I did my own research

Consider someone at your own table — a member of your family, an old friend, someone you love — who believes something that is false, and not harmlessly false. They have refused a treatment on the strength of it, or they are voting on it, or they are repeating it to your children. So you challenge it. And they answer, quite calmly, with the sentence that in most families ends the discussion: I did my own research.

You want to reply that their research amounted to a few internet forums and a man selling supplements on a podcast. But something holds you back. They are not lying. They are not inventing it to win the argument. They sincerely believe it. They did not sit down one morning and decide to believe it; it simply seems true to them, in the same way that the things you believe seem true to you. And a doubt occurs to you that takes most of the force out of your anger: a person cannot really be blamed for what they honestly believe. They are wrong. But they are not guilty.

This is Philosophy for Us — philosophy for everyone, no degree required.

This is the first of eight episodes on a single question — one of the most current questions there is, and one of the oldest beneath it: are you responsible for what you believe? We are going to take the argument you have at that table — who is to blame in an age of misinformation — and work down to the much older question underneath it.

Because consider what happens on the way home. The argument turns around and points back at you. You hold beliefs too — about your health, your politics, a former partner, your own competence — beliefs you would stake a friendship on, beliefs you would say you had looked into. And if you are honest about how you looked into them, some of the time you sought out the people who already agreed with you and stopped the moment the matter felt settled. You called that research. So did they. And if they are innocent because they sincerely believe what they believe, then so are you — always, about everything, however little work you did to get there. You arrived at that table certain that you were the careful one and they were the fool, and by the time you were home you could no longer locate the difference between you.

I am not going to supply that difference tonight. I am no exception to any of it; I have done my own research badly and called it being thorough. What I can tell you is what this series will not do. It will not tell you simply to trust the experts. And it will not tell you to go and do your own research. Both are ways of avoiding the question, and over these eight episodes we will work out why neither is as reliable as it appears. Tonight we only open the question. And find that it will not close.

Begin with why the defence works — because it does work, and that is the whole difficulty. If "I did my own research" were plainly foolish, there would be an easy reply at the table, and there is not one, and it is worth being honest about why there is not.

Take the sentence apart. When you accuse the person across the table, what exactly is the accusation? Not that they are lying. They are plainly not lying. Lying would almost be the easier case: a liar knows the truth and conceals it; a liar has done something, made a choice, selected the words. One can be angry at a liar without confusion. But this person is concealing nothing. They are telling you, as plainly as they can, what seems true to them. So the charge cannot be dishonesty. The charge, if there is one, has to be the believing itself — that they are guilty of believing something false. And said aloud, that has an odd sound to it.

Here is why it sounds odd. Consider the things we ordinarily blame people for. We blame them for what they do. He struck the man; she took the money; he drove drunk. Blame attaches to actions — to things that were within a person's control, that they could have done or not done, where two courses were open and they took the worse one. That is the picture of responsibility most of us have carried all our lives, and it is a good one; it fits almost everything. Belief does not fit it. Believing is not something a person does. It is something that happens to them.

You can test the point directly. Try to believe, simply by deciding to, that a third arm is growing from the middle of your chest. Not to pretend it, not to say the words — actually to believe it, as you believe you have two. You cannot. You cannot even begin. You could be offered a large sum to believe it and the belief would not come, because belief does not answer to a decision. It answers to how things seem, and it seems overwhelmingly that you have the usual number of arms, and that impression is not within your control. You did not choose it. It simply presented itself, the way a perception presents itself, and you believe it whether you wish to or not.

So here is the defence, assembled in full. Blame is for what a person chooses. No one chooses a belief. Therefore no one can be blamed for a belief. The person at the table no more chose to find that forum convincing than you chose to find your own eyes convincing. It simply seemed true to them. And one cannot put a person on trial for how the world looked from inside their own head. Lying is a fault, because a liar chose; believing is not a fault, because a believer did not. "I really thought it was true," said in that way, is not an excuse at all. It is a report of how things seemed — and we do not punish people for how things seem to them.

I want to be careful here, in the way I will ask us to be careful throughout the series, which means taking this defence as seriously as the person leaning on it before laying a finger on it. The temptation, when someone says "I did my own research" and is wrong and the error is doing harm, is to conclude that they must be stupid or dishonest, and to skip past the fact that their defence is a genuinely good one. It is a good one. There really is a difference between the person who lies to you and the person who is sincerely mistaken; you feel that difference, and you forgive the second in a way you do not forgive the first, and you are right to. There really is something outside a person's control in what seems true to them. A jury that convicted a man for the way daylight looks to him would be a mad jury.

