Go back to the kitchen table one more time. On the very first night, I described a friend of mine. She and her partner had sat down one evening with a spreadsheet open between them — actual columns, actual numbers, money in and money out, the years it would cost set against the years it would give back — because they were serious people, and this was a serious decision, and that is what serious people do. They were trying to work out, responsibly, whether to have a child. About an hour in, she closed the laptop, because the exercise had begun to feel obscene to her, and she could not say why.
Nine weeks on, we can say a lot about why. She was trying to do right by a person who was not there, and would only be there if she decided yes — and the simplest way of doing right by that person, the one she reached for first, better or worse for the child, turned out to have nothing in it to measure. That accounts for much of what felt wrong. The rest we will gather up before the end.
Tonight I want to point to the one thing we have walked past for nine weeks — the thing so ordinary it became invisible. Not what was in the spreadsheet. The spreadsheet itself. Two thoughtful people took the oldest thing living things do — something that had happened countless times before anyone wrote a word about it — and decided that the right way to face it was to build a case: to sit down and justify it, on the child's behalf, with a ledger, the way you might justify taking on a mortgage. Where did we get the idea that this is what having a child is — a private decision you argue yourself into, weighing costs and benefits for someone who does not yet exist?
It is a strange idea. It is also a recent one. And once you see how recent it is, the question we hand back at the end of tonight becomes harder, not easier — because you will not be able to retreat to the place most people retreat to. You will not be able to say: well, of course, you simply weigh what is best for the child. You will have learned that the simplest version of that weighing — better or worse for the child — has nothing in it to measure, and that the habit of weighing it in this way, privately, on a ledger, is younger than the electric light.
This is Philosophy for Us — philosophy for everyone, no degree required. This is the ninth of nine episodes on whether to have a child, and it's the last one.
Last time, we examined the idea that a life is a gift. You may have expected it to give way easily — the gentlest of the positions we had lined up, the reassuring thing people say once the arguing is done — and it did not give way. It turned into the deepest reply anyone has to the consent problem: that the one thing you could not have agreed to is the very thing that made you someone capable of agreeing to anything at all, so the demand to be asked first, followed all the way down, asks for something impossible. And at its full height it met the harm at its full height — David Benatar's claim that there is something bad in every life. The two turned out to be describing the identical act, one person brought into existence who never asked, and calling it, both of them truthfully, a gift and a harm. We named no winner. You were left holding the most consequential choice there is, with the simple comparison broken, the repair contested, and two true and opposite descriptions of the same act.
Tonight the series does the last two things it owes you. First, where we got the idea that this is a private decision you sit down and justify — because that habit of mind is younger and stranger than it feels, and seeing where it came from changes what it can claim over you. And then we hand the whole question back to you, for good, with the one tool your own reasoning has been using since week four. Tonight I give it a plain name and hand it back to you properly, and I tell you where you will need it next. And it is larger than one child you will never meet.
Let us find out where the spreadsheet came from.
Here is a fact that should strike us as stranger than it does. For almost the whole of the human story, nobody decided to have children. Children happened. They arrived as the ordinary, unremarkable continuation of a life that was already pointed that way. You married, or you did not; you shared a bed, and children came, and more children came, and some of them lived. There was no sitting down to it. There was no spreadsheet, because there was nothing to put in it. You did not weigh having a child against a holiday or a career or a smaller flat, any more than you weighed whether to keep breathing. It was fate, or it was duty, or it was simply what a household was for — the continuation of a family, a name, a farm, a line. And notice what matters most about that world, for us tonight: in it, you cannot run a cost-benefit calculation for the child, because there is no choosing. The question we have spent nine weeks on — should I — could not even be asked. There was no should. There was only did.
So how did did become should? How did a fate become a decision?
