There was a woman I worked with who said, for two years, that the building was making her ill. Headaches. A cough that went round her end of the floor and never quite cleared. I want to tell you what I did about it, because I did not do nothing. I did something worse. I treated her as not worth believing.
I never sat down and decided to. That is the part I keep returning to. At some point I had simply come to treat her as unreliable — as the anxious one, the person who always thinks something is wrong. And I went on trusting the facilities email, the one that said the air had been tested and was fine, because that is the kind of source we trust: the official one, with the letterhead and the measurements. I was not being cruel to her. I want to be honest that it did not feel like anything at all. It felt like simply knowing whom to believe.
Then they found the mould. Behind her wall. Exactly where she had been pointing the whole time.
What stayed with me afterwards was not mainly guilt. It was something more disorienting. I tried to reconstruct how I had worked it out — how I had concluded that she was the unreliable one and the email the reliable one — and I could not find the moment I decided. There was no such moment. The judgment was already in place. She was already discounted and the email already believed, before either of them had said a single word to me about that air.
And here is the part that is really why we are here. If that judgment was at work the whole time, underneath, and I genuinely could not feel it working — then how many people am I doing the same thing to now? Today? The honest answer is that I cannot tell. I have tried. From the inside, discounting someone feels exactly the same as being right about them.
This is Philosophy for Us — philosophy for everyone, no degree required.
This is the first of nine episodes, and the whole series comes down to a single question. Not "what is true." A harder one, underneath it: whom do we believe, and why — and could we tell, from the inside, when the answer is wrong?
Over these nine episodes I am not going to hand you a rule for whom to trust. Each time someone offers such a rule, I am going to show what it costs, and then ask you to keep the question rather than the rule. I am not standing outside any of this. I do it as much as anyone does. I did it to her. So tonight does one thing, and the rest of the series rests on it. It catches that sorting as it happens, describes it plainly, and sets out three answers people reach for to make the problem go away — and shows that none of the three actually works.
Set the woman aside for a moment. She is not really the point. The point is what she let me see — something we are doing constantly, today, to almost everyone around us.
Every day we decide whom to believe, and we almost never experience it as deciding. A doctor tells us one thing and a forum full of people tells us another, and we go with the doctor, and it does not feel like a choice; it feels like ordinary good sense. An official notice says the road is safe and a neighbour who drove it an hour ago says it is flooded, and we weigh the two in half a second without noticing that any weighing took place. Someone we have already marked as prone to exaggerate says they are genuinely in trouble this time, and we quietly treat what they say as less likely to be true, before we have checked a single fact. A man in a suit and a man in a hoodie say the same thing, and it does not reach us in the same way — and we know it.
We do this hundreds of times a day: whose word we take, and whose word we quietly set aside as probably overstated. We are constantly sorting the people around us into the credible and the not-quite-credible, and most of it happens beneath the level where we could catch it even if we tried.
Here is what we would say in our own defence, if stopped in the street and asked. We would say: that is not bias, it is calibration. I believe the people who tend to be right and I doubt the people who tend to be wrong. That is not a prejudice; it is a track record. And that is a serious answer. It may even be the correct one. We are going to take it seriously enough to give it an entire episode of its own, built as strong as I can make it, before we go anywhere near challenging it.
But the next question is the one that matters, the one underneath that answer. That track record — when did we actually compile it? When did we sit down and check, over years, how often the doctor was right, how often the forum was right, how often the official source was right, and how often the neighbour who was actually there was right? We did not. Nobody has the time, and nobody does it. We did not build our sense of who is worth believing by checking. We inherited it — from our schooling, our families, our trades, from whoever counted as serious in the place we grew up. We were handed a sense of who is credible long before we had any evidence of our own, and we have relied on it ever since, so automatically that it feels like using our own eyes.
And a sense of credibility we were handed and have never checked has one feature that ought to unsettle us more than it does. It tells us the expert is level-headed and the forum excitable, the official careful and the agitated stranger unreliable — and it is right very often. That is not the reassuring part. That is precisely what makes it so hard to see. And that is where we go next.
Here is why a sense of credibility we did not form, and have never checked, is so hard to recognise for what it is. It works. Most of the time, it genuinely works.
The expert usually is a safer bet than the forum. The careful official source usually is more reliable than the stranger with a strong feeling. Trust the doctor over the message board, time after time, and mostly you will be right to. So the judgment keeps being confirmed. Every day it delivers a verdict about whom to believe, and every day the world mostly agrees, and after enough of that it stops feeling like a judgment at all. It starts to feel like perception — as though we are simply seeing who is credible, the plain way we see that a wall is white.
This is the part that matters, because the whole series depends on it. A sense of credibility that is right most of the time does not feel like a judgment we are making. It feels like a fact we are noticing. And the reliability is real — which is exactly why it is so easy to miss that any judging is going on at all. A judgment with a long record of being correct stops feeling like a judgment. It feels like seeing straight.
Because here is the other thing a mostly-right judgment does. It makes its own mistakes invisible — not rare, invisible — because from the inside the mistakes look identical to the successes. When I discounted that woman as the anxious one, it felt precisely the same as the thousand times I had correctly judged someone to be overstating their case. There was no signal — nothing that marks the wrong ones out from the right ones. The judgment I applied to her was the very same judgment I apply to genuinely unreliable people, and from the inside I could not tell the two apart. The mould was behind the wall the entire time, and discounting her felt like ordinary good judgment — because most days, discounting somebody like her is ordinary good judgment.
