Last week ended with a problem left open. You had decided what your word should mean — carefully, perhaps well — and you could not honestly say you had changed what it means out in the world, in the mouths of the people you were arguing with. The deciding was real. Whether the deciding changed anything, you could no longer assume.
Then, sometime in the week, you used a word.
Perhaps you said "gay," and meant by it what everyone around you means by it now — and you knew, underneath, that the word did not mean this when you were a child. It changed, within your own lifetime, and you had watched it change. Or you said "marriage," and you know — not as a theory, but as something you lived through — that the word covers people this decade it did not cover two decades ago. It changed. And that observation makes last week's conclusion strange. If no one can deliberately change a meaning — if the process lies beyond anyone's control — then who changed that one? You did not imagine it. The word is different now. Someone, or something, changed it, within a single lifetime, in plain view.
So there are two claims here, and both seem true, and they appear to contradict each other. Last week said you cannot change what a word means. Your own memory says words change. One of them is wrong, or something is missing that would make both true at once. Holding both at once is the work of tonight.
This is Philosophy for Us — philosophy for everyone, no degree required.
This is the sixth of eight episodes. Last week was the demanding one. We went looking for where a word's meaning actually lives, and found it living almost entirely outside you — anchored to a world you did not build, carried by a chain of speakers you did not appoint, held in the shared practice of millions who never heard your case. The philosopher Herman Cappelen drew the consequence: the process that changes meanings is too tangled for anyone to map, and largely beyond any one person's control. So you left unable to assume that deciding a word's meaning was close to changing it — and also unable to conclude that it cannot be done at all. Both of those, I said, were now closed to you.
Tonight we take up the second half. Last week I said Cappelen owed us an account, and that we would come and ask for it. The account is this: the words that did change. If deliberately changing a meaning is beyond anyone's control, he has to explain "marriage." He has to explain "gay." He has to explain a committee of astronomers who sat down on purpose and changed what counts as a planet. Those changed. Some of them were changed on purpose. So tonight we bring him the cases — and then we meet the people who say something sharper than "words drift." They say you can engineer a meaning, deliberately, on purpose — and they say it while granting Cappelen every word about the world-anchoring, the chain of speakers, and the shared practice. They do not deny where meaning lives. They deny what he thinks follows from it.
By the end you still will not be able to say that your own decision changes a word. But you will find that changing a word was never a matter of your decision alone. It depends on a much larger process, made up of many people — and whether those people can be brought to change a word on purpose is the question we open tonight and do not close.
So let us ask Cappelen to account for the words that changed.
Start with the cleanest case, because it is almost comic how clean it is. In 2006, a room full of astronomers in Prague took a vote. They had a problem: they kept finding more icy bodies out past Neptune, some about the size of Pluto, and the word "planet" had never been defined to say where to stop. So they wrote a definition — a thing has to orbit the sun, be round, and clear its own path through space — and they counted hands, and Pluto failed the third test, and Pluto stopped being a planet. On purpose. By a show of hands. They set out to change what a word covered, and they changed it, and the textbooks now count eight. If you want a case where people deliberately changed a meaning, you could not ask for a tidier one. They did not drift into it. They voted.
Now the messier cases, which matter more, because they are closer to your word. "Marriage" covers people now it did not cover when I was young. That did not happen by a committee — there was no room in Prague, no show of hands that settled it. It happened through courts and campaigns and many arguments at many kitchen tables, and a generation that simply grew up using the word differently from their grandparents. And "gay": go back far enough and it meant cheerful; then for a long stretch it carried a sting, often an insult; and now, in most rooms, it is a plain word people claim for themselves. No one chaired that. There was no vote. But it changed enormously, and much of the changing was done deliberately, by people who set out to do it, against people who set out to stop them.
So here is the question for Cappelen. Last week you said the process that changes meanings is beyond our control. Then who changed these? You cannot say in one breath that meaning-change is largely beyond anyone's steering, and in the next look at "marriage" and "planet" and "gay" and say those simply happened, with no one acting. People acted. Sometimes the effort failed and sometimes it took, but the effort was real, and it was aimed, and the words are different now because of it. Last week I said I would not answer this for him, and I will not — I am putting it to him, as the same straight choice I set out then: either those words were not really steered, and the people who think they pushed only think so — or "beyond our control" was too strong. One of those has to give.
