A worse world, full stop

Last week I left you at an impasse, deliberately, and I should begin by admitting that was not a comfortable place to leave anyone.

Recall where we had reached. Over the first three episodes we built the strongest version of the conviction that we owe something to future generations: a real harm, a real victim downstream in time, even a genuine disagreement to be had about how much harm it takes. Then we tested that conviction on the case it was built for — the climate future we are all afraid of wrecking — and the victim it requires failed to appear. On the other path, the one where we hold back, the very same future people would never have been born; so no particular future person can say we made their life worse than it would otherwise have been. There is no one there to have been wronged. The one argument we were sure of returned no verdict at all. And I would not let you reach for a consolation. I left you holding the fact that the reason you trusted had come apart.

There was one thing, though, that I said little about, because the collapse was enough to be going on with. The moment that argument failed, an obvious conclusion presented itself — and after two hard weeks it arrived as a relief. You know how it runs. Future people don't exist; they can't be harmed; they owe their very existence to the choices we make; so we owe them nothing in particular. Close the book, we said, turn the heat back up, fly where you like.

A conclusion that convenient, arriving just as the argument collapses, is deeply tempting. All week it has been tempting.

Tonight we examine it — not accept it, examine it. Last week I promised we would not take that way out cheaply, and tonight I keep the promise. We are going to look hard at what actually follows from "they don't exist." And I will say now, so it does not take you by surprise: there is something serious on the far side of it. Not the slogan — a real argument, perhaps the most powerful one in the whole series, that does what nothing we have built so far could do. It survives the very problem that broke us last week, because it never needed a victim in the first place.

And then I will show you what that argument costs. Because it does cost something.

This is Philosophy for Us — philosophy for everyone, no degree required.

This is the fourth of nine episodes. Last week the argument we trusted collapsed: we went looking for the person our conviction needs — the one who could say you made my life worse than it would otherwise have been — and across the whole climate case there was no such person, and the argument fell silent. We stopped there, with one obvious conclusion in front of us: that we owe the future nothing. Tonight we put that conclusion on trial. We find the serious argument that survives the collapse — the way a duty to the future can hold even after the harm is gone — and then we find what that argument costs. Because it costs something. They always do.

Let us read that conclusion more carefully, because I think we have been reading it too quickly.

The words run together into one smooth thought, and that smoothness is exactly where the mistake goes unnoticed. There is no one there to harm, so nothing is wrong, so we may relax. Said quickly, it sounds like a single step. It is not. It is two separate claims, and the whole question tonight is whether the first really supports the second, or is merely carrying it along.

Here is the first claim, and it is sound — we built it ourselves last week, so I will not pretend otherwise. There is no one there we can harm. We established it slowly, following the climate case to its far end, and we found that the person living in the wrecked future cannot say the sentence the harm argument needs, because on the other path there was no better life of theirs to lose. There was no them at all. So: no determinate person, worse off than they would otherwise have been. No victim. That much is settled, and we are not unsettling it tonight.

Now take the second claim, because it follows immediately behind the first and it is not the same claim. So nothing is wrong. So nothing bad has happened. So we may relax. The first claim is about a person — is there someone who got a worse deal? The second is about the world — was what happened bad at all? Those are two different questions, and the inference we are being invited to make is that once we have answered the first, no victim, we have answered the second. No one was wronged, therefore nothing is wrong. Said quickly enough, who would stop to object?

But stop. Because we have drawn exactly this distinction before, in a quite different context, and drawn it cleanly. Think back to the question of free will — whether anyone is ever truly the author of what they do. We found that we could settle the deep question, whether the will is free, and discover that it did not settle the practical one at all — what actually to do on Monday morning. Two questions that had always felt like one. Two levels. Answering the first told us less than we expected about the second. This is the same structure, and we already know how to see it. Whether anyone was wronged is one question. Whether the world is worse is another. And non-identity — the problem that silenced the harm argument — only ever answered the first.