So simply resolving to be more careful, or more sceptical, or more evidence-based does not obviously reach it — because whatever is in that inherited sense is running underneath all of that. It is the thing quietly deciding which evidence we go and look for in the first place, and whose word we take about what we find. The discounting we should trust and the discounting we would be ashamed of can arrive looking exactly alike. That is what happened to me with her. Whether it is happening to us too, and how often, and how we would even tell — that is what we are going to spend nine episodes on.
Notice that I have not told you what follows from this. Perhaps it means the sense is mostly a real track record with the occasional honest miss, and the woman behind the wall is the rare exception. Or perhaps it means the sense is distorted by something that has nothing to do with reliability at all, so closely resembling reliability that we would never catch it from the inside. I have not said which, because I do not get to say which. That is the question the whole series is built on. So let us state it plainly.
So here is the question, plainly, the one the next nine episodes return to. Is our sense of who is credible simply a track record of who is reliable — a real, if rough, calibration to who tends to get things right? Or is it something else underneath: a distribution of belief handed to us by where we happened to stand, by who had the letterhead and who did not, which we mistake for a track record precisely because it is mostly right?
That question is genuinely open. It is not a riddle I have already solved and am slowly leading you towards. Reasonable people land in very different places on it; this is not a settled matter with an answer I am quietly keeping from you. Some hold that the only thing credibility should ever track is the truth, and that asking it to track anything else is a plain category mistake — and they have real arguments. Others hold that who gets believed is settled before the question of truth is even on the table, and is shaped by power, and cannot be read off accuracy at all. And others go further than either, and say that the very idea of a neutral reliability we could appeal to is itself something power built and handed us. I am going to give every one of those its strongest form in the episodes ahead. Not one of them is a straw man. Not one of them is the villain of this story.
Which means there is something I must do now, and you should hold me to it across the whole series. Our culture has three ready-made answers to this question — three ways of making the problem go away — and I am not going to take any of the three, because not one of them actually removes the problem. Let me name them, so you can hear me refuse each one.
The first answer says: trust the marginalised. Believe the discounted. The one who was not heard turned out to be right, so reverse the rule — side with whoever has been waved off. I will not take that answer, because reversing a judgment is not the same as understanding it, and the discounted are sometimes simply wrong. The woman behind the wall was right; the next anxious person down the hall may just be anxious.
The second answer says the opposite: distrust the experts. It is all power, the credentials are a racket, believe no one official. I will not take that one either. It merely rewards the cynic, and the expert really is the better bet most of the time. There is a deeper problem with it as well — but that is for much later.
And the third answer is the quiet, respectable one, the one most of us have reached for ourselves: do your own research. Be your own auditor. Get around the whole problem by checking for yourself. That one we will return to properly, later. For tonight I will only name what it promises — a position outside the whole business, from which we could check our judgments against the way things really are — and leave open, for now, whether any such position exists.
Three answers. I refuse all three. The problem stays with us.
So if I am not handing you a rule, and I have just refused all three ready-made answers, what are we actually doing here for nine episodes?
One thing. We are going to do one thing, slowly and at full weight, and it is the thing this whole series exists to teach. We are going to catch the sense of credibility we are already using — that inherited sense of whose word we take and whose we set aside — and catch it as it operates. We are going to see that it is something we bring to people, not something we read off them. And then we are going to try to step outside it, to grade it, to check it against how things really are — and find out, over these nine episodes, whether that can even be done. Not "here is whom to believe." Something slower and less comfortable: here is the very means by which we have been deciding whom to believe, all along — the thing we never knew we were using, finally set out where we can examine it. What we find when we try to test it is the rest of the series.
If you have been with the show a while, you have done a lighter version of this before — caught yourself standing somewhere you did not know you were standing, and watched it quietly shape what you were looking at. This is the same exercise, applied to the largest thing it can be applied to. Not your taste. Not even your sense of your own mind. Your sense of who, out in the world, is worth believing. It is the most demanding case there is. That is why this is nine episodes and not a single tidy tip.
And I am going to ask you to do one thing now, and then I am going to leave it untouched until we are ready for it. Bring a particular person to mind — the last one you quietly discounted. Not someone you had a row with, not someone you caught in a lie; just someone whose word you treated as slightly less credible, not-quite-reliable. You have one. Everyone does; I do. Do not defend it, and do not feel bad about it. Just hold on to the name. Because a few episodes from now I am going to ask you to take that exact judgment back out, and show — from where you stand, with only what you have — that it was tracking how reliable that person actually is, and not tracking where they happen to sit in relation to you. That is all for now. We will come back to them.
Here is what you have tonight that you did not have an hour ago. Not an answer — closer to the opposite of one. You came in able, you thought, to tell the reliable from the unreliable more or less just by looking. You leave knowing there is an inherited sense doing that telling for you — one you were handed, and never once thought to question. That is the gain: not the answer, but the thing any answer would have to be about. Whether that sense is tracking who is reliable, or tracking something else entirely, is the question you now carry — instead of a verdict you get to pocket.
That is the first of nine.
Next time, I keep my promise about the first answer of the night — the calibration answer. I am going to take the view most of us walked in holding — that our sense of credibility is nothing more loaded than a fair record of who tends to be right — and build it into the strongest case I can put in front of you. Built to win. Because that view deserves its best possible form, standing on its own, before anything else in this series touches it.
But that is next time. Tonight, just the name. The person you discounted. Keep them close. We are going to need them.
Thanks for listening. I will see you next time.