Last week ended with one reply still standing, and I should admit that I sat with it a long time myself before I was willing to put it to the test.
Here is the reply. All the way through this series we have been trying to judge our own sense of who is credible — and each time we reached for a standard to check that judgement against, the standard turned out to be drawn from the same judgement we were trying to check. So last week ended by naming one final way out — the strongest one, the one I had held back. It runs like this: set aside my impressions, set aside what other people say, set aside the whole tangle of who credits whom, and go instead to the plain fact of the matter — to how reliable that person actually is, regardless of anyone's judgement, mine or anyone else's. The flat truth of it. That stands outside all the crediting, and I can check my verdict against it.
So I tried it. I took the woman from the first episode — the one who said the building was making her ill, and was not believed — and I set out to settle it for good. Not whether she was right that one time, which I already know. The deeper question: was she a reliable source, full stop, as a plain fact about her — the kind of fact one could look up.
And I got about one step in before I hit something I could not get past. To find out whether she was reliable, I had to know what reliable even counts as — what would establish it, what kind of thing settles whether a person is a good source. A track record, yes — but measured how, weighed against what, judged by whom, and over which of the things she said? None of that was lying out in the world with a label on it, waiting to be read off. I was carrying it. The standard I would use to check my judgement came from the same place the judgement did.
And that is a different difficulty from last week's, and a worse one. Last week the problem was that I could not trust my own instrument. Tonight the problem is that when I go looking for a better instrument — a neutral one, outside myself, the standard everyone would have to accept — I cannot find the place where it would stand. Every place I reach for turns out to be some particular standpoint, not a neutral one. There is a name for the difficulty I ran into. It belongs to the one thinker this whole series has been moving toward, and he is the most unsettling figure in it. That is tonight.
This is Philosophy for Us — philosophy for everyone, no degree required.
Episode six of nine. Last week we took one of our own quiet judgements — a person filed, weeks ago, as not quite credible — and tried to grade it, fairly or unfairly, and could not, because every tool we would grade it with was the very thing being graded. And I left one reply unexamined: go to the plain truth of how reliable they really are, standing clear of everyone's judgement, and check the verdict against that. Tonight we examine it. I will say plainly that this is the hardest of the nine episodes — the point the whole series has been building toward. By the end, one reply will be gone: the large one, the idea that there is a neutral standpoint you could step out to and grade your own judgement against. But I want to be clear from the start about one thing, because it is what people most often get wrong when this argument circulates online. Removing that reply does not hand victory to the thinker who removes it. Watch for that. It will be the hardest thing in the hour to hold on to.
So let us examine the reply, slowly, because it really is the strongest one and it deserves a fair hearing.
Here is how it goes, stated plainly. We give up trying to clean up our own impressions. We admit that our feelings about a person are coloured. So we go around ourselves entirely. We say: I do not care how reliable she feels to me; I want to know how reliable she is — the real thing, the fact about her that would hold whether I liked her or not, whether I had filed her or not, the fact a fair stranger with no stake would read off the world. Let me get to that, and I will have my standard at last — a standard that does not come out of my own judgement, because it is out there, in the facts, where my judgement cannot reach.
It is the right instinct. It is the instinct behind every honest thing we do when we do not trust ourselves: we look for the objective record. So let us try to pick it up. What is "how reliable she really is" actually made of?
We would say: her track record — how often she has been right. Good — except that a track record is not a thing that exists on its own. Somebody has to decide what goes into it. Which of her claims count — only the dramatic ones, the building is making me ill, or also the hundred quiet, ordinary, true things she said every day that nobody recorded, because nobody records those? Somebody has to decide what counts as her being right — right on the headline, or right on the details, or right for the right reasons. Somebody has to decide how to weigh a near miss, a hunch that paid off late, a thing she got loudly wrong against a thing she got quietly right. The track record does not assemble itself. Somebody assembles it, by a rule about what counts. And the rule is not in the world. It is held by whoever is keeping score.
