Months later, the woman from the first episode is sitting across a desk from someone whose job is to help her, and she is about to be handed the thing this whole episode is about.
You may remember her. One in the morning, the phone lit, seven of the nine questions on the quiz ticked, and then the doubt: was I always like this, or did the name make me this? She did not sleep that night, and she did not let the question go. She is the kind of person who keeps working at a thing. So she read, and she talked to people, and she found what most people find: the moment you say a doubt like that out loud, two groups form, and each is certain, and they are certain in opposite directions. One group told her the name was a gift, that the doubt was only her old shame talking, and that she should accept the diagnosis and be done. The other told her the whole category was invented — that her grandmother had suffered worse with a fraction of the vocabulary — and that she should manage and stop reading about it. And she moved back and forth between them for months: some mornings sure it was real, a true condition with a name; some nights sure she had talked herself into an illness to avoid the work of her own life. Two answers, each apparently forced on her, and no way to settle between them.
And then the person across the desk says something quiet, almost in passing. She says: you keep trying to decide whether you are ill or whether you are fine. Those are not the only two options. Some of what you have described is probably a genuine malfunction — something that really does misfire, and that we have real ways to help. And some of it is probably just your life being hard in the way lives are hard, which picked up a clinical word along the way because clinical words are the ones we hand out now. You were never meant to choose one and accept it whole. The work — the only work — is telling which part is which.
And the woman lets out a breath.
That breath is worth attending to, because it is the most honest thing that has happened to her since the quiz. It is not relief that she is ill. It is not relief that she is fine. It is the relief of a person who has been told, for the first time, that she does not have to choose between two answers that both felt false — that the true shape of the thing is both, in some mixture she can work out for herself, case by case, rather than swallowing a single verdict. The pressure of the forced choice lifts. After months of being pinned between two answers, someone has finally given her a way to stand between them. And I want to be exact here, because being fair to this is the whole task tonight: she is right to feel that relief. She is not being fooled. This is the most mature thing anyone has said to her, and it is true, and tonight I am going to set it out at full strength and mean every word.
But notice — she will not, tonight, and you might — what she was actually handed, underneath the relief. She was handed a task: find a line. The gift was this: there is a boundary running through you between the part that is genuinely broken and the part that is merely hard, and your job is to locate it. And she is so grateful to have a definite task at last that she does not stop to ask the one thing standing directly behind the relief, the way a second question stood behind the first one months ago. She does not ask: who decided that the thing to do with a life is to find the line of the broken running through it? She simply takes up the task and begins measuring, and feels, for the first time in months, that she is doing it right.
Hold that task in mind. We are going to use it for the whole hour. And then, near the end, we are going to examine the assumption it rests on.
This is Philosophy for Us — philosophy for everyone, no degree required. This is the fifth of eight episodes on the most personal question this show asks: is something wrong with me? Last time, I left you in a bind, and I did it deliberately. You had the hard-headed answer — that something can be genuinely, mechanically wrong with a person, as real as a failing kidney — and you could not put it down. And you had the opposite answer — that these categories are partly made in the naming, that they spread across a generation, that a name can shape the very person it names, and that the quiet question of who gets to say has real force — and you could not put that down either. Two answers, both with something to them, both bearing on the same problem. And I told you that you would soon start looking for a compromise — the thought that surely the truth is somewhere in the middle — and that there was such a middle, and that I was refusing to give it to you yet, because the relief it offers is real, and resting in it would be a mistake, and you had to sit with the bind to see why.
Tonight I give you the middle. This is the mature answer — the one the woman was just handed — and I am going to set it out at full strength, in two different versions, and let you feel the relief she felt, because you have earned it and it is real. And then, near the end, I am going to show you the assumption underneath it. Not a verdict tonight. Tonight, the best the question can do — and the first sign of what it cannot do, however well you do it.
The person who worked out this answer has a name, and so does the answer, and once you have both you will notice it everywhere, because it is the position most thoughtful people arrive at on their own.
His name is Jerome Wakefield. He is neither a debunker nor a cheerleader. Over several years — the central statement is a 1992 paper in the journal American Psychologist called "The Concept of Mental Disorder" — he tried to say carefully what we are claiming when we call something a disorder. What he arrived at is designed precisely to end the dispute you have been caught in for two weeks. He calls a disorder a harmful dysfunction. Two words, and the point is that each word does one of the two jobs you could not do at once.