And the point reaches further than the obvious cases, which is what gives it force. The sincere believer is not only the person you disagree with. You are the sincere believer too, constantly, about the things you are most sure of. Every belief you would defend to the end feels, from the inside, exactly as that forum felt to them — not like a choice you are defending but like a fact you are reporting. You do not experience your own politics as a decision you could revise by an act of will, any more than they experience theirs that way. From the inside, each of us is simply reporting how things look. So if sincerity clears them, it must clear you, because there is no inner difference between the two. One honest believer and another are the same; they merely happen to be looking at different things.

Look a little longer at how strong this is, because I am about to start testing it, and I do not want anyone thinking I chose something flimsy to test. The defence is not a dodge. It is built from two things that are simply true. First, one cannot believe at will — the third arm proved it. Second, we blame people for what they could have done otherwise, and a thing one cannot help is, by definition, not a thing one could have done otherwise. Put those two true things together and the conclusion follows, cleanly: no one is to blame for a belief, ever. Not the relative at the table. Not the man on the forum. Not you. Each of us registers how things seem, and how things seem is no one's fault.

That is the argument the whole series is built to examine. I have now put it as strongly as I can, deliberately, because an argument this strong is exactly the kind worth examining closely. When something feels this watertight — when the reasoning is this clean and the conclusion this convenient — that is precisely the moment to grow suspicious. Not because the argument is obviously wrong; it is not obviously wrong. But because it does something a little too useful. It clears everyone of everything, permanently, the instant they meant well. And most of us have lived long enough to distrust any argument that hands out that much innocence for nothing.

Keep the suspicion; do not try to resolve it yet. It cannot be resolved with tools we have not built, and reaching for the answer now is exactly the impatience this series will ask us to give up. The defence looks watertight, and something in us does not trust it, and both of those are true at once. We stay there for the rest of the episode.

Because here is the first flaw — not in the logic; the logic is sound — but in what we actually believe when we stop arguing and look at an ordinary life. The argument says no one is ever to blame for a sincere belief. Very well. Consider, then, a sincere belief that killed someone.

Consider a man driving home at night. The road looks clear to him. It genuinely does — he is not drunk, he is not texting; he glances, sees nothing, and drives on. And there was a child. He did not see the child because he did not really look: a half-second glance at a bend in the road, where the bend was exactly the place a careful driver slows and looks twice. He honestly thought the road was clear. The sentence is true. He is not lying when he says it. The road did seem clear to him.

Now apply the defence to him. He did not choose to believe the road was clear; the belief arrived, as beliefs do, out of a glance and the look of things. He could not have believed at will that a child was there, any more than you could summon the belief in the third arm. So by the clean logic we have just built, he is innocent. He meant well. He sincerely believed the road was clear. He registered how things seemed, and how things seem is no one's fault.

And no one believes that for a moment.

No one does. We have a word for that man, and the word is not unlucky. The word is negligent. We say he should have looked harder. We say the belief that the road was clear was a belief he had no right to, given how little looking he had done to earn it — and that the harm traces back not to the bad luck of a child being there, but to the looking he skipped. He did not choose the belief. We blame him even so. And we are not confused when we do it. A jury would not release him because his belief was sincere; a jury would ask how hard he looked.

So something has gone wrong. A moment ago we proved, cleanly, that a person cannot be blamed for a belief they could not help. And here is a belief no one could help — the road seemed clear, and on the evidence he allowed himself he could not have believed otherwise — and we blame him regardless, and we are confident we are right to. Both of these are things we believe. They cannot both be the whole truth. Either the clean argument has a hole in it, or the man goes free. And we will not let the man go free.

Now bring it closer, because the driver is too easy — he is someone else. Bring it to your own case.

Think of a sincere belief of your own that cost something. Not a belief you lied about — a belief you held, honestly, the whole way through, and were wrong about, and the error landed on someone. Perhaps you were sure about a person, and said the sure thing to others, and it was not true, and it followed them. Perhaps you were sure about a treatment, a school, a decision made for someone who trusted your judgement, and you had looked into it in the way you look into things, and you were wrong, and they paid part of the price. Most of us have at least one such case. The real one — not a comfortable hypothetical, but the one with an actual face at the other end of it.

And ask honestly: does "but I sincerely believed it" close the case? When you hold the real one in mind, does it feel like a complete defence — as though you are cleared the moment you establish that you were not lying? Or is there a part of you, looking at that particular episode, that says: I should have looked harder. That sincerity is not the end of the question. That somewhere back there was perhaps something you could have done and did not — a source you did not check, a doubt you brushed aside, a person who tried to tell you and whom you decided not to hear — and that the honesty of the belief does not reach back and repair the part where you did not look.

