On 17 July 2025, I sat down with Kelsey Piper to chat about politics and social epistemology. You can listen to the audio file, or read the transcript below, which has been edited for clarity.
ZMD: Hi, I'm Zack M. Davis, here today with journalist Kelsey Piper to talk about how political pressures shape our speech and therefore our world-models: what gets said, what gets left unsaid, and how that changes over time. In particular, we had an election not too long ago, which had various impacts on our information environment that I'd like to try to make sense of with you.
KP: Yeah, I think the thing you initially reached out to me about was a Tweet that I sent a little while after the election, which was kind of lighthearted.
ZMD: I actually have that here. Quote,
having a woman as fire chief is I guess related to DEI but given that you have a woman fire chief obviously she'd be a lesbian so I don't think the lesbianism counts for any extra DEI here
End quote.
KP: So I showed this quote to some gay people, who all thought it made perfect sense, and I showed it to some straight people, who were honestly extremely confused by what I was trying to say. So in case you have any straight people in your audience, gay women are much likelier to be into traditionally masculine hobbies like carpentry and math and firefighting and stuff like that, and I would expect that we are wildly disproportionately represented among firefighters. I'm just adding this clarification because I mentioned this to several people who were like, I just don't get it at all; I don't understand that.
ZMD: Do you think they really didn't get it, or were they pretending to not get it?
KP: No, I think they really didn't get it. I think if you don't know very many gay people and just haven't thought very much about it, you have never organically run across differences in hobbies between gay women and straight women. I think they were sincere.
ZMD: So the reason I was curious about that is because it didn't seem to me like something I would have expected Kelsey Piper to tweet if Kamala Harris had won the election.
KP: Yes. So when you reached out to me and were like, would you have tweeted this if Kamala Harris won the election? I was like, I don't think so. I was quite upset about Trump being the president. I think most of his policies will be quite bad for various stuff that I care about: most importantly, disease control overseas, which just saves a ton of people, and I'm personally really enthusiastic about it, but also, I don't really think it's going to be great for AI. Why are we letting Nvidia sell chips to China again? Because Jensen Huang talked to Trump and Trump thought it sounded like a good idea. Why are we deporting tons of people who are working and not committing crimes? Because Stephen Miller wants it. Very few of the things that are happening are things that I'm happy about.
Given that, I try and take a step back and go, what could I and people in my general reference class have been doing differently over the last four years that might've caused more of the stuff that we cared about to happen in the world? And one of my diagnoses was, cowardice, being unwilling to get mild social disapproval. Not refusing to say true things we thought were really important: I actually think I always said true things I thought were really important, even when this involved disagreeing pretty substantively with progressives, but saying true things that aren't very important, saying true things that are just kind of small and trivial.
I feel like a lot of the stuff that happened over the last decade was stuff where there were just things you expected people to be a little annoyed at you about, or to get some people saying, wow, that was in bad taste or whatever. So you just wouldn't say them because that was a lot easier than pissing off ten people to no effect. But on a societal scale, if everybody gets shaped that way, then you only have various true, interesting, but not that important stuff being said by people who are in a totally separate information bubble, and that matters.
Also, you get a reputation for being humorless scolds. I think that's bad. It is hard to differentiate yourself from people who have quite different beliefs than you, if you only occasionally take stands about really big things and don't say a bunch of small incidental things which might nonetheless add up to part of your worldview.
Anyway, I thought the gay rights movement did better when people felt less like, "I'm scared that I'm going to get in trouble for saying the wrong thing about gay people" and more like, "Ah, gay people; this is a quirk of our society." And so I was intentionally like, what if I'm a little bit more willing to say things that I expect some people to have a negative reaction to, but that I think are true, and I think are fine, like gay women are more likely to become firefighters than straight women.
I wish, looking back over the last four years, not that I'd taken big stands on anything different—I did take the big stands I believed in—but that I'd said more small things. And I wish that it had been easier to differentiate me from someone who had different beliefs than me. And so I've been trying to do that. I thought it was funny that you thought of that Tweet in particular, because it did feel to me like—most Tweets, if you asked me, why did you Tweet this? I'm like, I don't know. But that, I had specifically been thinking about, what would doing better communication have been like?
ZMD: Yeah, I wasn't surprised that you would have that thought. I was surprised that you would say that in public. Because neoliberals and progressives are not actually the same thing, but there's this coalitional relationship.
KP: So in particular, I think that neoliberals have been reluctant to annoy progressives to a degree that is actively counterproductive. I think that the Democratic Party has two factions; they disagree on a ton of important stuff. I think that the neoliberals are right on nearly all of those disagreements, and the progressives are wrong on nearly all of the disagreements. In California politics, when you're trying to vote, it is useful to figure out who is the moderate Democrat on the ballot, and who is the progressive Democrat on the ballot. In my opinion, the progressive Democrats will make terrible policy calls in office, and the moderate Democrats will make good policy calls in office. But they both run as Democrats. This is kind of insane. You could imagine a world where the Republican Party in California was just very liberal, and these people call themselves Republicans. But because of the nationalization of our politics, that doesn't happen at all. They all run as Democrats; you just have to squint.
