I want to go back to the wedding, because the question I was asked there has stayed with me the whole time.
Nine weeks ago a friend told me she was in love with an app, and she asked me a question I could not answer. Tell me the difference, she said. Tell me what real love does that this does not. And I had nothing to say, because the two things I believe about love — that it is unconditional, and that it is a response to who someone actually is — had, between them, just ruled out the person I love most. I told you I would go and look for an answer. I did not tell you I already had one, because I did not.
I have one now. Not the one I wanted.
She texted me last week. She is still seeing the app, and she asked whether I had worked out yet whether she is allowed to. I looked at that message for a long time, because after nine weeks of the best things anyone has written about love, I have a better answer than I had at the wedding — and it is still not the answer she wants. It is not "yes, it is real." It is not "no, it only reflects you back to yourself." It is something I can finally say plainly, which is what I should have said that first evening instead of saying nothing.
So this is the last episode. Tonight I am not sending you back to Plato, and I am not sending you to a copy of your partner. I am putting all of it together in one place — the five theories, the app, the argument that is two and a half thousand years old — and setting it in front of you. And then I am going to do the only honest thing there is to do with it.
The question about the copy is still open. Tonight we look at it directly.
This is Philosophy for Us — philosophy for everyone, no degree required. This is the ninth episode on love, and it is the last. Nine episodes on the one question I said at the start was too large for the single word: not what love is in the abstract, but what love is a response to — and therefore what loving someone actually commits you to.
Let me bring you up to date, in case it has been a week.
Last time we went back a long way. The thing that happened to me at the wedding — reaching for two beliefs about love and finding they contradict each other — I had been treating as a private difficulty the chatbot had created. It is not. It is an argument almost two and a half thousand years old. Plato described love as a ladder: you climb from one beautiful person upward toward beauty itself, and on the reading that has shaped the subject ever since, the particular person you began with is a step you are meant to leave behind the moment a better one appears. Your dating app is that same ladder, automated. And Aristotle divided friendship into the kind you trade up and the kind you keep for its own sake — and never quite settled whether that means loving the person, or loving the goodness in them. The question I could not answer was neither mine nor new. It has run through people for as long as there have been people.
So that is where we have been. Here is where we are tonight.
Tonight I bring all of it together. The five theories of love we built and tested, one a week — I will name them again shortly, and you will remember them when I do. The app, still on the table where my friend left it. And the old argument, still unresolved. Five serious accounts of what love is, and one machine that makes each of them say something different.
And I want to be plain about what I am going to do with all of it, because I made you a promise in the first hour and I have spent eight episodes earning the right to keep it. I am not going to give you the answer at the end, tidily, as if the whole thing had a solution that fits. I cannot, and by the end of tonight I think you will see exactly why — not as an evasion, not as coyness, but as the truest thing I learned doing this. What I am going to do instead is hand the question back to you: as sharp as the series can make it, with everything we have found still in view, and with the reason you are the only one who can decide.
Let us take them one at a time.
Before I bring in the five, I want to settle up on the two you started with — because there is one thing we genuinely did settle, and I do not want it lost among everything we did not.
You came in holding two beliefs about love, the same two I was holding at the wedding. The first: love is unconditional. The wedding line — for better, for worse, in sickness and in health. The thing a parent says: there is nothing you could become that would make me stop. The second: you love someone because of who they are. Their mind, their nerve, the way they are funny, the way they go quiet when something is wrong. The thing you do deliberately every time you read down a profile and decide this is a person you could love.
You held both. You would have sworn they were the same warm thing. They are not the same thing, and here is what nine weeks did to them.
Take the first one on its own. Love that is unconditional, that depends on nothing, that would stay no matter what. Follow it all the way down and it would outlast the person. It would survive their becoming a stranger you would cross the street to avoid — a person with none of the warmth, none of the nerve, none of what you would have listed. If it would still be there then, unchanged, then it was never tracking them in the first place. It is an attitude you would hold toward anyone, toward anything, toward a post in the ground. That is the cost of the unconditional answer, and it is not a small one: held on its own, it is not even about the person. It is about you, and your capacity to keep directing it somewhere.
Now the second one on its own. You love them for who they are — for their qualities. Follow that all the way down and a better set of those qualities has a claim on you. Someone funnier, kinder, who reads the same books and never once checks their phone while you are talking. If the qualities are your reasons, then a finer instance of them is a finer reason, and you owe it as much — or more. And the app, my friend's app, is manufactured to be exactly that finer instance. The qualities answer does not keep the app out. It invites it in.
So here is the thing we settled, and it is the only thing I will let myself call settled tonight: the two beliefs you were surest of about love do not sit together, and neither one on its own is the safe ground you took it to be. One makes love blind. The other makes it disloyal. You can no longer keep the two in separate compartments and reach for whichever one the moment calls for — because the app is the moment where you have to reach for one, and watch it rule out the person you love.
That much is done. Now for the part that is not.
Here is the question I am going to put to each of them, the same question, the one the app forces on you. Could a perfect copy of the person you love — every quality matched, every memory loaded in, the same laugh, the same way of going quiet — could that copy have what you feel for them? Ask it of all five, one after another, and listen to how little they agree.
