There is a man — it does not much matter what we call him, because there are a great many like him. He is in his late sixties. His wife died three years ago. His children are good to him, in the way of grown children with their own lives two time zones away: a call on Sundays, a visit at Christmas. He is not abandoned. If you asked him, he would say he was fine.
Consider his week. He wakes and makes one cup of coffee, because there is only himself to make it for. He keeps the television on through the day, for the voices. Most afternoons he goes to the shop, and the reason, if you watched closely, is a particular one: so that one other person will say a sentence to him that is not coming out of a screen — that's three eighty, love — and on some days that is the whole of it. The entire share of another person's voice, from someone actually in the room. By Thursday he has not been touched by another human being in nine days. Not a hand on the shoulder. Nothing.
Here is the part that is less obvious. His doctor can see it — not his sadness, but his blood. The loneliness has reached his body and is doing measurable things there. His blood pressure runs higher than it should. His sleep breaks in a pattern the research has learned to recognise. There is a low, steady level of stress in him that never fully subsides — because for most of human history a person alone, outside the group, really was in danger — and that stress wears, year on year, on his heart, his arteries, and the parts of the brain that form memory. He is, measurably, more likely than the same man with someone in the house to have a stroke this year. More likely to develop dementia. More likely to die, and to die sooner.
None of that is a figure of speech, and I want to be careful about it from the first minute, because the previous two episodes may have primed us to hear loneliness as an account — a word of recent origin, a frame with a history and a politics. It is those things. It is also doing real damage to a real body tonight, and the body does not care that the word is young.
This is Philosophy for Us — philosophy for everyone, no degree required.
This is the third of eight episodes on loneliness, and tonight we take the frame we spent two weeks taking apart and restate it in its strongest form.
Last time, we ran what was in effect an attack on it. We established that the painful feeling has a relatively recent history — the word for it, in the wounding sense meant here, is barely two centuries old — and we found the conditions that produced it: a world that drew people out of their older networks of place, kin and faith, and left each of them standing alone, a single self expected to be sufficient in itself. Then we turned to Hannah Arendt, who argued that the ache is not a fault in the person at all, but what remains when a shared world falls apart. Two episodes, then, spent showing that "loneliness is a sickness to be cured" is not the plain truth about the feeling. It is a position — one reading among others, with a date on it and a view of the world behind it.
So tonight we do what a programme owes its listeners after two episodes like that. We hand that position its ablest defender, and we let the case be made as forcefully as it can be. The people who call this an epidemic are not fools, and they are not selling anything, and they hold the one thing we have not so much as touched in two hours of careful history: the bodies. The measured harm. Their case is that calling the ache "just a young word" or "just a broken world" abandons the people who are dying of it now, tonight, while we talk. If the medical frame is to be questioned honestly — and it will be; that is the whole point of these eight episodes — then it first has to be seen at full strength, with no appeal to the youth of the word to fall back on. So tonight the case is made in full. And it is a strong one. Stronger than you might wish.
Here is where it begins.
Begin with the part that is hardest to dispute, which is the body.
For most of history, to say that loneliness was bad for you was to say something like what you might say about a grey week: that it was a passing mood, unpleasant while it lasted and then gone, with nothing actually broken. What changed — and it changed mostly in the last fifteen or twenty years, which matters, because it means this is not old folk wisdom but something recent and measured — is that researchers began treating loneliness the way they treat smoking, or high blood pressure, or a poor diet: as a risk factor. Something you can measure across a whole population, follow over years, and, at the end, count the deaths.
When they did, the results were sobering.
Here is the finding you have probably heard, and it is worth slowing down on, because it is easy to let it pass as one more health statistic. A researcher named Julianne Holt-Lunstad gathered a large number of separate studies — hundreds of thousands of people, tracked over years — and asked a stark and simple question. Of the people who began lonely, isolated, short on real connection, and the people who began well-connected, who was still alive at the end? The answer was that being cut off in that way raised a person's chance of dying, over the years studied, by an amount close to what smoking does. The summary repeated everywhere afterwards — and it is a fair summary of the work — is that poor social connection is about as dangerous, in its effect on how long you live, as smoking fifteen cigarettes a day.
