After last week, I set out to do the very thing I keep asking you to do. Late at night, with a pen and a blank page, I wrote four words at the top: what would settle it. What would I actually have to see — evidence for, or evidence against — to close the question of whether there is anyone there. No more going round the question without settling it. I meant to be honest, to finish the sentence, and to have my standard written down at last.
And I wrote something. I put down: I'll believe it when there is evidence that would convince any fair-minded person — something public, something anyone could check, something that could not simply be explained away. That seemed right. Solid. The kind of sentence you could say at a dinner table and watch people nod.
Then, because it was late and I was not attending to myself, I read back what I had written — public, checkable, would convince any fair-minded person — and I thought about my wife, asleep in the next room, and about my belief that there is someone there. Not a body breathing under a blanket. A person, with an inner life I have never once seen and never will: a life I cannot photograph, cannot set on a table, cannot ask a fair-minded stranger to check. I cannot produce a single piece of public, repeatable evidence that there is anyone home in the person I have built my whole life around. And it has never once occurred to me to withhold belief until that evidence arrives.
So there were now two sentences on the page. The standard I had just written down for the question of God. And the way I actually live, everywhere else, all day, without apology. They did not match. They were not close. The rule I was about to judge the largest question there is by, I do not apply to the most important person in my life; I never have; and I had never noticed.
This is Philosophy for Us — philosophy for everyone, no degree required. This is Is There Anyone There?, and it is episode four of eight. Last time we built the strongest case against: the fawn that burns to death in a forest fire where no one will ever know, and the silence — the honest seeker who asks and asks and receives nothing back. Two arguments at full strength. Then the one reply available to the believer: the gap between our minds and the mind in question — you cannot move from "I see no reason" to "there is no reason." And then the awkward part: the same gap that rescued the believer from the fawn also removed the case for design. Nothing was settled. And one thing was left unresolved — the assessment you have been trying to make since the first week, the private weighing-up that would not come out even, however fairly you conducted it.
Tonight I stop putting arguments in front of you. For three weeks I have run what looked like a survey — here is the case for, here is the case against, weigh them. Tonight I tell you what that actually was. And then I turn it around and ask you the plainer question underneath all that weighing, the one you skipped: what, exactly, would settle this — for you? And we will find out why that weighing-up was never going to come out even.
Let me tell you what the last three weeks actually were, because it was not what it looked like.
It looked like a survey. In the second week I put the strongest case for in front of you — the fine-tuning of the universe, the conditions set precisely for life — and I asked you, in so many words, to weigh it. In the third week, the strongest case against — the fawn, the silence — and again, to weigh it. The case for, the case against, the arguments presented one after another, and you in your seat doing the natural thing, the thing I had set you up to do: judging which side each argument favoured. A point for one side, a point back for the other. That is what a survey is. Here are the arguments; here is how they fare; you decide. It is exactly what a series like this one is expected to be, it is what you were prepared for from the first night, and I let you think that was what you were getting.
It was not. Here is what I was actually attending to, the whole time, while you thought you were grading the arguments. You were judging them — yes. But to judge them, you need a criterion. You need to know what counts as a point: what makes a piece of evidence tell for a conclusion, or against it; how much weight the fawn carries; how much the fine-tuning carries; what would have to appear for the whole question to tip one way. You had such a criterion. You were applying it to every argument I handed you. You had to be — you cannot weigh a case without one. And I doubt you have ever set that criterion out and examined it. You assessed three weeks of arguments against a standard you never stated, never chose deliberately, and could not, right now, write down.
That is the real point of these three weeks. Not the arguments. The arguments were genuine, and I built them as well as I could. The point is that a survey makes you concentrate so hard on what is being judged — the question of God, the fine-tuning, the fawn — that you never turn your attention to the standard you are judging by, or where it came from, or who set it.
So tonight we stop surveying. I am not going to put another argument in front of you. I am going to turn the attention around.
And the first thing to notice, once you do, is that the question you walked in with was settled in advance — not by me, but by its shape. You came in assuming there were two ways this could go. Either the arguments succeed, and the honest response is to believe. Or the arguments fail, and the honest response is not to. Two outcomes; choose the one the evidence points to. That feels like the whole of it — the only choice available.