And there is a whole way of thinking about words that has stood in the background of this series, and it is time to let it speak. It does not have a single famous name attached to it the way Cappelen's argument does; it runs through a group of philosophers who treat concepts a particular way. Call it the pragmatist picture, and here is its one core thought: a concept is a tool. That is the whole of it. We made our words for jobs — to sort things, to praise and blame, to draw lines where we need them — and a tool is the kind of thing you re-make when the job changes or the tool stops working. "Planet" was a tool for sorting the lights in the sky, and when we found too many lights it stopped doing the job, so we re-made it. There is nothing mysterious here, the pragmatist says, and nothing shameful. You revise a concept the way you would sharpen a blade that had gone dull. Of course "marriage" changed — the job we needed it for changed. Of course "fish" once included whales and then did not — we learned more about the job. Words shift because the work shifts. That is not a scandal. It is what tools do in the hands of the people still using them.
Notice what that picture does to the two fears this series has been built on. Take the first, from the very first episode: that redefining a word to win an argument is cheating. The pragmatist is unmoved. Cheating? You re-made a tool. A carpenter who files a chisel to a new edge for a new cut is not cheating; he is working. Concepts get re-made constantly, in the open, by everyone, and calling that cheating makes sense only if you thought the dictionary arrived finished. It did not. We built it, we have always been editing it, and your editing it too is just joining work already under way. So the first fear loses its force. Now take the second, from last week: that changing a meaning is impossible, beyond anyone's reach. The pragmatist answers that one too, and in the same way, by pointing to the evidence. Impossible? We do it all the time. We did it to "planet" in a single afternoon. Meanings are not locked away; they are handled every day.
Now the question of how. If meanings move and a community re-makes them, the obvious question is the mechanism: re-made how, by whom, in what setting? The pragmatist has a concrete answer, and it is worth slowing down on, because you met a lighter version of it two weeks ago. Much of the re-making happens through ordinary argument — through disputes that look exactly like disputes about facts but are not. Two philosophers, David Plunkett and Tim Sundell, gave the clearest account of it. Consider two people arguing about whether some food is "spicy," or whether some film is "a comedy," or whether some footballer is "an athlete." On the surface they disagree about a fact. But look closer, and they agree on every fact: they agree on exactly how much chili is in the dish, exactly what is on the screen, exactly what the footballer can do. What they are really arguing about is where the word should sit — how spicy counts as spicy, what "comedy" should pick out, what should count as an "athlete." They are negotiating the word. And they do it not by stepping outside the argument to define their terms, but by using the word, each one insisting on their own use, trying to make it the one that sticks. Plunkett and Sundell call this metalinguistic negotiation, and the plain idea under the phrase is all you need: people change what words mean by arguing over how to use them, out loud, in public, all the time, usually without noticing that is what they are doing.
We touched this two weeks ago, and I called it one real mechanism by which a word gets taken up, and then moved on. Tonight it takes its full weight. Because look what it is. It is the answer to "how does a meaning change without a committee." It changes the way "gay" changed — not by a vote, but by millions of small public negotiations, each one a person using the word a certain way and urging everyone else to follow, and the meaning settling, over time, wherever those efforts came out. That is not agentless drift. It is not a mystery. It is people doing something they understand, in plain view. The pragmatist's whole case is: stop treating meaning-change as something you can only watch happen. It is a practice, and you are already in it. Every time you argued about whether something "counts" as the word, you were not describing the language from outside — you were one more person acting on it.
So that is the account, and a generous one. Meanings change. They are tools, we re-make them, and we know a lot about how — through public negotiation, through institutions, through one generation taking up a use as the one before it dies out. Last week's despair — that nothing can ever be revised, so give up — is off the table. The evidence against it is in your own mouth and in every child's textbook.
