You did the exercise — the harder version this time, the part that did not feel like power.
You went looking for the list: what it would take to make your word — the one you designed, the one you would defend — actually carry your better meaning out in the world. Not in your own head. Out there, in the mouths of people who use it and never asked you. You started writing, and the first step or two were things you could do: say it your way, say it often, say it where it counts. Then the list changed. Get the people around you to say it your way. Get the people who write the rules to say it. Get a generation of them to grow up saying it. Step by step the list stopped being a list of things you could do and became a list of things that would have to happen.
And there, with the pen still on the page, you looked for the connection you had assumed was there — the one that was supposed to take everything you had decided and carry it out into the language. You could not find what your deciding was joined to. You could feel yourself deciding, clearly and well. You could not tell whether anything at the other end was moving.
Your decision on one side; whether the word actually moves on the other. Tonight is about the distance between them.
This is Philosophy for Us — philosophy for everyone, no degree required.
This is the fifth of eight. Last week we found the second half of the move, the half you may never have noticed. Deciding what a word should mean is real, and it is yours; but it is only half the work. The other half is the word coming to mean it, out in a whole living language. And the news last week was good, and it was true: meanings do move. Communities change their words in plain sight, all the time, by a public process — people using a word a certain way until the use spreads or dies out. So the two halves can meet. They meet constantly. And we ended on a question that sounded almost friendly: what would it actually take to make a word mean what you have decided it should?
Tonight we answer it — or rather, we find that the answer depends completely on a prior question you have never once asked. Not can I redefine this word. Not even will my use win. The question underneath both: what fixes a word's meaning in the first place? The very same redefinition — your word, your case, unchanged — is either the easiest thing in the world or needs half the world to shift, and which one it is depends entirely on where a word's meaning actually lives. You have been assuming it lives somewhere you can reach. Tonight we check. And I will say it now, so it does not arrive as a shock: on the account that has dominated the last fifty years of this argument, it does not.
So keep the question in front of you for a few more minutes. By the end of tonight you may no longer be sure that deciding a meaning was ever joined to changing it. That is not me taking something from you. It is the thing we have to look at plainly, together, before either of us pretends to know whether you can do what you have spent years assuming you could.
Start with a strange fact about your list.
You wrote down what it would take to move your word. And the honest truth is you do not know whether the list is one line long or enormous. You do not know whether changing what "merit" means is the sort of thing you could manage in one good afternoon of clear argument, or the sort of thing that takes a war, a generation, and a lot of luck. Here is the strange part: it is not that the answer is hard to find. It is that the same redefinition — your word, your case, not one comma changed — could be either one. Trivially easy, or needing the whole world to shift. And nothing about your design decides which.
So something else decides it — something you have not yet looked at.
Herman Cappelen, a philosopher who has thought about this as carefully as anyone, names exactly this. In his 2018 book Fixing Language, on redesigning our words on purpose, he argues that the question of whether you can change a meaning has been approached from the wrong end. People argue about the redefinition itself: is it good, is it justified, is it worth doing. All of that, he says, comes second. The thing that actually settles whether you can move a word is a question almost no one asks first: what fixes the meaning of a word in the first place? Where does a word's meaning actually live? Until you know that, you do not know what you would have to change in order to change it. You are trying to move something without knowing where it is.
So let us ask his question. When the word "merit" means what it means — where is that meaning? What holds it in place?
Here is the answer you have been quietly assuming, and you have assumed it since the first night of this series. You have assumed the meaning lives somewhere near you: in what you intend by the word, in the definition you would give if asked, in the concept in your head — and in everyone else's head, one by one, so that to change the meaning you change minds, yours first, then theirs, one conversation at a time. Meaning as a thing made of intentions. If that is where meaning lives, then your list is short in principle: change enough minds and the word moves, because the word was only ever the sum of the minds.
Now here is the account that has spent the last fifty years making that first one hard to hold. It comes from two entirely separate directions — two philosophers, two arguments, built for different reasons — and the striking thing is where they both end up.
The first is Hilary Putnam, in his 1975 paper "The Meaning of 'Meaning'", and his argument concerns a word you would have called simple: "water." Picture a planet exactly like Earth — call it Twin Earth — identical in every detail but one: the liquid in its lakes and taps is not H2O but some other compound that looks, behaves, and tastes just like water, with a different chemistry. Now go back to 1750, before anyone on either planet knew any chemistry at all. A man on Earth and his exact double on Twin Earth each say "water." Same word. Same picture in the head — the planets are identical, so the two men are in identical states of mind. And here is Putnam's point: they mean different things. The Earth man's "water" refers to H2O; the twin's refers to the other compound. Same mind, different meaning. So the meaning was never in the head. It was fixed by what the word was attached to out in the world — the real substance, the actual chemistry, which neither man knew and neither man chose. Putnam's phrase for it is blunt and well known: meanings just ain't in the head.
