Last week ended on something that felt, for the first time in this series, like firm ground. Last week offered a hope, and it was an earned one. Meanings move. Not by one person pulling a lever — that possibility was gone, and it stayed gone — but through a shared practice: many people, using a word, and sometimes setting out on purpose to change how it is used. And you were one of those people. After two weeks in which each certainty gave way, that was something to hold. You cannot move a word on your own — granted. But you are not nothing. You are part of the practice that moves words. For a week, that seemed enough.
And then, sometime in the week, a word changed on you.
Perhaps it was a word that names your work, and a room you were never in decided it now meant something a little different, and the new meaning was simply there on the Monday, whether you had asked for it or not. Perhaps it was a line in a policy — what counts as "violence," what counts as "harassment," what counts as "essential" — redrawn by people with the authority to redraw it, and you woke up inside the new boundary whether you would have drawn it there or not. Perhaps it was a diagnosis. A category on a form. A word that once meant one thing about you and quietly came to mean another, and no one asked you.
And you noticed the thing the hope had passed over. Yes — the word moved, and it moved on purpose. Someone changed it. But the purpose was not yours. The power was not yours. And the word applied to you all the same. So last week's good news reads differently now. Meanings can be moved on purpose — and you were so relieved to hear it that you never asked the next question. On purpose by whom? For whose purposes? Ask that, and the hope looks entirely different, because you realise you have spent this whole series quietly assuming that the person with the better word and the power to install it was going to be you.
That is tonight. The hope you were handed last week has a second side, and it was there the whole time.
This is Philosophy for Us — philosophy for everyone, no degree required.
This is the seventh episode of eight, and I will say plainly, so you can prepare for it: this is the low point of the series. Last week three things turned out to be true together, and none of them defeated the others. You cannot move a meaning by deciding on your own — that held, from the hard week before it. Meanings do move, constantly, in plain view — the pragmatist was right about that. And some of those changes were made on purpose, by people who set out to make them — that was the new part, the part that felt like hope. You were left holding all three at once, and one question with them: not whether a meaning can move, because plainly it can, but whether the change can be aimed — and, the part I said we would return to, aimed by whom.
Tonight we return to it. We go straight at the "whom," and we do not look away from where it leads. If a meaning moves only when someone with the power to move it acts on the practice, then "let us engineer a better concept" was never only a question about the concept. It is a question about who gets to be the engineer — whose purposes the better word ends up serving, who has the power to make a use stick, and against whose objection. I am not going to lead us back out of that tonight. That is not this week's task. Tonight we go into it and stay there, because no one can honestly be shown the way out of a difficulty they were never made to feel. So let us feel it first.
Let us start with the sentence that did the quiet work last week, and examine it until it stops sounding like good news.
Here is the sentence. The community re-makes its words. That was the engineer-friendly camp's whole rescue last week. Meaning lives outside you, yes — granted, all of it. But the thing it lives in is not weather and not a machine. It is a practice, and a practice is people, and people coordinate, and use a word one way rather than another, and sometimes do so on purpose. So you are not helpless. You take part. The community moves its words, and you are in the community. I said that last week and I meant it, and I am not taking it back tonight. It is true.
But listen again. The community re-makes its words. It is a smooth, untroubled, friendly sentence. And the moment you put one question to it, the smoothness goes and you see what it was concealing. The question is: which part of the community? Because "the community" is not one thing moving in one direction. It is millions of people, wanting different things at once, for and against, hard and soft — and the word does not end up where the most people want it, or where the best argument points. It ends up where the most powerful users insist on it. And power is not handed out evenly.