Let me make this concrete, and let me choose an example with nothing to do with climate, so the shape is clear with none of the politics leaning on it.

Consider the last generation. Not a catastrophe — nothing is on fire. Simply the end of the line, a long way off. Suppose humanity arrives, gently, at its final generation: for whatever reason, this is the last one, and they know it. There are no more after them unless they choose it — and they could choose it. They could leave the world in good order, hand on what they were given, let the human story continue. It would cost them something, as it always does. Instead they decide against it. They are comfortable; they let things run down; they live out their lives pleasantly enough and let the whole thing quietly stop. No suffering, no one tortured — they are content, even. They simply shut out everyone who might have come after them.

Now apply the harm argument. Look for the victim — the person worse off than they would otherwise have been. And there is none. The people who would have come do not exist and never will, so they cannot be harmed; one cannot make a never-born person worse off, as we established last week. And the last generation themselves harmed no one — they are content, they chose their own comfortable end. So by the strict rule of no victim, no wrong, there is nothing wrong here. No one got a worse deal than they would otherwise have had. The harm argument returns no verdict.

And now the question that matters — and it is a question of judgement, not of theory. Is that acceptable? The human story simply petering out into a comfortable nothing, when it could so easily have continued? Does nothing bad happen there?

I do not think we can say yes. I have put this to a great many people and never met one who could say, with a straight face, that it simply does not matter — that a universe in which the whole human project ends in a shrug is exactly as good as one in which it continues. The honest answer is that it is a loss. It is worse. Something real has gone out of the world that need not have.

But notice what we have just done. We have judged the outcome worse — and we cannot point to a single person who was wronged. There is no victim in the story; we checked. The badness is not located in anyone's ruined life, because there is no person there whose life was ruined. It is located somewhere else. It is in the world — in the difference between the world that continued and the world that did not, a difference that is bad even though it is bad to no one.

That is what the smooth conclusion was relying on us not to notice. No one was wronged came out true. Nothing bad happened came out false. Both at once. Which means the second claim was never resting on the first. They came apart the moment we examined them directly. Non-identity removed the victim. It did not remove the badness. And if it did not remove the badness, then the conclusion does not say what we took it to say. They don't exist is true. So we may relax does not follow.

So before we accept it, we had better find out where that badness resides — the badness with no one for it to be bad to. Because if it is real, it is present in the climate case too, in the wrecked world we would leave behind: wrong of no one, and wrong all the same. And that changes everything about what actually follows from the conclusion. That is where we turn next.

So here is the thought, and it is the clearest and strongest we have held since the series began. Let me build it carefully, because for a few minutes it is going to hold up under pressure — the first argument in two weeks that does.

Return to what failed. The harm argument ran on a single part: a person who is worse off than they would otherwise have been. Victim first, badness second. On that account the badness simply was the harm to the person. No person, no harm, no badness — that was how it was wired, and that wiring is exactly what non-identity destroyed.

Now reverse it. Suppose we had it backwards all along. Suppose the badness was never located in the person at all — suppose it was in the world, and the person was merely where we had been looking for it.

Here is what that means. Stop asking who was hurt. Ask instead what kind of world we have brought about. Set the two futures side by side — not the people in them, the worlds. One world contains more: more health, more flourishing, more of whatever it is that makes a life good — more good in total, summed across everyone who is there. The other contains less: less flourishing, more suffering, lives more constrained than they needed to be. And now, without looking for a victim, ask the plain question: is one of them worse than the other?

Of course it is. The world with less good in it is the worse world. Not worse for anyone in particular — we have been through that; there is no particular person for it to be worse for. Worse, full stop. Worse in the way the world where the human story ends is worse than the world where it continues. A future with less wellbeing in it is simply a worse outcome than a future with more, and it is worse whether or not there is a single person standing in it with a grievance. The value was never a property of some victim's complaint. It was a property of the world.

That is the whole of it, and it can be put in one line: value belongs to states of affairs, not only to the ledger of who was wronged. A world can be better or worse as a whole, weighed entire — the way a life can be better or worse as a whole — and not only as a column of harms done to nameable people. And once we allow value to be seen that way, the problem changes shape.