We did this exact thing in How to Be Good, with right and wrong, so you have met it before: something that looks like a single line with two ends turns out to conceal a third question underneath. Look at your two outcomes. Arguments succeed — believe. Arguments fail — do not believe. Now ask what both of them rest on. They both assume there is a settled answer to what counts as the arguments succeeding — a criterion for how much evidence is enough. The believer who says "the evidence is there" and the unbeliever who says "the evidence is not there" are both reading off the same criterion; they disagree only about whether it has been met. Neither is asking who set that criterion, or on what authority, or what it actually measures. The real question was never which outcome. It was: who fixed the standard, and by what right.
So here is where I turn it on you — and I mean now, wherever you are sitting, not as something to consider later. Name the standard. Not mine; I showed you mine in the cold open, and it did not survive its second sentence. Yours. Say it aloud. Finish this sentence honestly: I would believe there is someone there if — and then the actual condition. Or, if you already believe, the other way: I would give the belief up if — and then what. Do not reach for the clever answer. Reach for the true one. What would it take? What would actually settle it, for you?
Take a moment and genuinely try to finish it.
I think one of two things has happened. Either the sentence would not finish — you reached for the condition and found nothing definite there, the same blank I reached years ago. Or, and this is the harder case, it did finish — you produced a clean sentence, something about evidence, something about proof, something a fair-minded person could not deny — and it felt solid, and you were a little pleased with it. Hold on to whichever one you got, particularly if it is the clean one. Because we are about to do to your sentence exactly what that late night did to mine. We are going to take the standard you have just named, and instead of applying it to God, apply it to everything else you believe, and see what happens.
Take the clean version first, the one that felt solid: I'll believe it when there is evidence — public, checkable, the kind that would convince any fair-minded person. Good. A strong standard. Now do not apply it to God. Apply it to your own mother.
You believe your mother has an inner life — that there is something it is like to be her, that behind her face there is a person who feels the cold, dreads the dentist, and privately thinks you do not call enough. You are as certain of that as of almost anything. Now produce the evidence. Public, checkable, the kind that would convince any fair-minded person. You cannot. You can point to her behaviour — she winces, she laughs, she says "ouch" — but a determined sceptic can reply that this is only a body doing what bodies do, and ask how you know there is anyone there behind it rather than a working organism with no one inside. You cannot observe another person's inner life directly. You never have. You cannot prove, to the standard you have just set down, that there is a single other mind in the universe besides your own. And you believe in billions of them, without apology, and you call yourself rational for it.
We have done this exact thing before, and you did it, so let me say where. Mind and Machine. There the question was whether a machine could really be conscious, and you reached for a test — some behaviour, some sign, that would prove there was someone inside. And then we turned it around, and you saw the difficulty: the test you would use to rule the machine out also ruled out everyone except yourself. You had been applying a standard for "is anyone there" to every mind you had ever met, you had never once stated it, and the moment you did, it excluded the entire world apart from your own head. The same thing is happening tonight, at a more general level. There it was a test for another mind. Here it is the standard for any belief at all. And it does the same work — it excludes almost everything you actually live by.
Because it is not only other minds. You believe the world is more than five minutes old. You cannot prove it. Every memory you would reach for as evidence — your childhood, this morning, the beginning of this sentence — could have been switched on five minutes ago, complete, and you would notice nothing; there is no experiment that gets you clear of it, because the experiment would be one more memory. You believe the floor will hold when you stand, that the past is a guide to the future — and the only argument for trusting the past in that way is that it has worked before, which is using the past to vouch for the past, and proves nothing to a sceptic. You believe there is a real world outside you and not a dream you are thoroughly committed to. None of it can be proven. All of it is held, rationally, every waking minute, by you — the same fair-minded person who was about to require evidence of God, and of God alone.
There is a sharper version of this that you also met, in The Trust Recession. There we found that almost nothing you know did you actually verify; you took it on trust — the doctor, the map, the date of a war you were not alive for. You had checked almost none of it. Tonight goes one step further. It is not only that you never verified most of what you know. It is that the rule itself — belief should wait for evidence — the rule you were about to apply to God and to nothing else, you have never applied to the things you most depend on, and you have never examined the rule itself. You took the rule on trust as well.