If that is present when you hold the real one — and for me, with mine, it is — then notice what has just happened. You have done to yourself exactly what you could not do to the person at the table. You have found a line. Not the line between the sincere person and the liar — you and your relative are both sincere, both honest, both reporting how things seem. This line runs elsewhere, somewhere beneath the sincerity, in a place you were not looking and cannot quite name. And once you admit the line exists in your own case, you cannot pretend it is absent in theirs. You arrived furious that they would not take responsibility, and the thing you were demanding of them is a thing you can barely face in yourself.

So we are caught between two things we both believe and cannot hold together. First: a person cannot be blamed for what they sincerely and helplessly believe — and that is not nothing; it survived the test; the third arm is real; one genuinely cannot believe at will. Second: we do blame the driver who did not look, and we blame ourselves in the case we have just called to mind, and the blame does not feel like an error — it feels like the most honest judgement in the room. Sincerity is a real defence. And sincerity is plainly not the whole defence, because if it were, the driver goes free and one's own worst mistake is excused, and we believe neither of those.

I am not going to resolve that here. Let me be plain about it, because the pull to jump to the answer — to say, well, obviously the difference is such-and-such, and settle it — is very strong just now. Resist it. Settle it now and you will settle it with the first thing that lets you feel innocent again, which is the very move we are here to study. The honest position tonight is to hold both halves and watch them fail to fit. One cannot be blamed for the belief. One is somehow answerable even so. Both are ours. The line between you and the person at the table did not hold — and the line that does hold is not where you expected it.

There is a word that sits directly over the division we have just uncovered. It is the word the whole argument is conducted in, the one your relative reached for and the one you reached for, both of you, without noticing that you meant different things by it. Let us look at it directly.

Research. That is the word. Your relative says they did their own research. You say you researched it. The man on the forum researched it. The scientist whose paper he half-read researched it. One word — and by the time it has passed through all those hands it is carrying at least two entirely different things, and no one at the table notices, because the word stays the same while the thing beneath it is exchanged.

Here is the first thing the word can mean. Research is the most careful thing a person can do. It is what you do when you distrust your first impression — when you go looking for what might prove you wrong, when you seek out the people who would disagree and make yourself understand why, when you check the source behind the source, when you notice that you want a particular answer and lean against the wanting on purpose, because you know it is there. That kind of research is the opposite of negligence. It is precisely the harder looking the driver did not do. When someone has genuinely done it, "I did my own research" is not an alibi. It is a credential. It means: I have earned this belief; I did the looking.

Here is the second thing the word can mean. Research is what you do when you already have the answer and go looking for support. You put the question in the form that returns the result you are hoping for. You read the three things that agree with you and feel the satisfaction of being confirmed, and the satisfaction is so agreeable that you stop — there, at the first agreement, you stop, and you call the stopping "being satisfied that I had looked into it." You never sought out the strongest person on the other side. You never went looking for the thing that would ruin your answer. You conducted a search, and the search was shaped, throughout, by the answer you meant to come home with. And then you stand up and say, in complete sincerity, I did my own research. And you did. That is the difficult part. By the loose sense of the word, you did exactly that.

Two different things, then, in the same five words. And from the inside — this is what makes it hard — the two feel identical. The person who searched for confirmation and the person who genuinely inquired both end the evening sincerely sure; both did something they would call looking into it; both can say the sentence without lying. The sincerity is the same. The sense of having done research is the same. What differs is buried back in how the looking went, and that does not show up in the sincerity at all. You cannot tell, from the feeling of being sure, which of the two you did. The feeling is identical. It has to be — producing that feeling is the whole point of the second kind.

So here is what I want to set out plainly tonight, so that it can be held and checked as we go, because we are going to spend eight episodes testing it. Each of us arrives carrying two assumptions we have never separated. The first: that sincerity is innocence — that if I really thought it true and was not lying, I am in the clear. The second, running beneath it, quieter and somehow even more trusted: that "I did my own research" is the proof that I was careful — my diligence, the part that shows I did not believe lazily, my defence. We reach for the second to shore up the first: I am not to blame, because I sincerely believed it — and I sincerely believed it because I looked into it. That is the whole structure — one claim resting on the other.