Take it back to climate. The reason last week was so painful is that we went looking for the harmed person and found no one. The person in the wrecked future owes their existence to the wreck; we cannot have made them worse off; there is no victim. All true. But we no longer need the victim. We were never owed one. Look instead at the two futures — the one where we hold back and the one where we burn — and yes, they are not one group of people graded better or worse; we know now they are entirely different people. But they are still two worlds. And one of them contains more good than the other. One of them is, weighed whole, the worse world. And bringing about the worse world when we could have brought about the better one — that is the wrong. Not a wrong to anyone. A wrong nonetheless. We do not owe future people a better deal, because there are no fixed future people to owe. We owe the future a better world.

There it is. The duty stands again. And I am not going to undersell it, because this is real. It is not the cheap way out — it is the opposite of the cheap way out. The cheap version said: no victim, so nothing matters, relax. This says: no victim — and it matters anyway, and here is exactly how it matters, and here is why the instinct to flinch was right all along. It rescues the conviction we walked in with. It keeps the future weighty. And it does so while granting Parfit every single thing he proved — yes, no harm to identifiable people; yes, different paths produce different people; yes, the harm argument returns nothing. It accepts all of that and replies: it does not matter. The badness was never in the tally of victims. It was in the world. And the world we would leave by burning is worse than the world we would leave by holding back — and that is reason enough to hold back. No one needs to have been wronged. A worse world is worse. So we should bring about the better one.

Let me be honest about how strong this is, because I have watched many people reach this point, feel the relief, and conclude the series should simply end here. They think: that is the answer. The talk of harm was a clumsy way of pointing at the real thing, and the real thing is the world, and now we have found it, so we can stop. And I understand the feeling entirely. After last week, this is the first secure position we have reached. Many who study this for a living end up holding some version of it — they will tell you that any honest ethics of the future has to be at least partly about the world and not only about the victims, precisely because non-identity leaves no other route.

So let us provisionally accept it, for a moment, and let it be the answer.

And then let me set out the difficulty this argument already carries.

The man who built this argument also identified the difficulty attached to it. The same man: Derek Parfit again — the one who set out the non-identity problem last week and spent his life unable to escape it. He saw, more clearly than anyone, that the impersonal view is where you have to go once the victim is gone. And he saw, just as clearly, what it costs to go there. He gave the cost a name: the Repugnant Conclusion. His own word — repugnant — not a critic's word for his view, but his own, for where his own best reasoning led. That is worth holding onto, because it tells us what kind of difficulty this is. It is not an outsider poking a hole in someone else's argument. It is the author of the argument reporting, in print, that his own best reasoning led to a conclusion he could not bring himself to accept.

Here is that conclusion. Follow the impersonal view where it actually leads. We said the better world is the one with more good in it, summed across everyone in it. More good, weighed whole, is better. Very well — then what is the best world? The one with the most good in total. So picture a world of a few billion people, every one of them flourishing, rich and full lives, about as good as human life gets. A great deal of good in it. Now picture a second world, vastly larger — not billions but trillions upon trillions of people — where every life is only just worth living. Not miserable — that is the crucial point, so we must not soften it into misery. Worth living, but only just: a life with barely enough in it to be, on balance, a shade better than nothing — thin, a little music and not much else, but a life its owner would, if asked, just barely rather have than not. There are so unimaginably many of these lives that when the good in them is added up — all that barely-good, multiplied across all those trillions — the total comes out higher than in the few billion flourishing lives. More good, in sum.

And the impersonal view — the one we were just resting on with such relief — has to say that the vast, barely-living world is the better one. It has to. It contains more total good. By the only measure the view recognises, trillions of barely-worth-living lives outweigh billions of flourishing ones. If we could choose, we ought to bring about the vast one. That is the better outcome.