Very well, we say — then take it out of my hands entirely. Use the experts. Use what qualified people say about who is reliable. And now the next plain question: qualified by whom? Who decided what makes a person qualified to judge reliability — what training counts, what credential, what method, whose say-so about another person one is allowed to lean on? Because that, too, did not fall out of the sky. Somebody established it. There was a long history in which it got settled what a real expert looks like, what a real study looks like, what counts as evidence and what gets waved off as anecdote, whose testimony one is supposed to take seriously and whose one may file as excitable. None of us derived any of that from first principles. We inherited it, fully built, the same way we inherited our sense of whom to believe.
And here is the thing I could not get past. Every single time I reached for the neutral standard outside myself — the track record, the qualified judges, the documented facts, the plain objective measure of who is reliable — I found that the measure itself had been made somewhere. It came already scored. What counts as reliable, what counts as evidence of reliability, who counts as fit to certify reliability — all of it is the product of some history, some institution, some long settling of accounts about whose word travels and whose does not. There is no standard floating free of all that, no measure that came from nowhere. The measure is always somebody's, made somewhere. And that particular somewhere is exactly what we were trying to get outside of.
That is the difficulty. And there is a man whose entire life's work was this difficulty — who said, in effect, of course you cannot find the neutral standard; the search was confused from the start, because the very idea of what counts as knowledge is something that gets produced, and produced by exactly the kind of power you were hoping to step outside of. His name is Michel Foucault. He is where we go next, and I am going to give him to you at full strength, because he is the hardest thing in this series, and if I soften him I cheat you.
Foucault's idea — the one that matters for us tonight — is usually put in two words joined together: power and knowledge. Michel Foucault was a French historian and philosopher; the argument runs through his work of the nineteen-seventies, in books like Discipline and Punish, from 1975, and the interviews collected as Power/Knowledge, in 1980. The ordinary way to hear those two words is the cynic's way: the powerful decide what counts as true, and they lie to keep the rest of us down. That is not it. That is much too small, and if that were all he were saying he would be easy to answer. He is saying something stranger and harder.
He is saying that power and knowledge are not two separate things, one pushing the other around. He is saying they are inseparable — that every society runs what he called a régime of truth. Not a régime that decides which statements are true; a régime that decides, beneath that, the whole apparatus of how you tell — which procedures count as finding out, which people count as qualified to speak, what gets to stand as evidence and what gets dismissed as noise, where the line falls between the serious and the unserious, the sane and the mad, the knowing and the merely opining. The régime sets the terms. And only then, within those terms, do particular claims get sorted into true and false.
Consider what that does to "follow the evidence." We had always taken the evidence as the way out — the neutral ground to stand on when people disagree, the thing that does not take sides. Foucault's claim is that what counts as evidence is itself one of the things the régime settled. It is not the exit from the system of who gets believed; it is one of that system's own products. When we say I just follow the evidence, we are not stepping outside the system to check it. We are making a move the system already defined. The rules that tell us a peer-reviewed study outranks a neighbour's account, that a credential outranks a lifetime of watching, that the calm institutional voice outranks the frightened one down the corridor — those rules feel like neutral good sense. Foucault spent his life showing that they have a history. They were made — in hospitals, in courts, in schools, in prisons. He dug up the actual records of how it got settled what counts as a real diagnosis, a real confession, a real proof, a real expert. None of it came from nowhere. All of it came from somewhere, and power was at work in that somewhere the whole time — not power lying about the truth, but power building the very framework within which truth would later be judged.
So when we went looking, last week and tonight, for the neutral standard — the place outside everyone's judgement where reliability is simply a plain fact — Foucault's answer is the coldest in this series. There is no such place. Not because nobody has found it yet, but because the idea of it is an illusion. Every standard is built. Every outside we run to turns out, on arrival, to be another inside — another particular standpoint within the same system, scored by the same kind of power that scored the judgement we were trying to escape. There is no view from nowhere — no vantage point standing clear of every framework. There never was one. We have been standing at some particular standpoint the whole time, and calling it nowhere.