So if you're politically informed, you're aware that these factions exist. You have maybe seen some of their veiled or less-veiled sniping at each other. But most people just look at the Democrats, and everybody I talk to who's right-of-center, this distinction is often extremely illegible to them. And they're like, you guys all think this. And I'm like, no, no, no: a bunch of people think that. A bunch of other people strongly disagree. But the neoliberals have tended to try pretty hard not to offend the progressives. Often, this has taken the form of neolibifying progressive ideas. You have this progressive agenda; we will find a technocratic set of reforms that we can sell as part of that thing. I think it would be much more productive, honestly, to just say: nope, that thing is stupid; however, here are some policy proposals which are not your thing, because your thing is actually stupid.
People disagree with me, but there's a lot of disagreement over how much to fight. But to me, the most decisive reason that this should be a fight instead of the neoliberals trying to play nice with the progressives is one, they don't play nice back; they're assholes. Two, I think that if you are right-of-center, and you want the Democratic Party to be an effective force in American politics, you shouldn't have to know a bunch of inside lore to figure out who your allies are. It should be pretty clear. I think people are getting a little bit better about this now. There was a bunch of Abundance fighting. But one thing that that did do is get some people to carve out stances on this one way or the other. I have, depending on when you publish this podcast, some news about upcoming stuff in this vein. But, yeah, it's a problem.
ZMD: I think part of the reason right-of-center people have that perception is because if there is active collaboration to downplay these ideological differences, then it's like you're on the same team. And so people who are trying to do different things on the different team, they're right to perceive that team dynamic.
KP: I've been watching the same thing play out on the right, and it has been revealing about why this happens and how much of it is culpably not having been responsible about carving out distinct identities, and how much is being a team player, and how much it's it just being always very hard as an outsider to parse other groups.
On the right, I've been watching it play out between Christian conservatives who are kind of anti- a big chunk of the MAGA base and will call it "woke right" or whatever. And there's a lot of the people who they call "woke right" would absolutely hate that term and do not use that term to describe themselves. I don't totally know what term they do use to describe themselves, but this division is real.
Reasonably often you see appeals that are like, the reason you guys have to stop doing what you're doing is because it is getting in the way of all of us uniting behind President Trump and his glorious agenda to do all of these things. Given that I don't like President Trump, and I think his glorious agenda is bad for our country, these are both the other team. I'm against both of them. I still would like to have a much better understanding of—I care a lot who wins that factional battle. I think that they are very different in terms of the effects of them winning the factional battle on the Republican Party and therefore on the country. But it is true that both of them are the other team.
And it is also true that some people want the worst elements of the right, especially Clinton's team in 2016 famously wanted this, right? They wanted Trump because he would be the weakest opponent. They specifically wanted the illiberal populist right to ascend because they thought it would be the easiest to beat. I think this was a correct assessment in that Clinton stood more of a chance against Trump than she stood against any other Republican nominee.
It was an incorrect assessment if you care about the country. I think from the values of all the people who made this bet, the thing that happened was much worse than if they had—to be clear, I don't think they were causal. But to the extent they had any effect, it was an effect that was very bad by their values and by my values. I think you shouldn't do this; you should try and get the better of your enemies to be ascendant in as much as anything you do is dedicated at intervening in conflicts among your enemies. Trying to get the worst one as a counterbalance or because they'll be weaker in the next—I don't know, when I look at foreign policy, when I look at like domestic group politics, it feels to me like that has a very poor track record and is corrosive and kind of evil, so I'm down on it. But you still want to know, even if they are the other team and it is in some context reasonable to treat them all as the other team, I think it is valuable to know which of the factions within the other team you want to be ascendant on the other team and which ones are more possible to compromise with. Similarly, I think it is useful to try and be clear about that to people who disagree with you, so that they can have that mental handle if it's useful.
ZMD: But what about when there's a conflict between maximizing clarity and securing victories for your team?
KP: A thing about politics is that a lot of it is about what you successfully communicate to people, and what you successfully communicate to people is always going to be an impoverished and horrible subset of what you actually believe. I'm tempted to be a purist about this and be like, fuck messaging, say what you actually believe loud and proud and clear. In the information environment I observe us to be in, this feels like fake posturing of honesty. The act of being loud and clear and honest does not feel closely related to the act of causing people to have an accurate model of you, or the stuff that you care about, or the effects that you're having on the world. This is not an argument for lying. I don't think that you should ever do that. But I think that the problem of causing people to have accurate beliefs in our current information ecosystem is like 20% a problem of loud and clear saying what you truly believe, and way more than that, a question of which circulating lies do you bother responding to if there are a million circulating lies and you want to respond to the important ones but not become someone who is defined by signal-boosting things that are lies even in order to refute them.