Now — and I have to put this carefully, because it is the part most often misread in the moment — none of this shows that belief in God is fine. I am not, by unsettling your standard, quietly letting God back in. It is entirely open to you to say: very well, but God really is different. Other minds, the past, the floor holding — I have to assume those simply to get through the day; they are the conditions of having any experience at all. God is not like that. God is an addition — a specific, enormous claim about what exists, and a claim of that kind should have to show its support in a way my mother's inner life does not. That may be exactly right. I am not telling you it is wrong. I am telling you it is a real position, and you are entitled to hold it.
But notice what you have just done if you reached for it. You have admitted that the standard is not automatic. You have conceded that whether the demand for evidence applies depends on what kind of belief is in question — and that means you owe a reason. You would have to say why belief in God goes in the "show me the evidence" category while your mother's inner life goes in the "of course, no evidence required" category. You would have to draw that line deliberately and defend where you drew it. And you never have — not from laziness or dishonesty, but because you never noticed there was a line to draw. You were applying the demand to the one belief and quietly exempting all the others, and calling the result simply being reasonable, and you never saw your own hand doing the sorting.
So that is the first flaw in the standard: you do not apply it to anything else you believe, and you have never said why God is the exception. Here is the second, and it is quieter and stranger, and I want to go carefully, because it can sound like a trick when it is not.
Take the rule itself. Whatever you wrote down — believe only when there is sufficient evidence. Now put to the rule the question the rule puts to everything else: where is your evidence? What is the proof that one should believe only on sufficient evidence? Is that claim public, checkable, the kind of thing a fair-minded person could not deny? Can you run an experiment on it? Point to it somewhere in the world? It is not written in the sky. You cannot put "always proportion belief to the evidence" under a microscope, or derive it from a measurement. It is a rule about how to handle beliefs — and by its own demand it would require evidence before you were permitted to hold it, and that evidence is precisely what it cannot supply. I am not going to pursue this tonight; I only want you to notice that it is there. The rule cannot meet its own demand. The one standard you were treating as the ground under everything else is the one standard that, held to its own terms, cannot produce the evidence it requires.
I will leave that where it is. It is both lighter and heavier than it looks, and we return to it. The first flaw is the one to carry with you.
Because put the two together and consider where you now stand. In the first episode I described the frame of mind you walked in with as a kind of trial: the question of God being tried, the arguments led in one at a time, and you presiding — weighing them, ready to deliver a verdict. It felt like justice. It felt like the most fair-minded thing a person could do: hear both sides and decide on the evidence. That was the picture, and it was the thing you were proud of.
Now consider the rules of evidence that trial ran on — the rule that says what may be admitted, what counts as a point, how much is enough. You have spent four weeks applying those rules: admitting the fawn, refusing "you cannot prove love either" as out of order, keeping the running assessment. And tonight, when you went to examine them, you found two things. You found they were written down nowhere; you had never actually stated them. And you found they were not even rules you live by; you set them aside, without difficulty, for your mother's inner life, and the age of the world, and the floor beneath your feet. So whose rules are they? Where did this trial get its law?
It got it from you. You set it. That is the whole of the difficulty, and it has been sitting under the entire series. You were never the neutral judge, hearing the case fairly and deciding on the merits. You wrote the rules of evidence, you appointed yourself to apply them, and then you forgot you had done either, and called the whole arrangement simply thinking clearly. You are not the judge in this case. You are one of the parties to it — one side, who also happens to have drafted the law the case is tried under, and then took the judge's seat as though the law had been there already.
And that, finally, is why the assessment would not come out even. Recall the strain last week, the weighing-up that would not balance however honestly you set the fawn against the fine-tuning. You supposed the trouble lay in the arguments — that they were simply matched, a genuine tie. It did not. The weighing-up would not resolve because the criterion for conducting it was never fixed in the first place. You cannot arrive at a total when you, the one totalling, also quietly decide what counts as a figure — and never wrote it down, and never checked it, and did not know you had done it. The assessment was not straining because the contest was close. It was straining because you were measuring with a standard you had set yourself, without examining it, and had forgotten you were using.
So here is where you actually are tonight, and it is not where you expected to be when you sat down. You came to weigh the question of God, and found you must first examine the standard you would weigh it with. You came as the judge, and find you are a party. You believed you were applying a rule of evidence you had found already in place. You wrote it, and had never once read it back.