Now I want to show where that feeling came from — the feeling that there was a nowhere to stand — because it has a history too, and seeing the history is half of getting free of it.
There is an old and admirable picture of what it is to know something for oneself: the lone thinker who clears his mind of everything he was ever told, every inherited opinion, and rebuilds his beliefs from the ground up on evidence anyone could check. René Descartes, in his Meditations of 1641, sitting alone, doubting everything until he reaches something he cannot doubt. Later — another country, another century — Immanuel Kant gives the picture its motto, in his essay of 1784 on the question "What is Enlightenment?": sapere aude, dare to know. Think for yourself. Do not take it on authority. Stand on your own reason, which is the one thing that belongs to no church and no king — the one ground that really is neutral, because it is just reason, the same in everyone.
That picture is the deepest root of the thing we walked in here believing. For on that picture, credibility is something a properly independent mind grants only after it has checked for itself, and leaning on another's word is a regrettable shortcut, a falling-short of the ideal. And you can hear the everyday, present-day version of it: do your own research. That is the lone thinker, two hundred years on, reduced to a search bar. The same idea exactly: I will stand on neutral ground — my own independent look at the facts — and I will not be at the mercy of anyone's authority. I will be the one who stands nowhere, and therefore sees straight.
And here is what the whole history was for. Foucault's point — the same point we reached the hard way with the woman from the first episode — is that the lone thinker was never standing nowhere. He was standing in seventeenth-century France, in a particular language, with a particular training in what a good reason looks like, inside a framework that had already decided what counts as clearing your mind and what counts as bedrock. The independence is the disguise. Someone certain that he stands nowhere is exactly the person who cannot see where he stands — because the whole feeling of standing nowhere is what stops him examining his own position. The promise of neutral ground is the very thing that hides the ground he is actually on.
We did a version of this once before, in an earlier series, with a machine. We tried to name the test we would use to deny a machine a mind, and the test turned back on the people we love. This is the same thing, carried all the way up and run on knowledge itself. We tried to name the neutral test we would use to grade who is reliable — and the test turned out to be one more standard of our own, occupying a particular standpoint, the whole time. There is no neutral standpoint outside every framework. That is the loss of the outside. I told you it would be the hardest hour.
Now. I promised to be honest about who closed that reply off, and whether they had the right to. So before we let Foucault walk off with the whole evening — and he is seductive; that cold completeness is seductive — I want to turn his own objection back on him, because there is an obvious and fair place to put it.
Here is the question to ask him. Foucault, you have just told me there is no neutral truth — only régimes, only what some arrangement of power lets count as knowing. Very well. Now: that thing you just told me — is it true? Is "there is no neutral truth" itself a neutral truth, a plain fact about all knowledge everywhere, the kind that holds whether anyone likes it or not? Because if it is, you have just done the very thing you said could not be done. You stepped outside all the régimes and reported back on how things really are. You found the neutral standpoint — and stood on it — to tell me there isn't one. And if that is what you have done, the claim contradicts itself: it is true only if it is false.
No, we say — Foucault is too clever for that; he would never claim a neutral fact. So take the other branch. Then his claim is not a neutral truth. It is just another move inside a régime — his régime, the seminar room, the particular circle that finds his method persuasive. But then why does it bind me at all? If "it is all power" is itself just power talking, then it has exactly the standing of the thing it was attacking, and not an inch more. The calibration thinker says I track reliability; Foucault says no, you track power — and if Foucault's own saying is only ever power talking, then the two of them are simply two voices in the room, and he has given me no reason to exchange mine for his. He has not defeated the plain calibration story. He has only joined it, on the same level, as one more voice with no special claim.
That is the objection Foucault aimed at everyone else, now turned on his own claim. "It is all power," if true, has no more right to your belief than anything else, because it has stripped itself of the very standing it was trying to strip from everyone else. The most all-conquering version of Foucault is the one that defeats its own author too.
