidson, one of the major American philosophers of the last century, in his essay "Deception and Division," and David Pears, an Oxford philosopher, in his 1984 book Motivated Irrationality — both writing in the nineteen-eighties, looked at the puzzle and said: very well. If self-deception requires a liar who knows and someone deceived who does not, and that is impossible in a single mind, then the mind must not be as single as it feels. It must be divided. Davidson's account is the more austere: the self is partitioned, with the painful true belief on one side of the division and the comforting false one on the other, the two kept from ever meeting, because if they met one would have to confront the contradiction. Davidson is careful — he does not want a little person inside making the decisions; only a division, with beliefs held apart on either side of it. Pears goes further. He describes a sub-system — almost a smaller mind, sealed off inside the main one, organised around the very thing you cannot face, a part of you that knows the truth perfectly well and works to keep the rest of you from discovering it. On Pears's account there really is something like a liar inside you. He knows about the relief, and about the contempt in "she couldn't handle it," and he keeps an accurate record of it; and his entire task is to see that the part of you which does the examining never learns of it.
Now consider what that does to the consequentialist's task. Go and check your motive, he said; audit the account. But on Davidson's and Pears's picture, the checking is done by the main system — by the part of you on the wrong side of the division. The one part of you that actually knows what was in your heart is precisely the part sealed off from the inquiry, because keeping it sealed off is the whole point of the arrangement. You can examine yourself as hard and as honestly as you like; the part that knows is unreachable, and introspection reaches only the part that does not.
That is the first picture. A good many people find it too strange — a smaller mind within the mind, keeping secrets — and so a second camp says: no, none of that. There is no inner liar at all. Its leading figure is the philosopher Alfred Mele, whose book Self-Deception Unmasked appeared around the turn of the century, and his account is deliberately undramatic. There is no second self, no partition, no hidden inner agent. There is only bias. Wanting a thing to be true, or dreading that it is true, tilts the way you handle the evidence — what you look for, what you allow yourself to notice, how charitably you read it, when you decide you have checked enough and can stop. Not a decision you make; a bias operating below the level at which you could notice it. I did not decide to score Naomi's calculation in my favour. I simply wanted, badly, for the comfortable reading to be the true one, and that wanting shifted every figure slightly in my favour, below the level where I could catch it happening.
And note — this is the crucial point — Mele's picture defeats the consequentialist's task just as completely as Davidson's did, but for the opposite reason. There is no inner liar concealing the record from you. The problem is that checking your motive is itself a piece of evidence-handling, and the same bias that tilted the original judgement is right there tilting the check. You sit down to audit yourself honestly, and the wanting reaches into the audit as well, and steers you gently to the conclusion you were always going to reach: no, I looked, my heart was in the right place. The check does not fail because the truth is held in a part of you the inquiry cannot reach. It fails because there is no clean instrument to run it with — the instrument is the very faculty the bias controls. "I examined my motives and they were pure" is not the end of the worry on Mele's account. It may be the bias's finest piece of work.
Two pictures, then, that could hardly differ more — a knowing liar sealed off inside you, or no liar at all and only a bias you cannot feel. And already the strange thing is beginning to happen: they give opposite answers about what you are, and the same answer about what you can do. There is a third picture still to come, stranger than either. But hold what is already here, because it is the pattern of the whole evening.
The third picture is the strangest, and I have kept it until now because it overturns something both of the others took for granted. Both Davidson and Mele assume that self-deception is, at bottom, a matter of belief — either you hold two contradictory beliefs in separate compartments, or a bias bends the single belief you have. The philosopher Herbert Fingarette, in his 1969 book Self-Deception, asked: what if it was never about belief at all?