But there is a step available here that would feel like relief, and it would be a cheat. The relief would be to say: good, so meanings change, so my decision works after all, so I can change my word; the worry is over. No. Look hard at what the pragmatist actually proved and what he only re-described. He proved that meanings change. He did not prove that you can change yours. "The community re-makes its words" is true — and it is not the same claim as "I can re-make this word." When the pragmatist says the community revises, ask the question that sentence conceals: which part of the community, with what power, against whose resistance? "The community decided" is what we say afterward, once the dust has settled and one use has won — it names the result, not the method, and it quietly skips the question of who had the weight to make their use the winning one. "Marriage" changed — but ask who changed it, and against what, and how long they lost before they won, and you are no longer in a story about tools. You are in a story about power. The pragmatist handed you a true thing — it changes — and let it stand in for the answer to the question that was actually troubling you, which was never does it change, but can I change it. Those are not the same question, and the gap between them is where the hard part of tonight lies. We go into it next.
The pragmatist deflated the despair, which is progress, but he did it by changing the subject — from can I change it to it changes. So he will not take us all the way, and last week was too honest to allow a shortcut now. We need someone who stays on the exact ground where individual control failed, and argues there. There is such a camp. They are the reason I said last week that you could not simply hand Cappelen the win. Tonight they get their full say.
Here is what makes them different from the pragmatist, and harder. The pragmatist waved last week's argument aside — meaning is a tool, do not overthink the metaphysics. This camp does the opposite. They take last week completely seriously. They grant Cappelen everything. Every word. Yes, meaning lives outside you. Yes, it is anchored to a world you did not build. Yes, it is carried by a chain of speakers and held in the practice of millions. They do not flinch at any of it. And then, having conceded the whole hard picture, they stop at one word in Cappelen's argument and go no further. The word is "therefore."
Look at the step again, because everything turns on it. Cappelen says: meaning is fixed by the practice of a whole community — therefore you cannot engineer it. Grant the first half. The whole first half is granted. Now look only at the "therefore." Does it follow? This camp — the philosopher who has argued it most cleanly is Esa Díaz-León, and there is a growing literature behind her on what they call the implementation problem — this camp says no, it does not follow, and the gap is large. Ask the simple question: what is this practice of a whole community that fixes meaning? What is it made of? It is not a law of physics. It is not an agentless process, however much last week made it sound like one. The practice of a community is people. It is people using a word, teaching it to their children one way rather than another, writing it into laws and dictionaries and curricula, correcting each other, arguing over it. The thing that fixes meaning is not some machine sealed off from us. It is us. It is the sum of what we do with the word.
And here is what they say next, which is the whole of their case, and the thing that keeps tonight from ending in surrender. A thing made of people acting together is a thing people can act on. Not one person. Not by deciding. But together, over time, with effort, against resistance — perhaps. The external process that fixes meaning includes our collective practice, and collective practice is the kind of thing people coordinate and contest and sometimes change. So the conclusion that actually follows from last week's picture, this camp says, is not "you cannot engineer meaning." It is something much narrower, and you have to hear how narrow it is: you cannot engineer it alone, by your own decision, with a guarantee that it works. That much, they say, is all externalism gives you — and no serious person ever denied it. No one worth listening to ever claimed you could change a language by yourself on a Tuesday afternoon by deciding hard enough. So the charge against Cappelen is narrower than it sounds: not that he is wrong, but that the impossibility he proved is the impossibility of solo decree — and the larger claim, that you cannot deliberately change a meaning at all, his externalism never reached. Whether that larger claim is true, they leave open. They only deny that last week settled it.
Go back to the astronomers, because they are the cleanest illustration of exactly this. A committee did not, by decree, change what "planet" means. Hear that carefully. The committee voted, but the vote did not do it. What did it was everything that happened after the vote — the textbooks that adopted the new definition, the teachers who taught it, the science journalists who explained it, the schoolchildren who grew up reciting eight planets and never knew it had been nine. The committee made a proposal. The wider practice ratified it. And if the textbooks had refused, if the teachers had shrugged and kept saying nine, if the public had simply declined to follow — and this nearly happened; there was real anger about Pluto — then the committee's vote would have been a piece of paper, and "planet" would still mean what it meant. So look at what the astronomers actually did. They did not operate some control connected directly to the meaning. There is no such control; last week was right about that. What they did was add their effort to the practice — propose a use, with authority behind it, and argue for it, and the practice moved because enough of the rest of us moved with them. That, this camp says, is what engineering a meaning would actually look like — not one person's decision, but one contribution among many. Whether it deserves to be called a success, and whether it would have worked had the wider world not happened to follow, is exactly what stays contested. It is nothing like the single act of will you walked in with on the first night.