The second direction is Saul Kripke, in his 1980 book Naming and Necessity, and it is a different argument — not about substances but about names, about how a word reaches the thing it names at all. On Kripke's account a word gets attached to something at a founding moment — a first naming — and then travels from person to person, each using it because they picked it up from the last. You call gold "gold" not because you have the definition right, not because you could tell it from a convincing fake, but because you stand at the end of a chain of speakers that runs back to people who were pointing at the actual metal. The word's hold on the gold is kept by that chain — by the whole community across time — not by what is in your head when you say it. You could be wrong about every property of gold and still mean gold, because the chain means it for you.
Two arguments. One about the world, one about the chain of speakers. Built by different people who did not set out to prove the same thing. And notice where they both leave you: the meaning of your word is not sitting in your head waiting to be revised. It is out there — attached to a world you did not build, carried by a community of speakers you did not appoint, most of whom have never heard your case and never will.
There is a third argument bearing on the same point, older, and reaching the same place for its own reasons — Wittgenstein, whom we have met before. His thought is not about chemistry or chains. It is that the meaning of a word simply is how a whole community uses it — the full range of times and ways it gets said, with no single essence underneath holding it together. Meaning as a public practice, not a private possession. That is a separate argument again, and it lands in the same place as the other two: a word's meaning lives in the shared life of millions of speakers. Not in any one of them. Not in you.
I am keeping these three apart on purpose, because it is tempting to run them together into one idea — "meaning is out there somewhere" — and they are not one idea. They are three different arguments that happen to converge. But converge they do, and the convergence is what matters tonight, because of what it does to your list.
If your assumption was right — meaning lives in heads, change the heads and you change the word — then your list is a list of people to persuade. Long, perhaps, but the kind of thing a person does. It is on your side of the line.
But if these three arguments are right — if the meaning of "merit" is held in place by the world it is attached to, the chain of speakers who carry it, and the shared practice of everyone who uses it — then your list is not a list of people to persuade. It is a list of conditions that would have to hold across an entire language. And you are one speaker. You begin to see the difficulty. We have not even reached the hard part yet. We have only found the question you skipped: not is my redefinition good, but where does this word's meaning actually live — and the answer with fifty years of argument behind it puts it almost entirely out of your reach.
In a moment we ask what follows from it — and what follows is the hard part of tonight.
So here is what Cappelen builds on that ground, and it is the centre of tonight.
Grant the account from a moment ago: meaning lives out there — in the world the word is attached to, in the chain of speakers, in the shared practice of millions. Cappelen grants it. He thinks this externalist picture is roughly right and hard to dislodge. Then he asks the only question left: if that is where meaning lives, what does it actually take to change one? He gives two answers, and each weakens the assumption you have been holding.
The first he calls inscrutability. It means the process that changes a word's meaning is too tangled for us to understand. Not hard, not merely unsolved — too tangled: too large, too many moving parts, too sensitive to things no one tracks. Think how "cool" came to mean good, or how "gay" came to mean what it now means. Could anyone set out the mechanism? Name the steps, in order, that did it? Say which push mattered and which did not? You can tell a story afterwards — there is always a story afterwards — but you cannot run it forwards. No one can stand at the front of one of these changes and say: do this, then this, and the meaning will move. The process that changes meanings is real, and it works, and it is, Cappelen says, beyond our power to survey. There is no manual to read — only a few hundred million people using a word, pushing on it from every direction at once, and a meaning that shifts as a result by no route anyone can trace.
The second answer is harder, and it is the one that does the damage. He calls it lack of control. Set aside whether we understand the process; ask whether we can steer it. His answer is: for the most part, no. Not because you are not clever enough or not committed enough, but because of where the meaning lives. It lives in the use of an entire community, and you are one speaker in that community. You can use your word your way, loudly, for years. You can be entirely right. And whether the meaning actually moves is settled by everyone else's use, by the world the word is attached to, by the chain — by the whole vast process you are a single voice inside. You do not control it. No one does — the process is not the kind of thing a single person can steer.
Now see what that does to the assumption.
You have carried it since the first night: that deciding what a word should mean and changing what it means are connected — that the one drives the other. That if you design well enough, argue clearly enough, hold to it long enough, the meaning moves; that your deciding and the word's changing are joined by something you could not see but were sure was there. That is the assumption. And Cappelen has just argued, carefully and twice over, that the join may not exist — that you have been deciding, and waiting for the word to follow, and your deciding was never attached to the word at all. It was attached only to your own conviction. You could go on deciding for ever.
So do the thing this whole series has been walking towards. Take your word — the real one, the one you designed, the one you would defend. Ask the question with Cappelen's two answers in hand. Did I change what that word means? Not should it change. Not is my case good. Did the meaning, out there, in the mouths of the people you argued with — did it move? And be honest about what you actually did. You decided. You decided well, perhaps. You worked out exactly what the word should mean and why, and you would defend it to anyone. Then you said it, and waited for the world's use to fall into line with your decision, as you have always quietly assumed it would. I decided the meaning. Now finish the sentence honestly: I decided the meaning — and I do not know that I changed the word. I may never have touched it.
That is the assumption going. Not the small satisfaction of learning a fact — the loss of something you had always assumed and never checked.