Now here is the difficulty, and there is no easy way past it. We said more good is better — we meant it; it is what rescued us ten minutes ago. The vast world has more good. So either it really is the better world and we must accept it, or "more good is better" was never quite what we meant, and the position we were resting on has a deep fault in it. Parfit worked at this for years — searched for the flaw, searched for some further principle that would let him keep the impersonal view and still say, no, the flourishing world is plainly better — and he could not find one that held. Near the end of his life he admitted, in print, the same way he admitted the non-identity problem: the search for a way out had failed. The conclusion remained repugnant and remained unavoidable, both at once. That is the bind: one cannot believe it, and one cannot find the way out.

Now the obvious repair, the same reflex as last week: very well, then do not sum the good — average it. Do not ask which world has more good in total; ask which has more good per person, the higher average. And the flourishing world wins that comfortably — billions of wonderful lives, a far higher average than the vast, thin one. Problem solved?

It is not, and here is why — briefly, because the point is only to show that this way out is not free either. Run the average view honestly and it grows a difficulty of its own. If all that matters is the average, then take a world full of people whose lives are wretched, well below the point of being worth living. Now add one more person whose life is also bad, also suffering — but slightly less wretched than the rest. You have raised the average. So by the average view you have improved the world by adding another suffering life, because it lifted the mean a fraction. And there is a stranger consequence still: on the average view, whether your own birth made the world better or worse depends on the lives of strangers you will never meet — people in the distant past whose welfare sets the average against which yours is measured. Whether it was good that you were born turns on the happiness of Bronze Age farmers. That is where averaging leads. It is not nowhere. It is simply a different difficulty.

And the next repair fares no better. Suppose we say a life counts toward the good only once it clears some threshold — some level of being decent enough to be worth adding to the world. That does blunt the vast, teeming result. But it has been shown to buy a result of its own, one with a name as ugly as Parfit's: the Sadistic Conclusion. On such threshold views there are cases where adding people with genuinely good lives — good, but not quite over the threshold you have set — comes out worse than adding people with bad lives, lives of real suffering. The arithmetic ends up preferring to bring sufferers into the world rather than people who are happy but not happy enough. Sadistic, the philosopher who named it called it, and he was not wrong.

I will stop the survey there, because the point is not the catalogue but the pattern, and it should be visible now. Every way of adding up the good across a world — every rule for making "more good is better" precise enough to act on — reaches some conclusion its own holder cannot accept. Sum it: the vast, thin world wins. Average it: you improve the world by adding a sufferer, and the worth of your birth hangs on strangers. Set a threshold: you prefer sufferers to the merely content. There is no clean version. The impersonal view genuinely gets us past non-identity — that was all true — but every form of it carries a cost of this kind, and one cannot choose a form without one.

And here is the thing to notice, because we have met it before in this very series, and the recognition is the point of tonight. This structure — a valid argument, one we cannot fault, leading somewhere we cannot follow. We know it. We met it in the discussion of what we owe a stranger, when each step in the argument for helping was one we could not fault, and together they led to a demand so total it would consume a whole life, and we could find no faulty step and also could not live there. The same structure, in different dress. Then it was one argument about our own sacrifice; tonight it is the value of whole worlds. But the bind is identical: the reasoning holds and the conclusion is unliveable, and no amount of scrutiny makes either fact go away.

So we do not get the clean relief after all. We needed a way out of last week's impasse, and this one is genuine — it is the most solid thing we have, and it really does survive the problem that defeated everything else. But it asks us to believe something we cannot, or to keep reshaping it into forms that ask us to believe other things we cannot. The position is secure, and there is a cost fixed to it. That is the first thing we have recovered, and the recovery is not free.

Let me set down plainly where the episode leaves us, because something real has happened in this hour.

We began at last week's impasse, with one obvious conclusion in front of us and the word relax attached to it. And we did examine it. The first thing we found is that the conclusion does not say what we took it to say. They don't exist — true. So nothing matters, so we may relax — that part never followed. It was two claims pretending to be one, and the moment we looked at them directly they came apart. We can tell them apart now, and it is hard to un-see: whether anyone was wronged is one question, whether the world is worse is a different one, and the argument that removed our victim never touched the second. That is ours now. It is what stops us from taking the easy way out on a technicality.