Stay with the size of that comparison — not how it feels, but how big it really is. We have spent decades, and large sums, and a great deal of law, treating cigarettes as a grave public threat: warnings on the packet, taxes designed to deter, bans in restaurants and offices. We did all of that over a mortality risk that loneliness, on this research, falls in roughly the same range as. If those figures are correct — and they have held up across many separate studies, run by many separate people who were not coordinating — then a society that treats smoking as a serious danger and files loneliness under private sadness is not being rational. It is being squeamish. It is treating the harm that shows up on an X-ray as real and the harm that does not as imaginary, and the mortality figures do not respect that distinction.
And it is no mystery how a feeling reaches down into a body and does this. The mechanism is fairly well understood. In short: a person who feels persistently alone is a person whose stress response never fully returns to rest. The body behaves, at a low and constant level, as though it were under threat — because for most of the time there have been human beings, being outside the group genuinely was a threat. So the stress hormones run somewhat high, day after day; the blood pressure with them; sleep comes broken and shallow and stops doing its repair work; and a slow, low inflammation settles across the system, the kind that wears on arteries, joints and the brain. None of it is dramatic in any single hour. It is the constancy that does the damage — not one severe episode, which the body recovers from, but a low strain that never fully lifts.
Put numbers to where that lands. The United States Surgeon General, Vivek Murthy, was concerned enough by all this to issue a formal public advisory in 2023 — the kind of document a government issues about an emergency — titled, plainly, Our Epidemic of Loneliness and Isolation. It gathers the findings into figures that are hard to set aside. People who are socially isolated carry something like a third again the risk of stroke, and close to a third more risk of heart disease. They fall into depression at markedly higher rates. In older people, the risk of developing dementia comes out raised by something on the order of a half. These are not risks you ignore. If a food additive did this, it would be banned within the week.
And then there is the scale, which is what earns the word epidemic honestly rather than as an alarm. In 2023 the World Health Organization — the body that coordinates the world's response to actual epidemics — launched a Commission on Social Connection, and put Murthy at its head alongside a co-chair from the African Union. Not a working group: a commission, the instrument reserved for something global and serious. When it reported, in the middle of 2025, the central figure was this: roughly one in six people on Earth is affected by loneliness, and it is linked to something on the order of eight hundred and seventy thousand deaths a year. Every year. That is the population of a city, annually, at the far end of a feeling we keep calling private.
Now, to be exact about what this does and does not establish, because the whole series depends on my not overstating it. It does not prove that the medical frame is the only way to read your loneliness — we will return to that. What it does, and I think it does this decisively, is end one habit for good: treating loneliness as merely a story, a matter of words, a thing that dissolves once you have seen how recently it was named. You cannot reason from the date of the word to any conclusion about a stroke. The pain in that man's chest, in the silent house, on the ninth day without human touch, is not waiting for us to finish being clever about etymology. Whatever else loneliness is — and it is several other things, one of which we have already met — it is also, for a great many people, a measurable medical danger. And a measurable medical danger is a thing you treat.
That is the first half of the case, the factual half. Now the part that is harder than the data — the part aimed not at your reasoning but at your conscience, and at mine.
The data is the easier half. The harder half is moral, and it is aimed at us — at this programme, and at you, comfortable enough to be listening to it.
Imagine the same two weeks, but set in a different place. Set it in the cardiac ward, at the bedside of the man from the start of tonight, three days after his stroke. Suppose you explained to him the finding from our second episode: that the word for what he feels is only two hundred years old, that older English had gentler words, that his ache is, in a sense, a product of modern life and of the idea of the self-contained individual. Imagine watching his face as you said it. Or take it to a doctor in one of the new loneliness clinics — they exist now — and explain Hannah Arendt to her: that her patients' suffering is really the symptom of a hollowed-out common world, and that to treat it in the individual is a category mistake. She will agree with every word. And then she will turn back to the individual in front of her, because the individual in front of her is the one who is dying, and the common world is not going to be rebuilt in time to save him.
That is the champion's first and hardest point, and you have to take its full weight before you are entitled to question it. It is this: a certain kind of thoughtful talk about loneliness is a luxury, available mainly to people who are not, at this moment, overwhelmed by it. The history we traced, the Arendt we admired — possibly true in every word, and also the kind of thing you can only sit and enjoy if your own stress response returned to rest at some point today, if someone touched you this week, if your loneliness is the bearable kind and not the lethal one. To a person in the lethal kind, an hour spent establishing that their feeling has an interesting history is not a gift. It is a delay. And delays, in something that kills at the rate we have just counted, themselves cost lives.