Take dysfunction first. This is the half you met in week three, the realist's half, and Wakefield keeps it. There is, he says, such a thing as a part or a process inside you failing to do the job it was built by evolution to do. The heart is for pumping; a heart that cannot pump is failing at something it was genuinely shaped to do, and that is not a matter of opinion. The mind, Wakefield says, is no different. Fear is for something: it is a response that evolved to fire in the presence of threat and to subside when the threat is gone. A fear system that fires in an empty room and will not switch off is a mechanism failing at its job, in the same way a knee that gives way on level ground is failing at its job. That is the dysfunction, and it is a fact about you, not a vote. This is the half that lets Wakefield agree with the realist that some suffering is genuine breakdown — that the brother from week three, nine days unable to get out of bed, the toast untouched on the plate, every system shut down, is not a social construct and never was.
Now the second word, harmful, and this is the half Wakefield takes from the other side, the constructionists you met last week. Not every dysfunction is a disorder. Your appendix does almost nothing; a quirk in your eye leaves you slightly colour-blind; there are mechanisms throughout your body that misfire, or never worked properly, that no one would call illnesses, because they cost you nothing your culture counts as a cost. For a dysfunction to rise to the level of a disorder, Wakefield says, it has to harm you — and harm is judged against the standards of the world you actually live in. That is the constructionist's central point, kept and honoured: yes, values enter the judgment; yes, the culture is in it; yes, what counts as harm in one time and place is not what counts in another. Wakefield does not deny an inch of it. He simply says it is the second half, and that it does not swallow the first.
Now consider what the two words do together, because this is the part that made the woman let out that breath. All real — the realist's claim — turns out to be only half right: real dysfunctions exist, but a real dysfunction that does you no harm your world recognises is no disorder. All constructed — the sceptic's claim — turns out to be only half right too: the harm is indeed judged by cultural values, but values judging the harm of a genuine breakdown is not the same as inventing the breakdown out of nothing. Neither claim was ever the whole truth. And — this is the key point — the two sides were never arguing about the same thing. The realist was describing the dysfunction. The sceptic was describing the harm. Each had hold of one word of a two-word idea and was asserting it as though it were the whole. The argument you could not win was one you never needed to have. You were handed a forced choice — all real or all invented — and a forced choice is exactly the thing you learned, early in this series, to stop accepting on sight. You may remember the first time you saw that our loudest arguments often run on forever because each side is holding half of something that was only ever true in halves. This is that pattern, at full weight. Wakefield takes the two-week dispute, examines it, and shows it was a false choice from the start — and the relief of that is considerable, and it is earned.
And it does real work; it is not just a soothing phrase. Run it across a few cases and watch it sort them cleanly, where two weeks of argument could not. The terror that overwhelms a soldier in a quiet supermarket years after the war: a fear system firing with no threat present — dysfunction — and it is wrecking his marriage and his sleep — harm. That is a disorder; treat it. The boredom of a teenager in a class designed to bore him: no mechanism is failing — the boredom is working, it is the correct reading of a boring room — so however unwelcome it is, calling it a disorder is the over-reach the sceptics warned about. Not a disorder; change the room. The grief of a man whose wife has died, in the second week, weeping, unable to work: a mechanism doing exactly its job — grief is the built response to losing what you love, it is meant to hurt like that — so it is not a dysfunction and not a disorder, even though it harms, even though from the outside it looks like the depression beside it. Notice what has happened. Three cases that the all-real side would have stamped ill and the all-invented side would have stamped fabricated, and the harmful-dysfunction test separates them and puts each where it belongs. Real disorders exist, and we over-apply the word constantly, and the one test tells you which is which. That is not a fudge. It is the most serious account of "is something wrong with me?" anyone in this series has offered, and the woman was right to take hold of it.
So before I question it — and I am going to, twice — stay inside it a moment, as she did. The two-week argument is over. You do not have to be the brave believer or the hard-headed sceptic. You get to be the careful one, the one who says: some of it is real and some of it is over-naming, and my job is to tell which part of my own suffering is which. That is a livable answer. It is, I think, the answer most reflective people reach and settle on. And this answer feels more secure than any position you have held in this series. Which is exactly why, in a moment, I am going to ask you to rely on it yourself, on your own case — and then to look hard at what it rests on.