On Twitter today, I want to say the majority of the posts I saw were clear provocations where it felt to me like the intent was to farm engagement. You know, the person who's going like, you won't be genetically related to your mixed-race children. I don't even think they care about that. I think they noticed that the fastest way to get 50 million views on Twitter is to say that and then have every single person on Twitter jump in to quote-Tweet you and explain how genetics actually works.
ZMD: And how there's different ways of quantifying relatedness such that if you actually understand the math, there's nothing to be confused about, but the math doesn't compress to a Tweet.
KP: Exactly. So that is a perfect example of me being like, okay, I feel like being loud and clear about the truth is inadequate. You should be loud and clear about the truth. That's good. I'm in favor of that. But trying to make sense of the world in this information environment, do you write a blog post trying to explain the different understandings of relatedness and then link that blog post every time someone tweets this for engagement? Which to be clear, I see it like every month or two. So there's clearly a bunch of people who have noticed this is a great engagement strategy. Do you get a lot of your own engagement by just sharing it to be like, this is dumb and bad, and I disagree with it?
I don't think you should generally just share things to say they're dumb and bad and you disagree with them without some reason to think they're also influential or important. But again, this is a bunch of calculations that are not just about, you should say the truth. Then you have to ask a bunch of questions about, is this an important misunderstanding people have? Are the people who misunderstand it, do they have any overlap with my audience? Am I going to be saying this to anybody who doesn't already know this? Or am I just saying this to an audience that already completely agrees with me about everything?—which is sometimes still worth doing, in part because sometimes you're wrong about whether your audience agrees with you about everything. And in part, because sometimes a lot of people believe something because they got the impression it's believed by people they trust, and they don't actually know the actual argument for it. It's worth doing that. I feel like telling the truth is important, and also, a ton of the decisions I end up making are not about, is this true? It's about, is this a representative fact? Is this a surprising fact? Is this a fact that should be surprising, even though people won't find it surprising for whatever reason? Is this lie, a lie that a lot of people actually believe or a lie that one bot came up with and that other bots are sharing and that probably no humans, or not that many humans, have been persuaded by, and it's not that worth the review? I don't know. I think about this a lot. It's kind of depressing and very different than how I thought about it like ten years ago.
ZMD: That's why I was so interested in having this conversation. Because I feel like there's alpha in making explicit that—I think a lot of the smartest people on Twitter, including people who are not on your team, see the same things you do. And yet you're playing different characters, deciding which true statements is it in my interest to talk about.
KP: Yeah, I think almost everybody is doing that. Some people are consciously thinking about which true statements are important and representative and worth talking about. But every single person is making that judgment on some level. I would just disbelieve anybody who was like, ah, yes, I share all true statements that come to my mind.
ZMD: That's not possible.
KP: You can't. It cannot be done. It would be really interesting. But it's not in fact possible.
ZMD: Stepping back a minute, about this team dynamic. Back in May, Scott Alexander told Curtis Yarvin that he's become more optimistic about liberal institutions because, quote,
The "vibe shift" against wokeness is as far as I can tell a genuine autochthonous liberal victory that predates anything Trump II was doing
End quote. I thought that was an interesting contrast with your self-report that you changed your Tweeting strategy after the election.
KP: I think a lot of this depends on what we mean by wokeness. I think peak fixation with identity politics was 2021 or so. I think I had already pre-election seen a bunch of people turn against a lot of stuff that I think of as peak wokeness in terms of identitarian framings of everything they do. I had already seen a lot of people going like, okay, so a lot of people in 2020 were preference falsifying, right? And going along with stuff that they didn't think was right. I think that had stopped before the election. But people mean different stuff by wokeness. It's certainly not true that all of the rollbacks of access to transition—today, Puerto Rico limited access to transition for legal adults [under age 21]. That would not have happened without Trump. If you're thinking of that as part of wokeness, then that certainly is a consequence of Trump and downstream of Trump.
If you're talking about vibes on Twitter, vibes on Twitter are dramatically different because Trump won. I think Scott has said, and I think I agree, that the vibes on Twitter are in fact liable to provoke a backlash in the other direction just because the woke right is speedrunning a lot of stuff that I thought was specific to the illiberal left until I saw it on the right. But certainly the vibes on Twitter are extremely different than in the hypothetical world where Harris won. I think the stuff at universities is very different. I think that the universities were already stepping back from a bunch of their most unpopular like DEI-type policies, but I think that it would have been a pretty limited retreat instead of a fire-tons-of-people-and-negotiate-settlements-with-the-DOJ-type retreat. My stance, which I think has been consistent for about the last eight years or whatever, on affirmative action, has been that if I happen to know of two kids with identical test scores and one of them had a life that was very neatly directed at optimizing for the things that my admission office wants to see, and the other one had a life in which there was nobody with the wherewithal or attention to optimize their life for what the admissions office wanted to see, then the second kid is a lot more impressive and admitting the second kid is a lot more valuable.