Here is the sentence the champion would put to us directly, and I will give it in full, because it deserves an answer rather than a dodge. To tell a suffering person that their loneliness is "the price of being a self," or "just the shape of being one human being and not everyone," is to present abandonment as though it were wisdom. It is to take the most consoling thing the comfortable can say — this is simply the human condition — and offer it, as comfort, to someone who needed a doctor. The phrase "the human condition" has served, over the centuries, to keep the comfortable untroubled by the suffering of others. The champion's charge is that "some loneliness is just part of being a person" is exactly that move, in philosophical language. It excuses us from acting. It turns a treatable killer into a noble and permanent feature of human life, and it does so in language gentle enough that no one notices the dying being quietly given up on.
That does something to the comfortable distance you may have been listening from. Two episodes in, you might have settled into an agreeable thought — I see now, loneliness is more complicated than I assumed; it has a history; it is partly the world's fault. That thought is itself a standpoint. It is the standpoint of someone with the leisure to find their loneliness interesting. The champion is not impressed by it, and says so: the view from which loneliness looks like a fascinating puzzle is the view of a person it is not, at present, killing — and that view should be very slow to lecture the view from the bed.
The moral case has a second part, and it is more uncomfortable than the first, because it is directed at something we have been quietly proud of. For two weeks we have congratulated ourselves on refusing easy answers, on declining to hand over the cheap cure. But the champion asks a cold question in return: refusing to act also has a cost — who pays it? When you are dealing with something that shortens lives at the rate of smoking, "let us not be hasty; let us first examine the frame" is not obviously the brave or honest position. It can be a way of keeping your own conscience clean while other people die. There is a kind of intellectual who would rather ask a fine question forever than risk a clumsy answer that might actually help someone tonight — and from where the champion stands, that is not depth. It is vanity, and vanity that costs lives.
I will be honest: I find this one hard to withstand, because it is partly right about us. It would be easy — it would be the cheap exit, the very one we have refused twice — to answer it cleverly, to say, "but you are assuming the medical frame, which is the very thing in question." And we will say something like that, carefully, in the weeks ahead. But not yet, and not as a reflex, to make the discomfort go away. First we let it stand, unanswered, because the discomfort is information. The champion has made a real point: a frame can be a modern invention, a position with a history, a thing the comfortable find suspect — and still be the one telling the truth about the body on the bed. Young is not false. We taught you that ourselves. The champion has simply picked it up and turned it to point at us.
A weak version of a view fails at the first objection. A strong one survives the best objection you can bring against it. So let me put to the medical frame the three hardest objections this show has already built — the ones some of you may have been holding, half-formed, since the data began. Not one of them unsettles it.
The first objection is the one we spent a whole episode assembling. The feeling is only two hundred years old. The word is modern. So this "disease" is really an invention, a thing a particular world taught us to feel — and you do not treat an invention with medicine. It sounds like a strong objection, and against a careless version of the frame it is. But the careful version answers it directly, and the answer is one we already hold, because we built it together last week. Young is not false. Made is not unreal. Consider scurvy. Sailors died of it for centuries before anyone had the concept of a vitamin, before "vitamin C deficiency" was a phrase a single person could have said. The word was not merely young; it did not exist. Did that make the bleeding gums imaginary? Did it make the limes fail to work? A condition does not wait for its name. It does its damage in the body whether or not the language has caught up. So the fact that "loneliness" is a modern word tells us something interesting about culture and nothing at all about whether the thing it names can kill. The champion grants the whole history and loses nothing by it. The date of the word was never an argument about the body.