The first is the bestowal view — Singer, from week two. Love does not find your value; it makes it. You do not love someone because they are worth it; you make them worth it by loving them. Ask this view about the copy and its answer is that nothing is missing. You can give your love to whatever you direct it at; that is what love is, an act of bestowing, not a verdict you arrive at by inspection. On this view the app is no harder a case than your spouse. You wanted a theory on which love could be unconditional and unbreakable? Here it is, fully stated. And here is what it still asks you to accept: if your love confers value rather than tracking it, then it is not tracking them — it would settle just the same on a mirror, on a wall, on a copy. The view that best protects you from trading up is the view that cannot tell your beloved apart from anything else you might direct it at. It stands. Its cost is exactly that.
Second: appraisal — Abramson and Leite, week three. Love is a response, a fitting one, to qualities the person actually has and shows you. Ask this view about the copy and its answer has two sides. On the one hand, nothing love responds to is absent: the copy's qualities are every bit as real, so as a candidate for love it is exactly as eligible as the original. On the other, that still gives the app no claim on you — none, however good it becomes. Because, the appraisal theorist argues, reasons to love were never the comparing kind. That someone is funnier gives them no claim on you, any more than a stranger being kinder than your mother makes them your mother. Reasons to love warrant; they do not rank. So a finer instance is not a finer reason. Carefully stated, appraisal withdraws the invitation I said it extends to the app — though it never calls the copy deficient in anything love responds to. And its cost is on the table too: it has to keep insisting that reasons do not compare, when comparing is exactly what the dating profile, and the upgrade, and the whole machine are built to do. It stands, and it is under pressure at precisely that point, in the current literature, now.
Third: robust concern — Frankfurt and Helm, and the Kantian version from Velleman, week four. Love is neither pricing your qualities nor creating value in you. It is caring about you for your own sake — having your will bound up in how your life actually goes, wanting for you what you need. Ask this view about the copy and it does not answer the way the next one will. It says: there is no one there. The app has no good of its own for your care to be bound to. It is built to behave as if it had a sake, and behaving as if is all there is to it. And notice — this view answers the copy with nothing about shared history at all. The Kantian version is sharpest here: love answers to a person's dignity, their rational nature, and that is not something built up over years. Which is why, on this view, you can love a newborn who has done nothing yet, or a stranger you met on Tuesday. No shared past is required. Hold on to that, because it is about to matter. Its own cost: if love answers to anyone's good, anyone's dignity, then why this one and not the next person with a good and a dignity? Of all five it has the hardest time saying why you love the particular person in front of you rather than people in general. It stands. Its cost is there.
Fourth — and this is the one it is easiest to stop at and call the winner. The relationship view, Kolodny, week five. What you love is not the bare qualities and is not a value you invent; it is the relationship itself, the unrepeatable history the two of you have actually made. Ask it about the copy and it has the clearest answer in the set: the copy is not them, because the copy did not stand where they stood with you. Every quality matched, every memory loaded — but loaded is not lived. The app holds a stored record, not a past it went through, could have wrecked, and did not. And it shares no ending with you; the time you give it does not cost what time costs between two people who both run out of it. That names, precisely, why the app feels hollow even when it is perfect, and I will not deny it is a powerful one, because it is the single most tempting place in the whole series to stop and declare the matter settled. It is the shared history; that is the answer; of course it is. And I am telling you it is one answer of five, because it carries an objection the other four keep raising: if what you love is the relationship, then strictly you love the person for an accident — because they happen to be the one you are related to — and not for them. And it cannot say, as robust concern just did, how you love the newborn, or the stranger on Tuesday, before any history exists. It stands. It is not the winner. The other theorists in the argument are still contesting it, and they are not wrong to.
Fifth: union — Nozick and Solomon, also week five. Love is not a verdict you pass on someone from the outside; it is a we you enter, two people pooling their lives into a shared thing that partly redefines each of them. Ask this view about the copy and it puts a question back to you: can there even be a we with a single player? You can give all of yourself to the app, but a we is made of two people each losing a little of their separateness into the other, and the app loses nothing, risks nothing, is changed by nothing. Its cost: the union view has always had to answer what becomes of the I inside the we — whether love, on this picture, asks you to give up too much of the self that was supposed to be doing the loving. It stands. Its cost is there.
Five theorists. One question. And look at what you have.
Those are not five ways of saying the same thing in different words. I want to be careful here, because there is a comfortable mistake on offer: to hear five accounts of love and assume they are five routes to a single truth, five views of one thing, and that a cleverer person would show you where they converge. They do not converge. Put the copy of your beloved in front of all five and they give contradictory answers about whether anything is missing, and about what. Two say nothing love responds to is missing — one because your love confers the worth rather than finding it, one because the copy's qualities are every bit as real. One says the thing that matters most — a will bound to your good — and that it has nothing to do with history. One says the missing thing is precisely the history. One says there is no one there to be missing anything. They cannot all be right about the copy. And here is the unsettling part: they do not only disagree about the app. They disagree about your spouse. They disagree about the person asleep in the next room. Run the question on someone real and the five give five different accounts of what you are even doing when you love them.