Now use the test where it counts. The work of the next few minutes is yours, not mine.
Take the word you brought into this series. Anxiety, depression, ADHD, trauma — the one you have, or suspect, or keep returning to. Last week I asked you to run the looping question on it: did you begin to live under the word? Tonight, a different and more useful question — two questions, in fact, the two halves of Wakefield's test, asked in order.
First question. Is something built-in actually misfiring? Not "does it hurt," but: is a mechanism failing at the job it was shaped to do? Is your fear firing when there is nothing to fear and refusing to stand down — or is it firing at things that are, if you are honest, genuinely frightening, which is exactly what fear is for? Is your attention unable to settle anywhere, however you steer it — or is it doing the sensible thing and refusing to settle on work that is boring or pointless? Is your mood flat against the facts of your life — the people who love you, the morning light, the things that used to move you, all gone dull for no reason your life supplies — or is it low because your life at present is, in plain fact, hard to be in? That is the first question, and notice that it is a real question with a real answer, not just a report of your feelings about yourself. Ask it honestly and you can often tell. Sometimes the answer is yes: something is genuinely misfiring. Sometimes the answer is no: the machinery is fine, and it is the situation that is bad. Telling those two apart is not a small thing. It is most of what a good clinician is for.
Second question, and only if the first was yes. Is it harming you — costing you, by the standards of the world you actually live in, things that world counts as worth having? Work, sleep, love, the ability to get out of bed. A misfire that costs you nothing is a quirk, not a disorder, and you do not owe it a diagnosis. Two questions, then. Is something built-in actually failing? And is the failure costing you a life your world would recognise as a loss? Run both, in order, on your own case. And notice — I think you will — that you can get further with those two questions than you got with two weeks of all real against all invented. You have a real instrument now, and it genuinely sorts. That is the relief, and you came by it honestly.
Now here is a case that shows the instrument working and, in the same motion, shows the first sign of trouble — and I have chosen it deliberately, because almost everyone meets it in a life.
Someone you love dies. For weeks you are flattened. You cannot work, you cannot sleep, food is tasteless, you cry in the car, nothing brings any pleasure. Run the test. First question: is a mechanism misfiring? The answer is no — a clear, clean no. Grief is not a malfunction. Grief is the built response to losing what you love; it is meant to hurt exactly that much; a person who felt nothing when his wife died would be the one with something wrong. The machinery is working. So: not a disorder. The test has just spared you a cruelty — being told, in the worst month of your life, that your sorrow is an illness to be cured rather than grief for someone you loved. The instrument works.
Except: for how long?
Because here is what no one tells you about that clean answer. Grief that is working at six weeks — is it still working at six months? At a year? At three years, still unable to work, still flat, still unable to get up? Somewhere along that line, most people want to say, the response that was for something has tipped into a response that is now failing at something: the same flattening that was healthy in the second week has become a depression by the second year. But where? Where exactly does the mechanism cross from doing its job to failing at it? And the honest answer is that no fact about grief itself tells you. The line has to be drawn. And we know it is drawn, because we have watched the people whose job it is draw it, erase it, and draw it again. For decades the standard American diagnostic manual, the DSM, largely blocked the diagnosis: it told clinicians not to call it depression in roughly the first two months after a death, except in severe cases. Grief had a protected window, a rule that said this is working, leave it alone. Then, in 2013, a new edition removed the rule. It judged the window arbitrary — real depression can begin the day after a funeral, and the calendar should not shield it — and many serious people called that a mercy, while as many others said it had just licensed calling the second week of a widow's grief a treatable illness. Same grief. Same mechanism. The line moved, because a committee moved it.
Hold on to your relief — it survives this; the test still sorts most cases, and that is real. But you have just watched the first half of the tool, the half that was meant to be the fact, the misfire you could read off like a failing kidney — you have just watched it turn out to contain a judgment, right at the hardest case: a point where "is the mechanism failing" quietly becomes "how long is too long," and that second question is not a fact about the mechanism at all. Somebody answers it. And we are about to find that this is not confined to grief.