So is he finished? No — and here I have to be careful, because the cheap move is to think you have killed him with one clean question and walk off satisfied, and you have not. Foucault has a reply, and it is serious. He would say: you have caught me only if I was claiming to hand you the truth from nowhere — and I never was. What I am doing is excavation. I am showing you the actual history of how a particular régime came to feel like plain neutral sense, so that the thing now presenting itself as the only way to know loses its air of being the only way. I do not need a neutral standpoint to do that. I need only the documents — the old records, the moment the line got drawn — and a refusal to take any régime's word that it is simply the natural order of things. I am not the new authority on truth. I am the one bringing the knowledges that were pushed aside back into consideration.
And that reply has real force. It means we cannot simply wave him away as a man who refuted himself. The honest position is in the middle. The strongest Foucault — power all the way down, nothing ever any better than anything else, the régime is the whole of it — does undercut its own standing, and you should not let anyone sell you that version, however bracing it sounds online. But the quieter Foucault — there is no neutral standard; every outside is some particular standpoint, built somewhere; the thing that calls itself plain reason has a history and a location — that one is still standing, and it survived the objection. And notice why it survived: we did not actually need Foucault to reach it. We got there ourselves, back at the start of the hour, simply by asking what a track record is made of. He did not smuggle it in. He named a difficulty we had already walked into on our own.
So the calibration thinker's position has been undercut, and the genealogist cannot cleanly escape the objection to his own either, and you might think that is the end of it — both positions defeated, nothing left to choose between, take your pick. But there is a third position that has been waiting for exactly this moment, and it is the one I most want you to hear, because it is the one the loud versions of this argument skip straight past.
It comes, in our own day, from philosophers such as Helen Longino — in her book Science as Social Knowledge, from 1990 — and Elizabeth Anderson, who has written on the epistemology of democracy and on epistemic justice in institutions. And it begins by agreeing with the one thing that survived the objection. Of course there is no view from nowhere, they say; we have known that for a century, and it was never where objectivity lived. The lone thinker on neutral ground was always a fantasy, and noticing that it is a fantasy is not a catastrophe — it is simply growing up. The mistake, they would say, is yours and Foucault's both: you both assumed that if there is no single neutral mind seeing from nowhere, then nothing is left but power. And that is a leap. Look at what you skipped over.
Here is what they put in its place. Objectivity, they argue, was never one mind standing nowhere. It is something a group achieves, when it is built well. Not nowhere — many particular standpoints, set against one another. You take many people standing in different places, with different stakes and different blind spots, and you arrange things so that they have to answer to each other — so that when one of them is wrong, someone standing elsewhere can catch it and has a real channel to say so and be heard. That is what a good court is trying to be. A good laboratory. A genuinely free press. Not a view from nowhere, but an arrangement in which no single view gets the last word and dissent has actual force. And the claim is that something built that way produces better knowledge than any lone judge ever could — not perfect, not neutral, but improvable, and demonstrably more often right. So "there is no neutral standard" does not deliver "every measure is as good as every other." A rigged scale and a scale that many independent people get to check are not the same scale, even though neither one came from nowhere. You cannot step outside power. But you may be able to build a better standpoint — and the right response to a judgement warped by power is not to throw out the whole idea of getting it right; it is to repair the procedure.
And that is a real answer. It is the live, serious thing a thoughtful person says at this point, and it is neither Foucault nor the cynic, and you should not let either of them shout it down.
But — and the reformer knows this; the honest ones say it themselves — it has a cost. Look at what it rests on. It rests on there being such a procedure in the first place: one that is not already captured. A free press that is not already owned; a court whose idea of a credible witness is not already the régime's idea; a community of inquiry whose real uptake for dissent has not been quietly tuned so that only the already-acceptable dissent gets heard. The reformer is betting that one can build a standpoint that is genuinely better, and not merely better at concealing that it is a standpoint. And that bet is precisely the thing Foucault spent his life saying you cannot safely make — because every such procedure he examined turned out to have the power already inside it, in the very structure, in what it had decided counted as a real objection before anyone walked in. She cannot prove him wrong. There is no neutral ground to prove it from — she, of all people, grants that. So she does not escape the choice between them. She stands at one end of it and makes a bet, in good faith, with her eyes open. It is a serious position, and it is still a bet.