And see what it does to the thing that troubled you at the very start. The fear, from the first night: that when you redefine a word to win an argument, you are cheating — changing the terms mid-argument so your side comes out ahead. Look at that fear now. You can only cheat by changing something you actually have the power to change. If Cappelen is right, you were never changing the shared word — you could not. You were doing something smaller and stranger than cheating: you were asking. Making a case — as, on the first night, I could only guess you were — to a process that was always going to decide for itself, in its own time, by means indifferent to the quality of your argument. The thing you feared might be a trick turns out, on this account, to have been a request. You were not changing the standard to suit yourself; you were only saying what you thought the standard should be, with no power to set it.
That is the heaviest thing tonight. But an objection is rising in you, and you should trust it, because something here is too strong, and you already know what it is. You are thinking: wait — words do change. "Marriage" changed. I watched it change. "Gay" changed. If no one can move a meaning, who moved those? Hold on to that. It is not a feeling; it is an argument, and a good one, and it is where we go next.
You are right to resist, and I am not going to talk you out of it, because the resistance is correct — it is what keeps tonight from turning into a cheaper story than the true one.
Here is the cheaper story, so you can see it and refuse it. It says: the connection is broken, so give up — meaning cannot be steered, redefining is a fantasy, go home. That story is wrong, and you already have the evidence in the words you use every day, because meanings move. Not in theory — in living memory, in front of you. "Marriage" means something in this decade it did not mean two decades ago. "Gay" travelled a long way within one lifetime. We even changed an astronomical word on purpose: a committee of astronomers voted, Pluto stopped being a planet, and now schoolchildren learn eight. Words move. Some of them were pushed, by people who set out to push them, and the pushing worked.
So set that directly against Cappelen, because it is a real objection and he has to answer it. If no one can steer a meaning — if the process is too tangled to grasp and too large to push — then how did those words move? Who moved them? You cannot say in one breath that meaning-change is beyond anyone's control and in the next that campaigners changed what "marriage" covers and astronomers changed what "planet" covers. One of those has to give. Either those words were not really steered — they drifted, and the people who think they pushed are flattering themselves — or steering a meaning is possible after all, and the lack-of-control claim was overstated. Cappelen owes an account of the cases where steering plainly worked. That debt is real, and it is exactly where we go next week, so I am not settling it tonight; I am only making sure you see that it is owed.
There is a second objection, harder than the first, and I will name it now even though we do not open it properly until next week — because without it you would leave here having quietly granted Cappelen a victory he has not won.
Watch the step he makes: meaning lives in the practice of a whole community, therefore you cannot engineer it. There is a camp of serious philosophers — we come to them next week — who grant Cappelen every word of tonight and then stop at that "therefore." What they deny is not his picture of where meaning lives; it is what he thinks follows from it. Next week they get their full say.
I am not going to settle that tonight, and I want to be exact about why, because it would be easy to settle it the wrong way and call the result depth. It would be easy to end on "the connection is dead, the project is over" — that has a grave, serious sound, and it would be false. And it would be just as easy to end on "do not worry, we push together, it works out in the end" — that is a comfort, and it would also be false. The truth is harder, and it is what I am actually leaving you with: you cannot assume that deciding a meaning changes it, and you cannot assume that it does not. Both are now closed to you. You came in able to assume, without noticing you were assuming, that deciding a meaning was most of the way to changing it. You cannot assume that any more — Cappelen has taken it. But you also cannot swing to despair, because the very words you use have moved, and serious people hold that the moving can be done on purpose. You are standing between two things you are no longer allowed to take for granted. That is not the episode falling short; it is the real shape of the question becoming visible. You had been treating as settled ground something that was never settled at all.
So look at what you are holding now, because you are not leaving empty-handed — you are leaving with more than you came with. You arrived with one assumption fixed so deep you had never seen it: that to decide a meaning is more or less to change it. You leave with that assumption taken apart, and a real question in the space where it was — where does this word's meaning actually live, and is it anywhere I can reach? That question is yours now. You can run it on any word, and it is sharper than anything you walked in with. This is the fifth of eight, and it is the first to take something back: the first four episodes handed you the power, and tonight is the first to return some of it — plainly, with its reasons shown, not to leave you smaller, but to stop you mistaking a request for an accomplished fact.
Next week we go straight at the account Cappelen owes: the words that did move, and the people who say they moved them. We meet the group that grants the whole externalist picture and still holds that you can engineer a meaning — never alone, never by decree. The question will not be whether meaning can change; it changes, we have settled that. It will be the one underneath: whether the change can ever be brought about on purpose — and if it can, who does it, and how.
For now, do this with your word — the one you have carried all five weeks. Run tonight's question on it: where does its meaning live? In the heads you could change one conversation at a time, or out in a practice you are one small voice inside? Do not reach for the answer that settles you; you do not have it. You decided what your word should mean. Whether you can change what it means, you do not know. An hour ago you would have said you did. That is what tonight has taken from you, and it does not go away.
Thank you for listening. I will see you next time.