And then, on the far side, we found the serious argument. The impersonal view: a worse world is worse, full stop — worse of no one, and worse all the same; we owe the future not a better deal but a better world. I will not take it back, because it is the strongest thing we have held. It survives everything that silenced the harm argument; it grants Parfit every point and still stands; many who work on this hold some version of it. But we saw its cost. We followed more good is better to where it actually leads, and it led to a vast, thin world we cannot believe in, and every repair grew a difficulty of its own. So we do not get to keep it clean. It is a real justification with a real fault in it — the first real justification we have found, and already, in the same hour, in trouble.

That is the shape of everything from here, and I would rather you knew it now. Last week the argument we trusted collapsed, once, hard. This week we began to rebuild — and the first thing we recovered both supported the duty and carried a cost we could not accept, in the same breath. That is not a detour. That is the route. There is no justification for this duty that does not cost something to hold. Tonight we found the first, and paid the first price.

Notice what I am not saying. I am not saying the impersonal view is the answer — that we have found what we owe the future and can stop. We cannot; it has a fault in it, and there are further justifications to come, each with its own weight and its own price. And I am not saying the opposite either — I am not sending you back to the easy way out. The serious version of we owe future strangers nothing in particular is still standing, exactly where we left it last week; some careful, clear-eyed people hold it to the end, and we have not refuted it and will not pretend to have. What I am saying is narrower and harder than either: the way out you wanted to take on a technicality is not available on a technicality, and the justification you would hold instead is not free. Neither is free. Hold both. That is the work.

And there is the conviction itself, which I have to mention, because it did the strange thing again tonight. Last week we took apart the one good reason we had, and the conviction did not weaken. Tonight we handed it a new reason — the impersonal one — and then found the fault in that too, in plain view, and the conviction still did not shift. Twice now we have given it a reason and twice watched the reason fail while it stayed exactly as strong as before. I will not say tonight what to make of that. I do not think we have earned it yet, and I will not dress it up as settled. But mark it on the way out, because it is the oddest thing in the room: we keep taking the argument away, and the conviction keeps not going with it.

This is the fourth of nine, and we are one justification into the rebuilding. Here is the next one, named plainly, so you can see where we are going. We have now watched the two largest and coldest answers fail in opposite directions — the bare no victim, so nothing is owed, which is too cold to live by, and the impersonal sum, a better world full stop, which is too vast to believe. And the natural response, when both of those break, is to reach for something between them: not a single victim, and not the whole universe weighed on a scale, but something on the scale of a person. Not the harm done to them — their standing. The thought that whoever ends up living in that future, whoever they turn out to be, has some claim on us — a standing of their own from which to make a demand — simply by being the people who will be there. That is where we turn next. And next week we follow it, and ask whether people who do not yet exist can hold a claim against us at all — or whether the same emptiness that removed the victim removes the claim too. I am not going to tell you tonight which.

For now, here is what to take away — and do not take it on my authority, because the worst way to hold something this strange is on someone else's say-so. Test it yourself this week, on the one case you cannot evade: your own judgement. Sometime in the next few days the thought will occur to you that the world would be better with more good in it — more people thriving, more life, more of the good, added up. It is a warm thought; we reach for it constantly. When it comes, do what tonight taught. Follow it. Do not stop where it is comfortable — follow it the whole way, to the vast world of trillions of barely-bearable lives that has, by that very measure, more good in it than everything you love. And watch your own judgement refuse the arithmetic it has just endorsed. That is the work: a reason you cannot fault, leading you somewhere you cannot go. When you have felt that for yourself, on your own example, with no one handing it to you, you will have met the strongest justification for the future at first hand, and the price it costs to hold. You will have found the fault yourself. It is yours.

Thanks for listening. I'll see you next time.