So the hidden judgment was there in the grief. Now I have to show you it is not only in the grief — that it runs through the whole test, and in particular through the half that was supposed to be the solid one. This is the first of my two challenges to the answer, and I want to be clear before I begin: I am not going to break Wakefield. I cannot, and neither can the people I am about to bring in. He does not break. He turns out blurrier than he looked — which is a different thing, and a more honest one — and you need to see exactly where the blur is.
Go back to the half that was meant to be the fact: the dysfunction. Wakefield's whole achievement was to secure one genuinely factual component — never mind the values for a moment, there is a plain matter of fact about whether a mechanism is doing the job it was naturally selected to do. Fear is for threat. Grief is for loss. Attention is for settling on what matters. A part that fails at its evolved job is failing, full stop, no vote required. That factual component is what let him keep the realist and answer the brother from week three. Remove it, and the whole account reduces to the values and becomes the very thing it was built to escape.
Here is the objection, and it comes from two psychologists, Scott Lilienfeld and Lori Marino, in a 1995 paper aimed straight at that claim. Their question is simple: what, exactly, was any of this selected to do? You say fear is "for" threat — but fear was shaped over a million years for the threats of a world with no supermarkets, no exams, no mortgages, no open-plan offices. So when a man's fear overwhelms him before a performance review, is that a mechanism failing — or a mechanism doing precisely what it was built to do, firing at a threat to his standing in the group, in a setting evolution never encountered? You cannot read the answer off the biology, because the biology was tuned for a world that is gone. And it is worse than that, because evolution does not hand you a clean list of what each part is "for." A trait might be present because it helped our ancestors, or as a side effect of something else that did, or for a reason that mattered then and is noise now. Sadness, anxiety, even some of what we call low mood — there are serious arguments that these were useful, selected for, doing a job, in the world that produced them. So the question "is this mechanism failing at its evolved function?" — the question meant to be the bedrock fact — turns out to require a prior answer nobody has: what was its evolved function in the first place, and against which environment do we measure it? And the moment you have to choose which environment counts, which jobs count as the "real" ones, you are making a judgment. A value has entered the half that was supposed to be free of values.
Lilienfeld and Marino put the point sharply. "Disorder," they argue, is not a clean category with a hard edge and a definition that either applies or does not. It is what they call a Roschian concept — a term from the psychologist Eleanor Rosch — a category organised around clear central examples, with borders that blur and shade off in every direction. Think of the word "game," or "chair," or "heap." You know a dining chair is a chair, and a beanbag more or less is, and a tree stump you once sat on is not quite — but there is no hidden definition drawing the line; there are only better and worse examples, and vagueness at the edges. Disorder, they say, is like that. The brother from week three is the dining chair: the clear case, which no one disputes. But out toward the borders — grief at month eight, fear before the review, boredom that might be despair, inattention that might be a disorder or might be a mismatch between a mind and a room — out there it is all beanbags and tree stumps, and Wakefield's test, which promised to find the line, turns out to be very good at the clear cases that never needed a test and vaguest exactly where you actually wanted help.
Now — keep hold of yourself here, because this is the point where a person gets careless and throws the good thing away. The vagueness does not mean Wakefield was wrong. It does not send you back to the bind. Hear what survives, because most of it does. There are still clear cases, and the clear cases are not a small matter: the test still tells you the brother is genuinely ill and the bored teenager is not, and that sorting spares real people real cruelty in both directions. The two-word structure still holds: a disorder really is some combination of breakdown and harm, and the false choice really is dissolved. What the objection takes from you is only the part the woman's relief oversold — the promise that the first half was a clean fact, a kidney you could read, a line waiting in the biology to be found rather than drawn. It is not clean. It too is a judgment — a careful, disciplined, evidence-soaked judgment, but a judgment, with values running quietly through both halves now, not just the one Wakefield admitted. The middle is still the most serious position on the table so far. It is simply not the solid fact it felt like in the first moment of relief. It is a wise way of drawing a line, mistaken for a moment as a way of finding one.
And if that were the only challenge, you could live there — a thoughtful person who knows the middle is a judgment and makes it as well as it can be made. Plenty of good clinicians live exactly there, and they are not fools. But there is a second voice in the discussion, and it does not test Wakefield from the inside. It walks up to the whole project — Boorse, Hacking, Wakefield, the hidden judgment, all of it — and asks whether we have been wasting our breath.