Go back to the astronomers, because last week I used them as the clean, hopeful case, and there is a part I let pass. A committee voted, "planet" changed, and Pluto fell out of the set. And I told you the vote did not do it — what did it was the textbooks and the teachers and the schoolchildren taking up the new use. The practice ratified the proposal. All true. But now put the question I have just handed you. Which people made that ratification happen? Why did the textbooks change? Not because the astronomers were necessarily right — plenty of astronomers were furious, the measure nearly failed, and there is a fair case that Pluto's demotion was always somewhat arbitrary. The textbooks changed because it was the astronomers who said so. They had the journals. They had the authority. They had the one thing that turns a proposal into the proposal rather than one more opinion: the standing to be taken up. When someone on the internet says Pluto is still a planet, that is noise. When the International Astronomical Union says it is not, that is the definition in the next morning's textbooks. Same proposal. Different people. And the practice followed the ones with the authority.
That is the thing the hope passed over. It is not enough to know that words move through the practice and that you are part of the practice. You have to ask how much power you have in it. Because the practice does not take up every use equally. It takes up the use it is already disposed to follow — the one with the institution behind it, the court, the platform, the textbook, the credential, the numbers, the money, the airtime. Whose use is treated as the proposal, and whose use is treated as a mistake to be corrected — that is settled before anyone weighs the merits, by who is doing the pushing.
Now take the pragmatist's term for all this, because he had one. Three weeks ago, and again last week, I gave you the idea from the philosophers David Plunkett and Tim Sundell — that much meaning-change happens through what they call metalinguistic negotiation. People dispute whether something "counts" as the word, each using the word their own way, each urging everyone else to follow, and the meaning settles wherever the dispute comes out. I called it the mechanism. It is real, and it is still real tonight; I am not unsaying it. But stay on that word "negotiation" for a moment, because it is doing the same smoothing-over the first sentence was doing.
A negotiation has parties. And parties to a negotiation almost never arrive with equal power. When a union and a single worker "negotiate," we do not pretend it is a meeting of equals. When a tenant "negotiates" with a landlord who has a hundred other applicants, the word "negotiation" is accurate and quietly misleading. So when the pragmatist says we negotiate our meanings — yes. We do. But a negotiation is not a vote, and it is certainly not a seminar in which the best argument wins. It is a contest, and "negotiation" can be a polite word for a contest the stronger side was always going to win. The pragmatist is right that this is how meanings move. He simply hands you the mild version of the word and lets you not notice that some parties can end the negotiation on their own and some cannot get a seat at it.
So the friendly sentence turns out to have quietly left the power out. "The community re-makes its words" invites you to picture the whole community, calm and entire, coming round together to the better use. But there is no single thing, "the community," that does anything. Take the picture apart and you find particular people, with particular power, pushing particular uses — and other people, with less, pushing back and losing. The smooth sentence is what remains once the stronger side has won and the contest has dropped out of view. It reports the result and hides the struggle. And the struggle was never a fair one.
And here I have to be careful, because there is a cheap version of this point and I do not want it. The cheap version is: it is all power, might makes right, words mean whatever the strong say they mean. Hold off. We are going to come right up to that claim tonight, and I am not going to let you take it early and easy, because it is comfortable in its own cynical way and we have not earned it. The point for now is narrower and harder. It is only this: the hope was real, but it was incomplete, and what it leaves out is power. Meanings move through the practice; some of those changes are aimed; and which aim the practice takes up is a question of who has the power to be taken up. You take part. Tonight you find out how much power you have.
But even that is not the bottom. So far I have only told you that the powerful win the contest over whose use sticks — and you half knew that walking in; it is grim, but it is no surprise. The surprise is underneath, and it concerns the one tool in this series I told you was clean: the engineer's method. The calm, principled move the messy disputes never had. I sold you that method as the sober one — the single part of all this that was not a fight. I need to show you now that the standpoint, the question of whose side you are on, does not enter only at the end, in the contest over uptake. It was there in the engineer's very first step, before any fight started. It was in the method itself. That is next.
For six weeks I have handed you a method and told you it was the clean one. Let me restate it the way I first put it to you, because it sounded good — genuinely good — and that is the point, before I turn to the part I did not show you.