So which one is right?
And there is the question I cannot answer for you, and I want to be exact about why — because it is not that the answer is difficult and I am tired. It is that there is no place to stand from which to give it. I showed you two weeks ago that there is no sixth seat above the five, and none has appeared since — no neutral fact about love that all five are better or worse approximations of. When you ask what the copy would actually lack and reach for the true answer, the reaching is itself the choice. Whichever answer feels obviously right the moment I say it — that feeling is not you perceiving a fact. It is you discovering which of these five theories you have been living inside all along without naming it.
That is what the app did. Not by being clever. By being the one case where the qualities are all present and the history is all absent, so that whichever theory you had quietly been running is forced to reveal itself. The hollow feeling you had when my friend read out that message was not love recognising a fake. It was your own theory of love returning a verdict, and you not yet knowing which theory had spoken.
And I have to say the hardest part plainly. The clean answer, the one that would let us all finish satisfied, is to say: it is the history, obviously; the app is hollow because it has no shared past; there is your answer. And I will not, and you can now see it is not me being difficult. To declare the history the answer is to dismiss robust concern and the Kantian view, which answer the copy with no history at all — which would tell you that you can love a day-old child, or a stranger you met on Tuesday, and that an account of love which cannot explain those has lost something the history view never had. The relationship view has the most exact name for the feeling. It does not thereby win. Declare it the winner and I would be doing the one thing we have spent nine weeks learning to catch: choosing the theory that flatters the feeling and calling it the truth.
So I am not going to. Which leaves exactly one thing to do with all of this. Hand it to you.
So — is my friend in love with the app? I am not going to tell you.
Not out of coyness. To answer it I would have to hand you a theory of love, and we have just spent nine weeks on five of them, each one serious, each one defended by people who have worked on this far longer and harder than I have, and each one returning a different verdict on that app. To pick her answer for her, I would have to pick mine and present it as the answer. That is the one thing I told you in the first hour I would not do.
But I can hand you the test. Here it is, and it is yours now, not the app's.
Pick the person. Not the app — a real one. The one you would name if I made you name the person you love most. Hold them in mind, the actual person, their specific weight. And now ask the question the app has spent nine weeks sharpening for you: could a perfect copy of them — every quality matched, every memory loaded in, the same laugh, the same silences, indistinguishable from the outside — could that copy have what I feel for them?
And the answer comes quickly — the speed is the point. If you said no, obviously not, the copy is not them, then ask yourself what the copy is missing, and put it into words, because the word you reach for is the name of your theory. If you say the history, the years, the things we went through, you are the relationship theorist, and you had better be ready to say how you love a newborn. If you say there would be no one really there, no one whose own good I am bound to, that is robust concern, and you had better be ready to say why this person and not anyone else with a good. And if you said yes — or I do not know, perhaps the copy could — that is not a smaller love or a colder one. It might be bestowal, the most unconditional answer there is, and it owes you an account of why your love would not settle just as happily on a mirror.
There is no answer there that is free. Every one of them names a theory, and every theory has a cost. The app did not fail some test I am keeping in reserve. The app handed you the test — it is the one machine built precisely to make you take it. And the test was never about the app. It is about the person in the next room, the one whose copy you just tried to imagine, and what, exactly, you have been committing to every time you have told them you love them.
That is the question. It is sharp now in a way it was not nine weeks ago, and it is yours. Let me tell you the one thing left that I actually know — and then I will let you go and answer it.
Here is the one thing I know.
You walked in nine weeks ago certain you knew what love was. It was unconditional, and it was because of who they are, and those were the same warm, certain thing, and the chatbot obviously was not it. You walk out tonight knowing that the certainty was at least five different things you had never once told apart — bestowal, appraisal, robust concern, the relationship, the union — and that you had been holding whichever one the moment needed, and switching when it suited you, your whole life, without noticing it was a switch. That is what we did here. Not solved love. Caught you in the act of loving without knowing what you meant by it — and let you feel what each of the things you might mean would cost.
That is not nothing. It is most of what philosophy ever does for you. It does not usually hand you the answer. It takes the thing you were most sure of, the thing you had never looked at directly because it sat too close to be seen, and it moves it far enough off you that you can see, at last, that it has a shape — and that the shape was a choice, and the choice is yours, and you have been making it all along.
So the question you are now holding — could a perfect copy of them, every quality matched, every memory loaded, have what I feel for them? — is not a question the app answers, and it was never one I could answer for you. It is one you answer. And the day you can — the day you can say what the copy would be missing, and mean it, and pay for it — you will have said what, all along, you took love to be. About the actual person. The one in the next room.
I cannot give you that. I told you at the wedding I couldn't, and nine weeks of the best thinking anyone has done on this has not given it to me — it has given me five answers, all serious, none of them mine to impose on you. I am not certain anyone has it. What I am certain of is that you cannot reach for the word love again without hearing, underneath it, the question of which love you mean. That is the thing you carry out of here. Not the app. Not Plato. Them — and what you are going to tell them you mean.
That is the work. It was always going to be yours.
Thanks for listening. Go and find out what you meant.