Let me state plainly where this leaves the argument, because it is stranger and more even than "Foucault won."
One reply is closed — the large one. The neutral standpoint, the standard of reliability standing clear of every judgement, is gone; and here is the part that matters: everyone at the table agrees it is gone. Foucault closed it. The reformer never wanted it. And the calibration thinker, once honest, cannot find it either — we watched him fail to, last week and tonight. Closing that reply crowns no one, because it is the one thing all three of them accept. It is the shared ground, not one side's victory.
What they divide on is whether there is another way through — not a neutral standpoint, but a built one, good enough to grade a judgement by. And that question is wide open. Foucault leans hard toward no: it is régimes all the way, and your better procedure is just a régime with better manners. The reformer bets yes: you can repair the procedures, and getting it right is still the aim. And I have not shown you which of them is correct. I cannot. Nobody in this argument can hand you that from outside it, because there is no outside — that is the one thing we did settle. So nobody leaves here crowned. The genealogist cannot, because his strongest claim undercut itself. The calibrator and the reformer cannot, because their answer is a bet on the very thing in question. All three are still standing, and not one of them is standing on secure ground.
So bring them back — the person you named, the one whose word you quietly turned down, the one you could not grade last week. Last week you tried to grade them from the inside, and the instrument was the very thing on trial. Tonight you have the strongest reply in hand, so use it. Go outside yourself. Say: I will find out how reliable they really are — the documented fact, the public record, the verdict of the people qualified to judge — and I will check my verdict against that.
And watch where your hand actually goes. To find their real track record, you have to decide which of the things they said get counted, and how a hit weighs against a miss — and you are deciding that with the same sense of what matters that did the filing in the first place. To find the qualified judges, you have to decide who counts as qualified to judge them — and you are sorting those people into the serious and the unserious with the very judgement you came to check. To find the documented facts, you have to decide which documents are sound and which sources credible — more testimony, sorted by the same hand. Every route you take to the outside lets you out into another position within the same system. There is no way out at the back. You already knew the inside was unreliable. Tonight you find that the outside you were going to correct it with is just more of the inside, scored by the same kind of power, as far down as you can reach.
That is the step, at the highest point of the series. Not you have a judgement — you had that last week. It is this: you reached for the one thing that was supposed to stand outside your judgement and grade it, and your hand closed on more of the same judgement. The outside was a place you had never left.
So here is what you leave with tonight. You do not get it is all power — that version undercut its own standing in front of you, and you should distrust anyone selling it, however clear-eyed it sounds. You do not get just calibrate carefully and you are fine, either — not cleanly, not any more; that is a bet now, a real one, on there being a procedure that is not already rigged, and you have seen the size of the bet. What is actually gone is the outside — the neutral place to stand. And the loss is quiet, but it does not grow back: the next time you go to settle whether someone's word is good by checking the facts — calmly, sensibly, the grown-up move — you will feel it. You will know that the facts you reach for were scored somewhere, by someone, in a history you did not write, and that you are standing inside that scoring while you do the checking. There is no position marked you are not here. You are always somewhere. And from somewhere, a judgement shaped by power and a judgement shaped by accuracy can look exactly alike — they present the same picture, with the same confidence.
I have not told you that power wins. I have taken away the place from which you could have proved it does not.
Next week we stop standing back from the whole structure and look instead at how it works from the inside. Because there is a fair question you should be holding right now, almost angrily: all right, there is no clean outside — but my reasoning still feels like reasoning; it does not feel like power pushing me around; so where exactly does the power get in? By what mechanism does it reach a thought that felt, the whole time, like me simply working things out for myself? That is next week — not whether power is in our judgement, we are past that, but the actual machinery of how it gets there.
Thanks for listening. I'll see you next time.