The second voice belongs to a philosopher named Peter Zachar, whose 2014 book is called A Metaphysics of Psychopathology. What makes him dangerous to everything you have heard tonight is that he agrees there is a hidden judgment — and then declines to be troubled by it.
Consider his move, because it is the most mature-sounding thing in the whole hour, more mature even than Wakefield's. It goes like this. You have spent two weeks, Zachar says, and three or four able thinkers, trying to settle what a disorder really is — whether the line lies in the biology, or in the values, or is braided through both — and look where it got you: a hidden judgment, a fog, beanbags and tree stumps. Perhaps, he says, the reason you cannot find the true definition of disorder is that there is none, and there was never going to be, and the search was a mistake from the start. Psychiatric kinds are not natural kinds. They are not like gold, or like electrons, sitting in the world with a hidden essence, waiting for the right definition to divide them correctly. They are what Zachar calls practical kinds — categories we built, and keep, because they do something for us. A diagnosis earns its place not by being true to some deep fact about the furniture of the universe but by working: does it predict what is coming for this person, does it tell us what is likely to help, does it let a clinic organise care and a researcher run a study and a suffering person get help rather than be turned away? That is the whole job. That is the only job.
And so, Zachar says, the question you have been agonising over — is something really wrong with me, really, is it a real disorder — is not a deep question we have failed to answer. It is the wrong question, a category mistake, like asking whether the line of longitude running through your house is really there. The right question, the only one that pays, is the practical one: does treating this as a disorder help, here, for this particular person, in this actual life? If naming the woman's restlessness ADHD gets her the medication that lets her finish a degree and stop hating herself, then the name is doing its job, and whether it divides nature correctly is a question for people with time to spare. And if naming an ordinary grief a disorder gets a widow pills she does not need and a label she will carry for years, then the name is failing at its job, and again the deep metaphysics never came into it. Stop asking what it is. Ask what it is for, and whether it is earning its keep. You can hear how much easier everything becomes. The hidden judgment stops being a crisis — of course there is no clean fact; there is no clean fact about whether a beanbag is a chair either, and we sit on it anyway when it does the job. This is the working clinician's deflation of the whole two-week argument, and on a Tuesday afternoon with a real patient in the room it is probably the wisest single thing anyone has said tonight.
So that is two answers one could stop at now — Wakefield's middle and Zachar's deflation — and the woman could settle on either and call it a good night. But I told you I test everything I build, and the case against the pragmatist is not that he is wrong. It is that his calm depends on quietly leaving the decision to a question he refuses to ask — and you have met that question before.
Start with the smaller objection. "Whatever helps" cannot, by itself, stop the creep — the very creep you spent last week learning to see. Because if the test for a real disorder is whether calling it one gets people care, then the incentive runs one way only: outward. More categories, milder cases, wider nets — every expansion helps somebody, gets somebody access to care, so usefulness never says stop. What was meant to restrain the spread of diagnosis turns out to encourage it. Zachar's deflation does not check the medicalising of ordinary life; it gives it a clean conscience. Whatever helps — and it all helps somebody.
But the real objection is the second one, and it is a single word hiding inside his question. Does it help — help whom? Because "does treating this as a disorder help" only sounds neutral until you ask who is in the room when the help is judged. Help the patient — or help the clinic bill the hour? Help the child — or help the school that wanted the loud boy quiet, and the exhausted parent who needed the loud boy to be a diagnosis rather than a verdict on the home? Help the worker — or help the company that would rather treat his burnout as a fault in him than look at the sixty-hour weeks that caused it? You felt this exact question last week, pointed at the clinic: who gets to say, and what does the yes permit them to do? The pragmatist promised to make that question disappear by moving from "what is true" to "what works" — and instead he has brought it straight back, because "what works" means nothing until you say for whom, and the moment you say for whom you are back inside the oldest and hardest question this show is built on: whose standpoint, whose interests, whose power, gets to count as the medical judgment and call the result help. He has not escaped the question. He has hidden it, just beneath a word that sounds like kindness.