Here is his thought, and it is subtle, so let me go slowly. Most of what we do, we could put into words if asked. We are engaged in the world — doing things, wanting things, minding about things — and normally we can spell it out: yes, I am angry; yes, I want the job; yes, that hurt. Spelling it out — saying it to oneself in plain words, avowing it — is, Fingarette says, most of what we mean by being conscious of our own lives. And his claim about self-deception is that it is not a false belief you are hiding. It is an engagement you refuse to spell out. There is something you are doing — enjoying the power, avoiding the difficult hour — and the manoeuvre is simply never to put it into words, because to put it into words, to avow it, would break the picture of yourself you cannot afford to lose. You do not believe the lie and you do not disbelieve the truth. You decline, quietly and continuously, ever to state the real thing to yourself. It never gets formulated at all.
And do you see what that does to "go and check"? On the first picture, the truth was held in a part of you the inquiry could not reach. On the second, a bias steered the search. On Fingarette's, you sit down to examine your motives and there is nothing stated there to examine, because the whole manoeuvre was never to let it become a stated thing in the first place. You look for a guilty belief, you do not find one, and you read the absence as innocence — I searched my conscience and there was nothing bad in it — when the absence is exactly the point. The thing was never a belief you could turn up. It was a refusal to speak, and a refusal to speak does not show up under examination as an object; it shows up as nothing. The very emptiness you would take for a clear conscience is what the self-deception was always going to leave.
Now stand back and look at the three of them together, because this is the whole episode, the thing I have been leading toward all evening. Davidson and Pears say the mind is divided — beliefs held apart, and at Pears's end of it something like a knowing part sealed off inside you. Mele says there is no liar at all, only a bias you cannot feel. Fingarette says it is not even a belief; it is a refusal to spell out. These are not three versions of one idea. They genuinely contradict each other; the people who hold them have spent decades unable to settle it; and I am not going to pretend to settle it for them in an hour, on a podcast, between a story about Naomi and the end of the programme. I do not know which is right. No one does, yet.
And it does not matter. That is the finding. Consider what happens to your plan to check — to audit the account, to examine the motive, to confirm it was clean — under each of the three. On Davidson and Pears, the check fails because the part of you that knows is sealed off from the asking. On Mele, the check fails because the bias that bent the judgement bends the check along with it. On Fingarette, the check fails because there was never a stated thing to find, and you will mistake its absence for innocence. Three completely different accounts of the mechanism — and every one of them, by a completely different route, arrives at the same conclusion: there is no procedure you can run on yourself that certifies your own sincerity. You do not need to know which picture is true. The result is the same whichever of the three is correct. That is a strange and powerful kind of finding — not "here is the answer," but "every available answer leads to the same place, so you can stop hoping the answer will rescue you."
And if you are looking for one last way out — suppose you drop the mechanism; I will simply hold myself to the standard, be true to myself, and trust that — remember what we found last week, with Trilling. Even the standard is compromised. "Be true to yourself" is a recent and inward ideal, with no witness to it but the very self on trial. Which is why Trilling's line stands over all three pictures at once: we sincerely act the part of the sincere person. You can perform sincerity, to yourself, with complete conviction; there is no one in the audience but you; and you are also the actor, and also the critic writing the favourable review. The standard cannot catch the self-justification, because applying the standard depends on the same self-inspection that is in question.
So that is where the evening has taken us. You came in with the consequentialist's reasonable-sounding task — confirm that your calculation is honest. And we have now watched, three different ways, the word check fail. Not because you are a bad person, but because of what a self is, on every serious account we have of it. This kind of self-justification is real, and you are giving yourself one sometimes, and you cannot examine your way out of it, because the only instrument of examination you own is the same faculty that produced the justification.
Your turn now — you did not sit down tonight to think about Naomi and me. And this week the case is harder to find, because the kind silences do not stand out in memory the way the blunt words do.
Find your own instance. Not the brave one from last week — the kind one. A time when someone you love was in a difficult stretch and asked you something directly, and you gave them the soft answer. Do you think I'll be all right. Was I a good partner to him. Should I keep trying. Do you think I made the right decision. And you said the warm thing, the encouraging thing, the thing that kept the evening gentle. You have one. Choose the most real — a real person, a real question, a real difficult stretch — and put yourself back in the moment just after you answered, when they looked relieved and you felt whatever you felt.