The short answer is that we gained control. Across the nineteenth and twentieth centuries, in fits and starts and then all at once, reliable contraception arrived — and with it, something no human beings before us had really had: the power to separate the having of children from the ordinary conduct of a married life. For the first time on a large scale, procreation stopped being something that befell you and became something you could choose. You could say no, reliably, for years, and still have the rest of the life. And once something becomes a choice, something else follows. The moment you can choose, you are liable to be asked to justify the choice. A thing that simply happens to you needs no defence. A thing you elect, you can be called to account for — to others, and, harder, to yourself. The freedom to decide and the demand to justify the decision arrive together. We gained the one, and the other came attached to it.
And here is the difficulty this whole series has been about. The demand to justify the decision — make the case, on the child's behalf, that this is the right thing to do for them — arrived at the very moment we gained the power that makes that demand impossible to meet. Because the justification it asks for is a justification for the child, and we have spent nine weeks finding that the simplest version of that has nothing to measure. The same century handed us the duty to argue this on the child's behalf and the power that empties the argument of its object. We are the first people ever required to justify having children, and we are required to do it in the currency the simplest version reaches for — better or worse for this child — which the nature of the choice has quietly emptied of anything to count.
Then the child changed too — changed in meaning. The historian Viviana Zelizer wrote a book whose title states the matter: Pricing the Priceless Child, published in 1985. Her finding, in brief, is that across that same stretch of history the child underwent a striking reversal of worth. The old child — the child of the world where children simply happened — was, among other things, an economic asset. Hands for the farm, a wage from the mill, security in old age; a child was, in part, something that paid. Over a few generations that reversed completely. The modern child became economically worthless — a pure expense from cradle to graduation, earning nothing — and, at the same time, emotionally priceless, treated as beyond any price one could name. Both at once. That is the pattern that underlies every modern conversation about this. It is why "children cost a fortune" and "you cannot put a price on a child" come from the same person in the same breath without seeming to contradict. The child became a line in a budget and a sacred object at once — something you must plan for financially and something it seems obscene to plan for financially. No wonder the spreadsheet felt obscene. It was trying to price the priceless, which is precisely the difficulty the last two hundred years have built for us.
And then, to complete the circle, the matter returns to something like where it began — but in a new form. We spent a whole episode, two weeks ago, on this: the elective, planned child, the one you are meant to justify privately on a ledger, has lately become the object of public management. Baby bonuses. Fertility targets. Medals for mothers. A government with its hand on your timeline, treating the national birth rate as a figure to be steered and your own body as one of the levers. So we have come the whole way round. We began in a world where having children was a duty — to the family, the line, the continuation. We turned it into a private choice you justify to yourself. And now, in 2026, it is being turned back into a duty — not the inherited kind, the old obligation to the dead and the unborn, but an administered kind: a duty to the state's accounting rather than your own. This time the duty is one the state lays on you.
Now, I have to be careful here, because there is a cheap inference on offer, and I am not going to let us make it. It would be easy to hear all that and conclude: so the whole idea of deciding this for the child's good is merely a recent invention, a two-hundred-year-old habit dressed up as timeless wisdom — so I can set it aside; it was never real. Do not. That does not follow. Showing that an idea is recent does not show that it is wrong. Reliable medicine is recent; it is still right. The germ theory of disease is younger than some of the buildings in your town, and it is true. That "weigh what is best for the child" is a habit our great-great-grandparents mostly lacked tells you it is contingent — that people lived and bore children and loved without it — but it does not tell you it is false. Perhaps we learned something. Perhaps the old fatalism was the error, and the modern demand to take responsibility for the lives we bring about is hard-won moral progress, even if its simplest form runs into the non-identity problem.
So the genealogy does not, on its own, refute anything. That is not its purpose. What it does is more useful, and quieter. It takes the frame you had assumed was simply given — of course you sit down and justify this, on the child's behalf; what else would you do — and shows that it was made, and made recently, and that you can trace where it came from if you look. It does not remove the frame. It stops you treating it as a permanent given. And it fits with everything else we have found: the simple comparison has nothing to measure, and the demand that you perform that comparison is younger than you supposed. Two separate findings, consistent with each other, neither one displacing the other. You do not get to weigh this on the child's behalf in the simplest way — and the conviction that you must weigh it in that way turns out to be a recent and particular standpoint, with a history, and now with a public budget behind it. Both are true. Hold both.