The method is the engineer's — Sally Haslanger's, most sharply. Stop asking what a word really means, she said; that question chases a folk concept that may be confused, or unjust, or simply old. Ask instead what concept we should have — given the legitimate purposes the word is meant to serve. Define the better word by the work we need it to do. Then the right definition is not the one that fits how people happen to talk; it is the one that best serves the purpose. That was the calm, orderly alternative to the disputes. The arguments about whether something is "really" violence or "really" a disorder are a fight; this was the sober move that stepped out of the fight and asked the orderly question — what do we need this word for, and what shape should it have to do that job well. You did it on "race" at one of our earlier stops, and on "disorder" at another, and it felt like the most honest thing you had done in either dispute.
Now look at the point on which the whole method depends. The legitimate purposes the word is meant to serve. Read it again slowly, because there is a word hidden in it doing enormous work while drawing no attention to itself. The word is "legitimate." And behind it a second, quieter one: our. The concept we should have, given our legitimate purposes.
Whose?
Whose purposes are the legitimate ones? Who is the "our"? Because the entire method — calm, principled, sober — does not begin once you have answered that. It begins by assuming it has been answered. The engineer starts work at step two. Step one — these are the purposes worth serving, this is the work the word should do, this is who the word is for — step one is already finished before the engineering begins. And step one is not an engineering question at all. There is no neutral, technical, view-from-nowhere way to do it. Choosing which purposes count as legitimate is taking a side. It is a standpoint. It is the most standpoint-laden part of the whole procedure, and the method reduces it to a single word — "legitimate" — and moves on to the satisfying part, the designing.
Let me be exact about what I am saying and what I am not, because here it would be easy to overstate. I am not saying Haslanger has no answer to "whose purposes." She has one. She would say the purposes can be argued for, out loud, defended, contested — that asking the word to help locate and undo a real injustice is a purpose you can make a public case for, and she makes it. That is fair. The charge is not that the method is secretly empty. The charge is that the method's first step is a standpoint, and that the method's whole appeal — that it was the sober procedure which stepped out of the fight — was always partly false, because the fight is in the first step. "What should this word mean?" only sounds neutral. Unfold it and it is "what should this word mean for the purposes I have already decided are the legitimate ones" — and the moment two people disagree about that, you are back in the dispute the method promised to rise above. The method did not escape the standpoint. It moved it one step earlier and stopped examining it.
And here is the part that should cost me something, not only Haslanger, because I am the one who has been running this on you. Go back to "race." When you ran the ameliorative move on "race" — should the word pick out a biological fiction, or should it be rebuilt to name a position in a social hierarchy so that it can do justice-work — that "so it can do justice-work" was the purpose. And it is a purpose I share, and you probably share, and I think we can defend it. But it was a choice of purpose, made from a standpoint, before the engineering started, and I walked you through it as though the hard part were the design. It was not. The hard part was the part I skipped — whose justice, named by whom, against whose account of what justice even is. Same with "disorder." Function or flourishing — that whole dispute was a dispute about which purpose the word should serve, and "flourishing" is not a neutral term either; someone's picture of the good life sits inside it. I told you those were clean uses of the method. They had a standpoint in them the whole time, in the first step, and I pointed you at the second step so that you would see the method's power and not its politics. So when I say the standpoint was hidden in the engineer's first move — I am not catching Haslanger out from a safe distance. I am telling you that the move this show taught you to trust has a stance built into it that I let you not see. I am in this. I made the move too, on you, four times.
Now I have to take you one level lower, because there is a thinker who says that even this — even "whose purposes, within our shared language" — is still too comfortable, still too small a version of the question. He is Kwasi Wiredu, a Ghanaian philosopher, and in his 1996 book Cultural Universals and Particulars he asks a question that goes underneath everything we have said tonight. We have been arguing about whose purposes get to steer our concepts. Wiredu asks: and where did the concepts themselves come from? And where did the very idea you are using right now — that a concept is a tool you can redesign for a purpose — come from?