Now run the two questions. The first is the old one, and it keeps its full standing here, because it is real: was the soft answer the right one? Sometimes it plainly is. Sometimes the hard true thing would have been a gratuitous cruelty, and withholding it was you being a decent friend and nothing more. If that is what it was, good; keep it. This whole line of thought falls apart the moment I pretend that kindness is always a deception.
But now the second question, the one that turns back on you. After you gave the soft answer, what arrived? Was it the ordinary tenderness of having carried something difficult for someone? Or was it that other thing — the light, clean relief that means a difficult moment is over and you have got out of it? And then the worse question, the one specific to this kind of lie: when you told yourself you were sparing them, did you also, somewhere underneath, decide that they could not handle it? If you did, look hard at that, because inside "they couldn't handle it" is a quiet verdict that they are fragile, a person to be managed — and that verdict is remarkably convenient, because judging them too breakable to hear the truth is also what gets you out of the long, difficult conversation you did not want. Was it care? Or was it contempt and self-protection, folded together so closely that they felt like care?
And here is the part that cannot be settled, and tonight we have earned it three times over. You cannot resolve it by looking harder. If there is a sealed-off part of you that knows, your looking cannot reach it. If it is a bias, your looking runs on the same tilted instrument and returns the answer the bias wanted. And if it is a thing you simply never put into words, your looking finds nothing, and you call the nothing innocence. So I cannot hand you a test. If I gave you one — feel for the relief, and if it is there, you were a coward — you would only become skilled at performing a person who feels no relief, and the self-justification would simply absorb the test as well. There is no clean instrument. We do not have one. Not because I am being bleak, but because of what every serious account of a self says a self is.
So let me be plain about what the evening settled, and where it stopped short. It did not tell you that the kind lie is wrong — the consequentialist's case for it is still sitting there, undefeated, and it may be the truth of the matter. It did not tell you which account of your motive was the true one, or even that you were giving yourself a false one at all; perhaps you were simply being kind. And it did not — I want to be careful here — settle the deep question of how we deceive ourselves; three serious people gave us three contradictory answers and I crowned none of them, because the thing they agree on needs no winner: whichever is right, the check fails. What tonight took away is smaller and harder than any verdict. It took away the ability to inspect. You used to be able to say I didn't want to hurt them and let that be the end of it. You leave knowing there is no record you can read and trust — and that this is the opposite of being let off. A person who could audit himself clean would at least know where he stood. You cannot have even that. You are left holding a motive you cannot certify and a self-justification whose edges you cannot see, and no honest way to set either one down.
Which is exactly the wrong place to end a series — and we are not going to. This is the seventh of eight. Next week we stop. And I want to tell you what the finale is not going to do, because you have earned the right not to be taken by surprise by a false ending. It is not going to give you a rule for when the white lie is permitted. It is not going to hand you, at the last moment, the method for checking your motives that I have spent tonight showing you do not have. And it is not going to rank the inward question — was I honest with myself — above the plain outward one — what did I owe her. Because there is a voice we have kept alive since the night this turned inward, and next week it is finally heard: the voice that says the whole inward turn may itself be the evasion — that all this examining of one's own motive may be a way of not answering the plain question one was actually asked. That is the position. I am not going to make its case tonight, because making it is next week's work. Next week we let it speak, and we hand all of it back, sharper than we found it, and then we stop.
For tonight, carry this. The next time you hear yourself say I didn't want to hurt them, and feel that small relief pass through you, do not try to settle it on the spot. You cannot, and a person who thinks they can has already missed the point. Simply notice that you cannot. The sentence that used to close the question no longer does, and that not-closing is the most honest thing available to you — not a clear conscience, and not a guilty one, but a question you have finally lost the tools to answer too quickly.
Thanks for listening. I'll see you next time.