Now the tool itself. You have been using one all series, and it is time to know it by name, and to know exactly what it can and cannot do.
Think back to week four, the hardest episode. I took you slowly through a set of cases — the couple advised to wait a few months for a healthier baby; the fourteen-year-old who has a child far too young — and somewhere in there you said the thing yourself: that when your choice is what determines which child exists — not how a fixed child's life goes, but whether this particular one is the one who gets a life at all — then "better or worse for them" no longer applies, because there is no same person, otherwise, to be better or worse than. I gave you the formal name once that night. Philosophers call it the non-identity problem; it comes from Derek Parfit, in Reasons and Persons, published in 1984. Tonight I want to give it a plainer name, the one that does the work — call it the swap — because that is the move: change the choice, and you do not improve a person or harm a person, you replace them with a different person, and the comparison you meant to run has no one to run it on.
That is the tool, and tonight I hand it back to you to keep. But I hand it back with a clear statement of its limits, because it is precise and it is narrow, and almost everyone who picks it up applies it too widely and draws a false conclusion.
Here is the mistake. The mistake is to take the swap and conclude: so there is no one there, so nothing matters, so do as you please — to hear "the comparison has no one to run it on" as "no one can be wronged, no one can be given a gift, the whole question was a confusion." That is the one conclusion tonight rules out, and I want to be as plain about it as I have been about anything. The swap does not say that. It settles one specific thing, and settles it for good: the comparison. The idea that you could justify this by showing the child will be better off than they would otherwise have been — that fails, because there is no otherwise-version of them to compare against. Everyone serious agrees the comparison fails. But that is all it settles. It does not touch the thought that you might wrong someone by bringing them into a bad life. There are careful philosophers, Elizabeth Harman among them, who hold that you can drop the comparison entirely and the wrong remains — that causing a person to exist in a harmed condition can be a harm to them, with no better-off version needed anywhere in the picture. I told you in week four that I had leaned on that reply and could not defeat it. I have not defeated it since, and I am not going to stand here at the end and quietly pretend I did. It is still standing. It may simply be right.
So use the tool within its limits. It does not dissolve the whole question. It settles one precise thing: that the simplest way you wanted to settle this — the comparison, better for this child than the alternative for this child — has no purchase, ever, on a choice that determines who exists. That is the finding. Beyond it, everything remains open. Whether you can wrong the one you would make in some non-comparative way — open. Whether a life is a gift you would be giving — open. Whether you owe the future a yes — open. The swap removes one kind of faulty reasoning from the table. It leaves more of the question open, not less, because now you must face the harder grounds without the easy one to fall back on. Anyone who tells you the non-identity problem means "so nothing matters, do as you like" has taken a narrow result and stretched it into a conclusion it cannot bear.
And one more thing — the reason I put it in your hands rather than just mentioning it exists. You will need it again, and soon, on something much larger than a single birth. Everything you did tonight you did with a single person — the one you might or might not bring into being. But the same move operates at scale. There is a question coming, later in this whole project, about what we owe the people who will live one and two hundred years from now — the billions who will exist, or will not, and who will be these particular billions rather than those, entirely because of what we do now. Our energy, our emissions, our choices this decade do not merely make the future better or worse for a fixed set of future people. They determine which future people there are. So the swap waits there too, at the scale of whole generations, and those who try to reason about the future without it walk into the same difficulty you met in week four, and make the same over-extension. So this is not a curiosity about your own kitchen. It is a tool you will carry into the largest moral question there is. Tonight you applied it to one person. Next time it is everyone. Learn its correct use now, while the stakes are one life and not a world.
So return, one last time, to the question that would not leave you alone. Should I have a child. You came in meaning to answer it in the sensible way — work out whether you could give a child a good life, and decide for the child's own sake. Over nine weeks we took the simplest version of that away from you. Not by talking you out of wanting it, but by showing that the child you meant does not exist until you say yes, and would be someone else entirely if you said anything else — so the comparison you reached for first, is this better or worse for this child than it would have been for this same child otherwise, has nothing to read. There is nothing for it to measure. And that holds no matter how long or how carefully you consider it, because the trouble is not that the measurement is difficult. The trouble is that there is nothing there to measure.