Here is his thought, and it is quiet and it is serious. The words a tradition reasons with — and the deeper picture of what words and thoughts even are — are not the neutral inheritance of all humanity. They are the particular inheritance of one language, one history. Wiredu spent years working between English and Akan, his own first language, and he kept finding places where a problem that looked like a deep universal puzzle in English simply did not arise in Akan, or arose in a wholly different shape, because the concepts were cut differently. So he asked what it would take to think a problem in another tradition's terms — and he called the project conceptual decolonisation: the work of noticing which of your concepts are universal and which are just your own language's furniture, mistaken for the structure of reality.
Now turn that on tonight. We have spent seven weeks inside one picture — that words are tools, that meanings are things you engineer, that the smart move is to ask what a concept is for and redesign it to serve the purpose better. That picture has felt, all this time, like plain good sense — like simply seeing clearly how language works. Wiredu's question is: is it? Or is "a concept is an improvable tool" itself one tradition's way of understanding the world — a way of thinking that grew up in one place, in one history, with one set of assumptions about purposes and persons and what it is to improve a thing? Because if it is — if even that picture belongs to one tradition and not to everyone — then the deepest standpoint in the room tonight is not whose purposes win the engineering contest. It is the standpoint built into the idea that there is an engineering contest to win at all. We picked up the engineer's tools to get underneath the fight, and Wiredu asks whose tools those are, what they quietly assume, and what other ways of understanding a word they have already ruled out before anyone spoke.
That is the bottom of it. Not just that the powerful win the uptake contest — but that the method we trusted to rise above the fight had a standpoint in its first step, and that the whole picture of words-as-tools may itself be one standpoint presented as plain common sense. There is no firm place to rest down here. I told you we were going down. This is as low as it goes.
So now we come up to the claim I would not let you take earlier. We have earned the right to look at it. Here it is, at full size, with nothing softened.
If a meaning moves only when the powerful act — if "the better word" is just the word that served the purposes of whoever got to set "legitimate," and won the uptake contest to make it stick — then what, exactly, is "improvement"? Say it without the flattering words, and "we engineered a better concept" reads as "we had the power, so our word won." And then the whole thing this series taught you to trust — the careful move, the ameliorative method, the sober procedure — starts to look like something much uglier presented as a respectable method. Not a method for finding the better word, but a method for making the word that won look like the better word. A polite, philosophical name for the oldest thing there is: the strong deciding what the words mean, and calling it progress. Conceptual engineering collapses into conceptual politics — power, with a footnote.
That is the worst version. And it is entirely serious — not a straw man, not paranoia. People who do this work for a living take exactly this danger seriously, and the honest ones lose sleep over it. So I am not going to insult you by knocking it down with one quick argument. I genuinely cannot, not tonight, not cleanly. And I want to be just as clear about the other side: I am also not going to tell you the worst version is simply true — that it really is all power, that the method is a fraud, that you should throw it out. Because that is its own cheap comfort. It excuses you from trying. It is all power, nothing to be done, no point even trying to fix the word. That is not the bottom telling the truth; that is the bottom handing you an excuse. So I am leaving you in the exact place I have you: unable to wave the worst version away, and not willing to let you settle into it as the final answer. The worst version is still in play. We do not get to decide tonight whether it is right.
Now run it. Not on a word I have prepared for you — on a fresh one, cold, so that the work is your own.
Take "essential worker." You watched this word get made. A few years ago it barely existed as a category, and then, almost overnight, in an emergency, it was everywhere — and it was a deliberate revision if ever there was one. Someone sat down and drew the line: these jobs are essential, these are not. And it stuck, fast, harder and quicker than "marriage" ever moved — because it was installed by people with enormous power. Governments with emergency orders. Agencies with lists. Employers deciding who received the letter that let them travel to work. That is the practice being moved by the most powerful bodies in the country, on purpose, in weeks. By the plain question of whether a meaning can be deliberately moved — the question that occupied us for two weeks — "essential worker" is a resounding yes. It moved, it was aimed, it landed.