I am not going to knock either of them down tonight. I am going to do something I think more honest and more uncomfortable, which is simply to mark them — to say: there are the two assumptions you have never once examined. You have never asked whether sincerity actually clears you or merely feels as though it should. And you have never asked whether your research was the genuine article or only the appearance of it — the harder looking, or the moment of confirmation you stopped at and dressed up as diligence. Both will be examined. Not now. But you should at least know what they are.

Let me also say, plainly and briefly, what this series is not about, because there are two large questions that sit immediately next to ours and are not the one we are asking, and unless I mark them off we will keep falling into them.

We are not asking whom you should trust. That is a real and difficult question — how to tell a good source from a bad one, how anyone learns what they cannot check for themselves, how to trust well in a world where one has to trust almost everything. It is a whole inquiry, and it is not this one. We will lean on it and point back to it, but a guide to identifying reliable sources is a different subject.

And we are not asking who has the power to be believed — whose word is taken seriously and whose is waved away, how that is settled by things that have nothing to do with the truth, who gets silenced. That matters enormously, and we will touch its edge. But it concerns the structure around the believer — how credibility is distributed. Ours is the question that remains when all of that is set aside and a person is alone with their own case: was I answerable, and for what? Am I answerable for how I came to believe the things I believe — and if I am, for what, exactly?

That is the question. Its most current form is being shouted from every feed you own — who is the fool, who is to blame, do your own research, no, trust the experts. We are going to walk straight past both slogans, because both are ways of avoiding the question, and find the older one they are both running from. Tonight we have only located the question. Let me set out what there is to take from it.

So here is what there is to take from tonight — less than you came wanting, and more than you had.

You came in with a defence. The defence was: I sincerely believed it, and behind that, I did my own research. You took it to be solid, and you meant to use it as your relative used it, to end the argument. The defence has a flaw in it. Not because sincerity is worthless — it is not; it survived; one really cannot believe at will, and one really cannot be tried for how things seem. The flaw is that sincerity does not reach as far as you needed it to. The driver was sincere, and we blame him still. You were sincere in the worst thing you got wrong, and you cannot quite forgive yourself, and you may not be able to release yourself as quickly as the sincerity would like. So sincerity was always going to let something through. The question is what.

And you have the word now as well, divided down the middle. "I did my own research," said one way, is the proudest thing a careful person can say. Said the way it is usually said at the table — yours and theirs alike — it may name the moment the looking stopped, when the wanting was satisfied. You cannot tell the two apart from the inside. The feeling is the same feeling. That is what there is to take: not an answer, but a flaw you can no longer overlook, in the one sentence you most relied on to feel innocent.

That is the gain, and it is a real one. You came in able to say "I sincerely believed it" and have it mean the matter is closed. It is harder to say it now without hearing the question underneath — the one you cannot yet answer. That is worth an hour.

This is the first of eight, and it is the easy one. It only asks us to notice that the defence has failed, not yet to put anything in its place. From here it grows harder, and it grows harder in a particular direction.

Next time, the hardest form of the case against you. A Victorian named William Clifford will tell you about a shipowner — a man who sent an old and doubtful ship to sea, full of emigrant families, having allowed himself to feel sure that she was sound. The ship went down. Everyone aboard drowned. And Clifford will make an argument you will not easily get past: that the shipowner was guilty — not when the ship sank, but at the moment he let himself believe she was safe on evidence he had not earned. That he would have been just as guilty had the ship, by sheer luck, come home safely, because the wrong was in the believing, not in the drowning. Clifford holds that every careless belief is exactly this — a small version of the shipowner — that you are doing it constantly, and that it is a wrong against everyone, not a private mistake. He will be hard to answer. Hear him at full strength, because if he is right, the person at your table is not innocent at all. And neither are you.

But that is next time. Tonight the only thing I will ask you to take away is a question, and I am not going to answer it, because you are not ready to and neither am I.

Choose one. One belief you would stake something real on — your health, your politics, the thing you are sure of about a person, the thing you are sure of about yourself. One you would tell me, with a straight face, that you had looked into. Do not go and re-argue it in your head until it feels safe again; that may only be the confirmation again, the search for support, and you would simply be doing it once more. Hold the belief still and ask the single question: when I say I looked into this, which kind of looking was it? The kind that went hunting for what might prove me wrong? Or the kind that found the people who already agreed and stopped at the first agreement?

You do not have to answer tonight; you could not yet. And the answer you would reach for now is the one that lets you off, which is the one I trust least. Only carry the question to your own table. You came to that table sure that you were the careful one. See whether the word holds when you turn it on yourself.

Thanks for listening. I'll see you next time.