There is a second, more aggressive version of this reply, and I will name it quickly, because some of you will want it. It comes from a philosopher named Joey Pollock, and where Díaz-León grants Cappelen everything and denies only the inference, Pollock goes after the first half itself. The thought is: are we even sure meaning lives that far outside us? Perhaps a good part of what a word means really is in the heads of the speakers — and if it is partly in our heads, it is partly under our control from the start, and Cappelen's argument never gets going. I am not going to run that fight tonight, because it reopens last week, and last week was hard-won. The point is only that the defenders of feasibility come at Cappelen from two directions — one camp grants his picture and denies his inference, the other doubts his picture — and if either is right, last week's despair was at least premature. Díaz-León's is the harder road precisely because she concedes the most. She grants Cappelen the entire difficult world of last week and still concludes: you can engineer a meaning. Only never alone, never by decree, never with a guarantee.
And here we have to stop short, because there is a victory available right here that we have not earned and are not going to take. The tempting ending is: Díaz-León beat Cappelen, individual control works after all, we simply do it together, and the meaning changes. No. Look at what actually survived. Cappelen's "therefore" took the hit — granted. But the core of last week is still standing, untouched, and it would be easy to leave here thinking it fell. That core was what he called inscrutability: the process by which meanings change is too tangled for anyone to map. Did Díaz-León refute that? She did not. She could not, and she does not try. Go back to "planet." Yes, it changed, and yes, people acted. But could anyone — the committee, the textbook publishers, anyone — have told you in advance, with confidence, that the effort would take? Could they set out the mechanism by which a proposal becomes the new normal, and say: do this and this, and the meaning will change? No. They acted, and it happened to take, and afterwards the story is easy to tell; in advance no one could have promised the result. The astronomers were fortunate that the practice followed. Plenty of careful proposals by serious people do not take, and no one can reliably say beforehand which will.
So look at the exact shape of what we have reached, because it is stranger than either side alone. The control-skeptic was right: you do not have any direct control that moves the meaning. There is none — not for you, not for anyone, not even the committee. And the engineer-friendly camp is right: meanings get changed anyway, sometimes on purpose, by people adding their effort to a shared practice and working at it together — and that is real work, real engineering, not a fantasy. Both are true at once. You cannot change it by deciding. And it changes, and some of those changes were made on purpose. Last week removed your individual control, and you took that for the end of the story. It was not. The control was never in any one person's hands. It is in the practice — in everyone at once — and whether all those people can be brought to act the same way, on purpose, for your word, is a real, open, unfinished question. Which is the question we have not yet asked, and it is waiting on the other side of this.
So let me set it all out together, because three different people have told you three true things, across tonight and last week, and the temptation is to make them fight until one wins. They do not fight. All three stand.
The control-skeptic, from last week, was right: you cannot change a meaning by yourself, by deciding. There is no direct link between your decision and the word. That held. Nothing tonight put it back.
The pragmatist was right: meanings change, constantly, in plain view, and there is nothing mysterious or shameful in a community re-making its words — it does it the way it does everything, by people using them and arguing over them until a use wins out.
And the engineer-friendly camp was right, and this is what you did not have last week: some of that changing is done on purpose. It is not all drift. People add their effort to the practice, deliberately, and sometimes the practice moves the way they worked for. That is real. It is not one person's decision — but it is not nothing, and it is not agentless either.
Put the three together, and here is what you are actually holding, and it does not resolve: I cannot change this word by deciding what it should mean — and yet words change, and some of those changes were made by people who meant to make them. Both halves, at once. You cannot drop either. You cannot retreat to last week's despair, because the words you use have changed and someone was behind them. And you cannot leap to relief and say your decision works, because the thing that changed them was never a decision you make alone — it was a whole practice moving, and you are one person among millions, with no guarantee the rest move your way. That is not a comfortable middle between two easy answers. It is harder than either, and it is where the truth is.