The second objection is Arendt's, the one we admired last week. Loneliness is not really a condition inside the individual; it is what a broken world does to people. So treating the individual is a category mistake — you are medicating a person for a problem that lives in society. This one is deeper, and the champion takes it seriously, and then shows that it rests on a false choice. Who said it had to be one or the other? Consider how public health actually works, and has always worked, at its best. Cholera is caused by filth in the water — a structural problem, out in the shared world, in the pipes and the politics. And the response to it was both to build the clean water system and to treat the patient dying of cholera that night. No one stood over a dying man and said, "I am afraid your illness is really structural, so medicine would be a category mistake; do come back once we have reformed the water board." That would not be philosophical rigour. It would be letting a man die for the sake of a tidy diagnosis. The champion says: yes, rebuild the common world, fight the isolation at its roots, I am entirely with Arendt on the cause — and while that long work goes on, I am going to treat the person in front of me, because they cannot wait the decades that rebuilding a world takes. Structural cause and individual treatment are not rivals. To refuse the second in the name of the first is simply to let people die for the elegance of a diagnosis.
The third objection is probably the one nagging at you even now. Fine, but "treat it" sounds so cold, so reductive — as though you would hand a lonely person a pill and call it solved, as if the human need for other people were a chemical imbalance. And here the champion takes the very warmth you thought you were defending and claims it for its own side. The treatment, the champion points out, is almost never a pill. Look at what the frame actually prescribes. It prescribes connection. It builds befriending services, in which a volunteer visits the man in the quiet house so that Thursday is not nine days without a voice. It funds community centres and men's sheds and choirs. It has doctors write what they now literally call social prescriptions — go to this walking group, join this kitchen. It redesigns towns so that the old and the isolated are not stranded behind six lanes of traffic. Ask what is cold about any of that. The medical frame, taken seriously, does not reduce the human need for others to chemistry. It takes that need so seriously that it organises a whole society around meeting it. The reductive, isolating thing is the world we already have; the medical frame is the one arriving with a plan to change it.
So there it stands, having met our three best objections. The data says loneliness measurably damages bodies and shortens lives. The moral case says that calling this "the human condition" abandons the people it is killing. And the frame has answered the history, answered Arendt, and answered the charge of coldness without being weakened by any of them — and in answering the last of the three, it came out looking warmer than its critics. I said at the start it would make a good case. I have now made the best version of it I know how to make, in good faith, and nothing I could set against it tonight knocks it down. Standing here, at the end of this part, the honest report is that the medical frame looks less like a suspect modern invention and more like the one position in this argument that has actually been counting the dead.
There is exactly one thing I would ask you to hold as we move on — not an objection; I have not earned an objection yet; only a question I find I cannot stop asking. Every example that made this case overwhelming had the same shape: the man on the ninth day, the body on the bed, the lethal kind. About that loneliness the frame is plainly, decisively right. The question I cannot answer tonight is whether that loneliness is the only kind there is, or whether the one word covers more than one thing, so that the cure so obviously right for the man in the ward may or may not fit the rest. I am not saying it does not fit. I am saying I do not yet know, and neither do you. We will need that.
Now, what you are left holding — and I will not tidy it.
You may have come in tonight half-hoping we would finish taking the frame apart and let you leave with nothing to answer for: loneliness is just a young word and a broken world; there is nothing in me to fix. You cannot have that exit. It is the easy one, and tonight should have closed it for good. The bodies are real. A frame can be modern, and suspect to the comfortable, and still be the one telling the truth — and about the lethal kind of loneliness it plainly is. Anyone who dismisses the medical frame after tonight is not being deep. They are being cruel with a clean conscience.
But the other exit is just as easy, and it is the one you are now tempted by. Having seen the ablest version, the tempting thing is to give in: all right, you have convinced me, loneliness is a disease, hand me the prescription. Not yet. The force of it is real — the case is that strong. But strong is not the same as complete. There is a step from this kind of loneliness is a treatable killer to every ache that calls itself loneliness is the same kind of thing, answerable to the same cure — and the second is a larger claim than the first. Tonight was about the first. Whether the second holds is exactly the sort of thing you learned to watch for on your first Friday night.
Where we honestly stand, then, is this. Two serious positions, and tonight neither one has lost. Arendt's account, still standing, saying that some of this is the world. And the case for public health, just as strong, saying that some of this is a killer in the body, and shame on anyone too refined to treat it. Neither can be crowned.
Next time we take up the one assumption no one has thought to question: that the answer to being alone is, simply, less of it. More company. More connection. We pick that up and ask whether the whole thing might run the other way.
Thanks for listening. I'll see you next time.