Now put tonight's questions to it, and see what becomes of the triumph. Whose purposes did that word serve? Read it honestly: in large part the purpose was keeping things running — the supply chains, the shelves, the economy — and the word's job was to name the people who had to keep turning up so that everyone else could stay home. Whose power drew the line? Not the workers'. The word was applied to them — it named them, it governed their week, it sent them in — and the people it named had, almost by definition, the least power over the label now running their lives. "Essential" sounded like honour, and for many of the people wearing it, it worked like an order. And against whose objection? Ask who would have drawn the line elsewhere — the worker who would have said, if I am so essential, where is the hazard pay, where is the protection, why does "essential" amount to "expendable" — and notice that their use of the word counted for almost nothing, because they had no power over it. The word was theirs to wear and never theirs to write. So: was "essential worker" engineered? Yes — cleanly, by tonight's standard. And whose word was it? Not the people it was about. That is the whole episode in one word, and you ran it yourself.
And now the one that costs you, because so far this still has villains in it — the powerful drawing lines on the weak — and it does not implicate you. So bring it home. Bring your word. The one you have carried since the first week. "Merit," or whatever yours is — the word you most want to fix, and you want to fix it for good reasons, real ones; you are not the villain here. Fine. Now put to it the questions you just put to the emergency. Your better "merit" — whose purposes does it serve? Yours, and the people you are trying to protect with it, certainly — but say that out loud and notice it is still a whose. It is still a side. Someone disagrees with your purpose, and not all of them are fools or bigots; some of them simply stand somewhere else. And here is the hard one, the one with no exit: even if you are right — even if your purpose is genuinely better, even if your design is sound — for your word to win, you have to have enough power, or join with enough others, to make the practice follow your use rather than theirs. Which means winning the meaning, even for the best reason in the world, is still winning an uptake contest. It is still power. Being right does not excuse you from needing the power. There is no version in which the good revision rises up and installs itself because it deserves to. It wins the way "essential worker" won, or it does not win at all — and the only thing your good intentions change is which side you are pushing for. Not the fact that it is a push.
So here is where we are, the seventh episode of eight. You walked in standing on a hope — meanings move, and you are one of the people who move them. You walk out still holding that it is true, and holding three harder things on top of it. That whose use counts is decided by power, and power is not handed out evenly, and you may not have much. That the calm method you trusted to rise above the fight had a side chosen in its very first step, before any engineering started — and I am the one who taught it to you without pointing that part out. And that even the picture underneath it all — words as tools you redesign — may be one tradition's inheritance, mistaken for the way the world simply is. That is what you have now that you did not walk in with. It is not comfort. This is the hardest place the series goes, and you are standing at it, and I have not built you a way up.
Next week I will not pretend to hand you one cheaply. Next week is the last one, and it does the only thing left to do. It takes a single word — the one word you most want to fix, the real one — and runs the whole method on it, end to end, every question we have raised in seven weeks, all at once, on your word: whether the question is even the right kind; whether the concept is actually broken or you just dislike it; what fixes its meaning, and so what moving it would really take; whose purposes, whose power, against whose objection. And then it refuses to tell you the answer. I am not going to promise you, walking into it, that there is a rescue waiting at the end. I do not get to promise that. What I can tell you is that you will walk out holding your own word, and the questions, and no one handing you a verdict — which, if you have been listening, you already half suspect was always the most honest thing this show could do.
For now, do this, and do not soften it. Take your word. And do not ask whether you are allowed to fix it — we buried that question weeks ago. Ask the one tonight has handed you. If your word is going to move, whose use makes it move — and is yours among the ones with the power to be taken up? And the one underneath, the one I would carry carefully: if your good word can win only the way every word wins — by power, by the stronger side pushing harder — then is what you have been calling fixing the word any different, underneath, from what the people you are fighting are doing to it? You do not get an answer to that tonight. You are not supposed to. That is where we are. Stay with it a while.
Thanks for listening. I'll see you next time.