Because, two kids of identical talent, the one who everybody knew exactly which notes to hit is going to have an application that looks a lot more optimized for the results, and the second kid is not like that, means that given approximately equivalent impressiveness of application, the second kid is in fact more impressive. For a while, this was my understanding of all we were trying to do with affirmative action, find the people who did not have as many opportunities, such that you were in fact improving the quality of your class by admitting people who had fewer opportunities.
So I know people who were very smart and didn't have as impressive scores. There is a completely separate thing that you could also do and call affirmative action, which is, we would like the following balance of our student body, and so we will pick the most impressive people from this background, and the most impressive people from this background, and the most impressive people from this background, to create the student body that has the resemblance we like the best. I think the Supreme Court has said you can't do that. I think the Supreme Court was right to say you can't do that. I think that is not a good thing to do. I think the thing that is actually valuable is actually also almost impossible to do because, think about how I literally defined it. I said the people who know which set of targets to hit to impress an admissions office. If the admissions office changes which set of targets to hit to impress it, those are the people who will know how to hit the new ones, too. This is how we get, you know, Zohran Mamdani is technically African-American or whatever. But also, oh, is it helpful to have studied in a foreign country? Guess who studied in a foreign country? Is it helpful to have done a missionary trip? Guess who does that missionary trip? You cannot actually rig the system in favor of the people who didn't have dedicated adults putting in tons of optimization effort on their behalf, because rigging the system is the thing that the dedicated adults will do. So I've kind of gone, okay, it's not that I don't think of that as a fundamentally worthy endeavor. As alpha lying on the ground, frankly, if an admissions office could figure out how to do it, it would make their university stronger and better and good, and they should. But they can't figure out how to do it. Nobody is. It may be fundamentally the kind of thing that cannot be figured out how to do. We shouldn't instead just aim for a kind of demographic—I think the Supreme Court got that one right.
I think that based on the Supreme Court ruling, there was already some movement at universities to do this, but I think that that is going to be a lot more all-encompassing and involve a lot more open retreat and a lot more exposés of practices that were going on, as a consequence of Trump. So I think I more than half disagree with Scott, but I think there is a bunch of wokeness that was already on the retreat regardless. And then there is a bunch of stuff that other people mean by wokeness that either would have stayed or would have been retreated from much more incompletely and in a way that didn't require anybody to acknowledge as much that they've been doing it.
ZMD: So part of my motivation for asking is, I'm wondering from the perspective of trying to triumph in the neoliberal versus progressive conflict, it might be an advantage in the conflict, once the other guys win, you say, ah, well, we were already fighting wokeness this whole time; it was already in retreat. Even if that isn't actually true.
KP: I agree that that is totally a move to make. I also feel like it is important to say that both neoliberals and progressives have spent the last year trying to pin woke on the other, and they are both wrong. I'm not the first person to say this. I have like seen other people point this out. Woke was its own weird phenomenon, that was neither in line with traditional progressive objectives. A lot of the progressives are socialists, or at least capitalist-skeptical. But they weren't particularly identitarian before woke became a big thing. I think that identitarian stuff came out of Tumblr and academia and was embraced by nonprofits, both the more neoliberal nonprofits and tech, which is pretty neoliberal, and the progressive nonprofits. At this point, because the identitarian stuff did not make the world a better place in any way for anybody, as far as I can tell, except individual people who got jobs out of it, since the identitarian stuff was bad, then all of the factional conflict, everybody's like, well, you guys did this dumb identitarian stuff, which was bad. In some ways, I have tried not to say the progressives did the identitarian stuff, because I don't think that's true. Like I said, I think you should say true stuff. I do think the fact that everybody starting a year ago was trying to blame the identitarian stuff on their political enemies was in fact a form of wokeness already profoundly being in retreat. The fact that people were saying, you guys ruined the party with this shit. That is an important kind of wokeness being in retreat that was already true.
ZMD: Already—what month/year did you see this?
KP: I am not going to be very reliable at dating internet arguments that I saw on Twitter. I can try and dig up some Tweets for you. I do think that people were saying peak woke has already passed by 2022, depending on what you define as peak woke. They were referring to a real thing. This is not just something that was made up later to rationalize it. There was a real thing that did peak in 2020 or at latest 2022, and that people were talking about by then as on the retreat. And there's other stuff that didn't.