Now hold on, because here is what the failure of that comparison did not prove, and you will be tempted — everyone is tempted — to let it prove too much. It did not show that no one can be wronged, or given a gift, by being brought into existence. I said this a few minutes ago; I will say it once more, here, where it matters most: perhaps you still hold that a life worth living is a real good to the one who lives it, and that a person can be wronged by being drawn into a bad one. There are careful people who think the swap only ever caught the simplest form of that thought — that you can drop the comparison and keep the wrong — and we did not defeat them, not in week four and not tonight. So do not leave saying there is no one, so nothing counts. That is the one move this whole series rules out. The comparison failed. The question did not.
And here is what else we did not settle, and will not. We are not going to tell you that the absence of a person is a mercy — that the kindest thing is the child never made — even though one of the most careful thinkers ever to work on this argued exactly that, and we could not defeat his argument cleanly, and neither can the field. And we are not going to tell you that a life is a gift and therefore you should give it — even though that may be the truest thing anyone has said about being born, the gift of a beginning, the very gift you are using, this moment, to ask the question at all. Both describe the same act: a person brought into existence who never asked and never could. One calls it a harm. One calls it a gift. They are looking at the identical thing, in full, and neither is the other's mistake. We named no winner last week, and I am not going to install one tonight.
So we hand it back, and it is heavier than when you picked it up. You are going to make this decision — to have the child, or not — and the simple way of making it on the child's behalf is gone, because there is no prior child whose otherwise-life you could weigh. So on what ground do you make it?
If the ground is your own meaning — that you want this, that a life with a child in it is the life you mean to have — then very well; but notice what has happened. The child's good is no longer doing the work. It is your own meaning you are serving, openly, in your own name. That is allowed. It may even be more honest than the spreadsheet ever was. But you no longer get to call it selfless.
If the ground is a debt — to the species, to the future, to the world you were handed by the dead and owe forward — then remember what we found two weeks ago: the body asked to pay that debt, the one that can be scolded and bonused and given medals into paying it, is part of the question, not the answer. A debt may be real. But a debt collected by the state out of your particular life is something to examine very hard before you call it your own.
And if the ground is the repaired version of the child's good — that you can wrong, or give a gift to, the one you would make, even with no other version of them to compare against — then you have kept the hardest thing the whole series left standing, the thing we could not defeat. Good. Hold it. But you still owe an account of what a good life, owed to a person who does not yet exist, actually amounts to — because that account is exactly what the simple comparison used to supply for free, and you do not get it for free anymore.
And if the ground is no one's — if you decide that a choice for no one is still a choice you are entitled to make, on no one's behalf but your own — then a choice for no one is still the most consequential one you will ever make, and you must decide, yourself, whether that emptiness is a release or the heaviest part of the whole thing.
And I am not going to tell you which. That is not coyness, and it is not that I have run out of things to say. It is that there is no answer I could hand you that would not be a lie about how settled this is — and a lie about how settled it is would be the one real betrayal open to me at the end of nine weeks. The honest report is this: you are standing somewhere real, with a decision in your hands, and the methods you trusted to make it either failed or turned out to have a history, and the grounds that remain are all live and none is safe. That is not nowhere. That is the actual place. Most people never find it; they decide from inside one comfortable position or another and never see how unsettled it is. You saw it. That is what tonight was for.
So here is what I would have you carry out, and it is the same thing I asked you to carry out of the very first night, only now you know why. The next time this question comes near you — your own, or someone you love wrestling with theirs — do not begin with what would be best. Begin with the question underneath it, the one my friend struck by accident with her laptop open on the kitchen table and could not put into words. Who am I deciding this for? And are they here to be asked? Watch what happens to the whole decision when you cannot find the person at the centre of it. That is not confusion. That is the real weight of the most private choice you will ever make, uncovered at last, with every comfort removed.
That is the end of this series. Thanks for listening. I'll see you next time.