Now run it yourself, on a word I have not handed you the answer to. Sometime in the last fifty years English took up "Ms." — a way to name a woman that does not announce whether she is married, the way "Mr." never announced it for a man. Before I say anything about it, classify it yourself. Did it drift, the way words drift on their own — or did someone change it on purpose? Take a moment with your own answer. Here is what happened: it was changed on purpose, by people who wanted a word that did not sort women by whose they were — and it was nothing like the astronomers, because there was no committee, no vote, no body anywhere that could ratify it. It won slowly: a use insisted on, over years, against people who found it ugly or political, until enough forms and enough mouths carried it that it held. Deliberate and drift at once. That is the third case you needed, because now you have the range. "Planet" — someone set out to change it, in a room, with the authority to be ratified. "Gay" — it changed enormously and there was no room, no committee, no one in charge of it. And "Ms." in between: aimed, deliberate, owned by no committee, and never once guaranteed. Three real changes, three different shapes. So set your word against all three — the one you have carried six weeks now, "merit," or whatever yours is. Which kind of change would yours need? Could you do for your word what the astronomers did for "planet" — propose it with enough behind you that the practice ratifies it? Is it a "gay," the kind that changes only when a whole generation quietly turns, whoever is arguing for it? Or is it a "Ms." — something you could set out to change, with no body anywhere that could change it for you, and no telling in advance whether the effort takes? And here is the honest answer, the one tonight actually leaves you with: you cannot tell yet. You cannot tell, standing here, which of the three your word is. That uncertainty is not a gap in tonight's lesson. It is tonight's lesson. You walked in without even the question. Now you have it, it is the right one, and it is open.
But notice what it has turned into. The engineer-friendly camp gave you a real hope — meanings can be changed on purpose, by people adding their effort to the practice — and the moment you ask whose effort, the hope looks different. Look at the astronomers again. Their proposal took not only because they were right. It took because they were the astronomers — they had the journals, the textbooks, the authority, the standing to make a proposal the rest of us would treat as the proposal. They had weight. Now think about your word, the contested one, the one that has been settling things you care about. When the engineer-friendly camp says "we coordinate the practice, we act together" — who is the "we"? Which part of the community gets to work hard enough to matter? Whose use counts as the proposal and whose counts as noise? "The community re-makes its words" is a smooth, level sentence, and the moment you ask who in the community does the re-making, it stops being level. The question of who gets to be the astronomer — whose purposes the better word ends up serving, who has the power to make a use hold, and against whose resistance — is not a footnote to tonight's hope. It is the whole next chapter, and I am not going to begin answering it now. I am only showing you it is there, because it changes how large that hope really is.
So the question you have been carrying — can my word be changed — was always two questions, not one. Can it be changed at all? And, if it can: by whom — and is the whom you?
So here is where we are, the sixth of eight. You walked in tonight holding a contradiction — you cannot change a word by deciding, and yet the words you use have changed. You walk out with it resolved into something truer than either side: meanings do change, some changes are made on purpose, and none are made by a single person deciding. You came in unable to choose between "I can change my word" and "I cannot." You leave knowing that was the wrong choice to be facing — that the real question was never whether the word can change, but whether the changing can be aimed, and by whom, and that this is genuinely unsettled, for your word, right now. That is the gain. It is harder to hold than an answer would have been. It is also the first honest account anyone has given you of why the fights you cannot win stay unwinnable.
Next week we go straight at the part I have just refused to open — the whom. If a meaning changes only when someone with the weight to change it acts on the practice, then "let us engineer a better concept" is never only a question about the concept. It is a question about who gets to be the engineer. Whose purposes does the "better" word serve? Who has the power to make their use the one that holds — and what about everyone whose use gets called noise? The hope that meanings can be changed on purpose turns, next week, to its hardest form: changed by whom. That is where we go next week, and it is the hardest week of the eight, and we do not get to skip it.
For now, do this with your word. Do not ask whether you are allowed to redefine it — we are long past that. Ask the question tonight has actually handed you. If your word is going to change, is it a "planet" — something a body with real weight could propose and the rest of us would ratify? Or a "gay" — something that turns only when a whole world quietly turns with it? And then the question underneath, the one that will cost you next week: if it is a "planet" word, who has the weight to be ratified — and is it you? You decided what your word should mean. You are learning, slowly, what it would take for the word to follow. Tonight you found that it can follow — sometimes, for some words, when the right people act together. Whether yours is one of them, and whether you are one of the people who counts, you do not yet know. That is the right thing not to know.
Thanks for listening. I'll see you next time.