Before you go, here is the thing to take and use — because you do not own this by watching me do it to you; you own it when you have done it to someone yourself, with no one arranging it for you. So here is the exercise, and it is not about God. Deliberately. Do it on neutral ground first, so you can see the shape of the thing on its own.
You know someone — most people do — who is proud of being evidence-based. The friend who says, I just follow the data; I don't believe anything I can't back up; show me the evidence and I'll change my mind, and that is the difference between me and the people who believe what is comfortable. It is an admirable sentence. They mean it as the mark of an honest mind, and for the most part it is one.
Same ground rules as the first week: do not argue, do not raise God, do not try to catch them out — that only starts a quarrel and ruins what you are watching for. Ask them the prior question — what would actually count, though? State the standard. Then stay with it. Ask whether they applied that standard to the belief that their husband actually loves them, or whether they simply know it, the way one knows things one would never think to check. Ask whether they demanded data before trusting their oldest friend with something they had told no one else. Ask whether the belief that their own life is worth getting up for arrived on evidence, or was underneath all the looking to begin with. And then, most lightly of all — and this is the one that matters most — ask what the evidence is for "believe only what the evidence supports," and whether that rule arrived in the manner it says everything must.
You are not trying to win, and you are certainly not trying to show them they are wrong — they may not be; their standard may survive the examination. You are trying to watch one specific thing happen: the moment a person takes up the standard they have applied to the whole world, examines it for the first time, and realises they have relied on it for years and never examined it. That is the whole of it. When you can watch that happen in another person, and stay kind while it does, the method is yours.
So let me say plainly what you now have, walking out, because it is not an answer, and it is worth being honest about its size. You came in able to say "the evidence is not there, so no," or "you do not need evidence, so yes," and to feel, either way, that you were simply reading off what is the case from a fair and neutral position. You are leaving unable to locate that position. You went to weigh the question of God and found you must first examine the standard you would weigh it with; you went to read the rules of the court and found them in your own writing. That is smaller than a verdict, and it changes more than a verdict would. You cannot un-see it. The neutral position you thought you occupied is not there, and you will feel its absence every time you reach for "well, where is the evidence" from now on.
Now — there is one last thing you will want to say — I know, because it is the one I reached for too, and it is the only reply still standing. You want to say: proportioning what I believe to the evidence is not some position I picked up and could put down. It is not one option among others. It simply is what being reasonable means. You have not shown my standard is wrong — only that it is mine; and so what, it is mine because it is the only sane one there is. Hold that. Say it to yourself in exactly those words, and do not let me talk you out of it tonight, because it is the strongest reply you have, and it deserves its own hour. That is the last reply left, and it is where we go next week. We will walk straight up to it — the claim that wanting evidence is not a standpoint at all, but simply what being reasonable means — and test whether it holds, as we have tested everything else. I will not tell you tonight whether it does.
Let me say what I have not done tonight, so that no one leaves with the wrong conclusion. I have not proved there is anyone there. I have not lowered the bar for faith by an inch — and if you are a believer and feeling vindicated right now, thinking, you see, you cannot prove your other beliefs either, so mine is fine, you have taken the wrong lesson; next week your side has a reply of its own to answer for. And I have not shown that the unbeliever is irrational for wanting evidence. All I have done is take the one thing you were sure was neutral — the standard itself — and show you that it was yours, and that you had never read it. The believer who cannot say why God should meet a bar his own mother's inner life need not meet, and the unbeliever who could examine that standard and keep it — knowingly, deliberately, with open eyes — are both still standing at the end of tonight. I have knocked neither one down. I have only taken away the position that let you suppose you were above the dispute rather than inside it.
So — is there anyone there? I still cannot tell you, and tonight I did not try. Tonight I did something stranger, and you felt it. I took the question you walked in with — is there enough evidence? — and showed you that the word enough was doing all the work, and that the meaning of enough was something you had settled yourself, without examining it, and had been applying to the largest question there is as though you had found it fixed and given. The question is still open. The case for God, the case against, the fawn and the fine-tuning and the silence — all of it still stands. What is different now is that you have turned around and seen that the neutral position was never under you, and that the standard this whole question is judged by is one you set yourself. Next week we ask the last thing: whether, in setting it, you were only ever writing down what reason already required — or whether you have been the author all along.
Thanks for listening. I'll see you next time.