So look at where you actually are. Two answers one could stop at, both of them real — I mean that; you could stop at either and be wiser than you came. The middle tells you a disorder is a combination of breakdown and harm, and gives you a way to sort that genuinely sorts. The pragmatist tells you to stop chasing the essence and ask only what helps. Neither breaks tonight; both go home standing. But notice what they have in common, because in a moment it will matter more than anything either of them says alone. Wakefield asks where the line of the broken falls. Zachar asks whether treating you as broken helps. Different questions — but listen to them again, side by side. They are both questions about the broken. Both of them — the careful one and the deflating one, the realist's heir and the realist's critic — already agree, before they begin arguing, that the thing to do with your suffering is to work out how broken it is and what to do about the broken. And that is the assumption underneath both. Both of them rest on it. And it is that assumption I am about to disturb.
So what do you leave with. More than you came in with, and — this is the strange part, and I am not going to hide it — less rest than you came in for.
The more, first, because you earned it and it is real. You came in caught between two answers — all real against all invented, with no way to settle. You leave with that bind opened. You have the mature answer: a disorder is a combination — a real breakdown that does real harm by your world's lights — and the work is telling, case by case, which of your sufferings is which. And you have a way to do that work: two questions, in order — is a built-in thing failing, and is the failure costing you a life your world would count as a loss — questions with real answers, that genuinely separate the brother from week three from the bored teenager from the second week of a widow's grief. And if even that feels too heavy, you have the clinician's lighter version: stop chasing what it really is, and ask what helps, here, for this person. Two livable answers, where two weeks ago you had none. Most thoughtful people reach one of these and settle, and they are not fools to. Take it. It is yours, and it is good.
Now the less rest, because I will not give you the relief and quietly keep back the difficulty. Twice tonight an answer felt secure, and you thought: here, at last, is a position I can settle in — once at the middle, once at the deflation. And twice I told you to hold your relief and look closer. The first time: the half of the test that was supposed to be a plain fact, the misfire you could read like a kidney, turned out to have a judgment running through it — how long is grief allowed before it becomes depression, and the answer is set by a committee, not found in the biology. The second time: the pragmatist's whatever helps turned out to mean nothing until you ask help whom, and the moment you ask it you are back inside the question of who has the authority to decide. Neither of those broke the answers. Both did something quieter and worse. They told you that the firmest position in this whole series is still one that somebody drew — a line laid down by people, for reasons — and not the certainty it felt like in the first moment of relief.
But that is not the deepest difficulty tonight. Here it is. Go back and set the two most serious answers side by side, and hear what they never once disagreed about. Wakefield asks: where does the line of the broken fall in you? Zachar asks: does treating you as broken help? They argue about everything else — and neither of them, for one second, questions that the thing to do with your suffering is to work out how broken it is and what to do about the broken. That is not one of the answers. That is the assumption underneath all the answers — under the realist, under the sceptic, under the wise middle, under the deflation, every position you have held for five weeks. They all rest on it. You have been resting on it since the first night, with the phone lit in the dark. And you never once chose it, and you have never once looked at it, because it is very hard to see the assumption you are reasoning from.
Next week, I am going to ask you to turn and look at it.
That is where the sixth episode goes, and I am going to name it plainly, so you can feel it coming, the way you should always be able to. We are done arguing about where the line falls. Next week we ask the question standing behind that one — the question no one in tonight's discussion would raise, because raising it means giving up the whole framework they share: was "how broken am I" ever the right question to ask about a human life — or is it just the only question anyone ever handed you? You met its twin in the very first hour: the second question, the one that stood behind is something wrong with me so quietly you never turned to look at it — am I living well? I told you then that I would not yet say whether that question had gone missing or had only been sitting there, waiting. I am still not going to tell you tonight. But next week the two of them finally meet, and the answer you found this week — the good answer, the mature one, the one you were right to be glad of — stands in the exact spot where that meeting happens.
So take the answer with you. You have it now — a real way to think the question through, where two weeks ago you had only the argument. But take the difficulty with it, because they come as a pair, and I will not let you have one without the other. Even the best this question can give you is a line that careful people drew, and that you can learn to draw a little better — a line, drawn, resting on an assumption you never chose and never once turned to examine. You are not at rest. You are not meant to be. You are one week from the sixth episode.
Thanks for listening. I'll see you next time.