The New Yorker’s Head of Fact Checking on Our Post-Truth Era

0:06Hey, Fergus. Hi, Tyler. Thanks so much for being here. My pleasure. So people likely saw one of three headlines about Trump attending the U.S. Open. Either he was booed, I think that was what was reported by The Guardian. He was cheered, that was what was reported by Fox News. Or it was a mix of both. I think the New York Times went with something kind of diplomatic, like Trump was met with mostly boos at the U.S. Open. And so all of these things could technically be true at the same time. But depending on your feed, you'd only see one of these things. And then depending on your worldview, you might only accept one of these things as the actual truth.

0:41And so what do you think that this says about how we experience facts and truth today? You can have three headlines all saying a variation of the same thing in some ways contradicting each other. And yet there's also evidence supporting each and every one. I think what it says is that facts, which are something that people tend to think of as really singular and extant in the world, things that exist by themselves, are actually often a matter of interpretation or selection or focus or what you choose to look at and how you choose to look at it. People want to think of facts as something that are in themselves fair, something that are in themselves accurate or balanced.

1:23And I tend to think maybe this is something, you know, I'm kind of too pilled by having spent 10 years in the New Yorkers fact-checking department, but this is the material that we work with every day, factual material. And you can do a lot with information that is accurate, depending on what you choose to use. That's Fergus McIntosh, who leads the fact-checking department at The New Yorker. Over the past decade, questions about what counts as a fact, who decides, and how much institutions can be trusted have only grown more complicated. At the same time, Donald Trump has continued to bend data to his political needs, sometimes ignoring it, sometimes suppressing it, and more recently, trying to control it outright.

2:09I wanted to talk with Fergus about how we understand facts and data in 2025, what it takes to verify them, what it takes to check them, and how Trump's second term has reshaped the relationship between truth, information, and political power. This is The Political Scene. I'm Tyler Foggett, and I'm a senior editor at The New Yorker.

2:32So you've worked in fact-checking for a decade now, and you oversee our fact-checking department. And, you know, I guess I'm wondering, like, in that time, we have gotten so much more access to information. And given that we all have access to everything all the time, is that something that makes your job easier or does it make it harder? And how so? Okay, definitely both, obviously. So, yeah, I can pull up, like, 100 cell phone videos of Trump at the U.S. Open, more probably, with absolutely no trouble at all. and that means that sourcing can get a lot easier. So 10 years ago, I could probably have still pulled up quite a lot of videos.

3:1420 years ago, I definitely couldn't have done because, you know, unless someone took a camcorder, those videos wouldn't have been available. I might've got stuff from news networks. Okay, so that's really helpful for sourcing. That's helpful for the person writing the story, helpful for the person checking what's in it. On the other hand, it means that anyone who wants to poke holes in what you've written and what we're publishing have a much, much easier time finding ammunition to do that. So it means that I think that we have had to... The New Yorker has been very careful about facts for a long time.

3:44I think that the increasing access to information just means that the holes in the net have to get smaller and smaller and smaller to try to catch things that people might see. And like the really, really new thing might not necessarily be sort of variety of videos that you can see online, because as you said, that's been happening for years now. But I feel like as of, I guess, the last year, definitely not like the last several months, like we have been seeing a lot of AI generated videos and deep fakes, which are getting so much better. And so, you know, I guess I wonder like how you personally deal with misinformation or disinformation.

4:19Like you're someone who, you know, spends a lot of your time thinking about facts. You've gotten very good at being able to look at something and kind of determine whether or not you should immediately be skeptical of it. And what is your experience going through social media like? Well, confession to start off with, I'm not on any social media. I got up years ago. So I'm professionally on Twitter, on X, I should say, but basically not on anything else. My experience dealing with misinformation is something that I really have to caveat by saying I come at it from a position of enormous luxury in that I am paid to spend a lot of time thinking about the quality of information.

4:59And the people who I work with are paid to do the same thing. And, you know, that can get kind of maddening sometimes when you kind of can't switch off that part of your brain. But it does mean that while other people are, you know, saving lives and stocking the shelves at supermarkets and whatever, we are sitting at our laptops thinking about trying to dig into the reality of what's in front of us. So with that caveat, what this makes me think is that fact-checking is a process. It is a really iterative process of saying, first of all, what are the claims of the thing that I'm reading or watching?

5:38What is being claimed? It's normally like 10 different things at once, but, you know, say it's Trump was booed at the U.S. Open. That's a reasonably simple claim. First of all, you have to isolate that. And then you have to think, okay, where is the information coming from? um so is it from a bunch of reporting from people who were there on the ground who saw it or is it from like one video um if it's from one video that's something but it's definitely not nearly enough to make me think that we've got that we've that we have it um and that's because it might be a deep fake like you said um it might be something that deliberately avoids the context of the situation.

6:22It might be that it's easily contradicted by something that shows something a few seconds later. All that say, when I'm thinking about misinformation or disinformation, and the possibility that it might affect what we put out in the magazine, I mean, it's definitely something that keeps me up at night. But I hope, I believe that the best answer is you just keep trying, you keep digging, and you look again and again and again, and you keep asking yourself, how do I know this? Okay, I know it because this source, this video, this thing that I read told me, how does that source know what it is telling me about?

7:00What is the basis for what it says? It's really a simple question, but if you keep asking yourself that question, you can really go a long way. And sometimes you get to a point where you feel like you've hit really solid ground. Quite often you don't. And that's when it becomes an editorial process And you need to think about telling the readers with an appropriate level of confidence what it is that we think that we know. So when fact-checking something like, yeah, the claim that Trump was either booed or cheered at the U.S. Open, that's the kind of thing where we would refer to videos of Trump at the U.S.

7:33Open or maybe talk to people who were there who could talk about what they heard in real time. But there are plenty of times where we're checking something where like we're referring to, you know, it's like a number, like this number of people crossed the U.S. border illegally last year. And that's the kind of thing where we would go to a, you know, we would use government data. You know, you're leaning on reporting from other publications. And so I guess I'm just wondering, like, in this kind of changing information environment, what has the gold standard been for data in the past and for finding truth in the past?

8:10And how many of those sources are still trustworthy today and I guess would you say are still, like, uncorrupted? Well, as a start, I would say that you shouldn't ever assume that any data is totally uncorrupted. I think that any data is potentially flawed or skewed. In fact, I think that probably all data is necessarily flawed and skewed in some way. How so? Simply because it has been selected. You know, when we have a data set, we don't have like an exact replica of the world in front of us. We have a snapshot, a series of snapshots, a series of impressions, maybe a huge number of series of impressions of the world.

8:51Those snapshots, those impressions, those data points have been taken in order to try to get an answer to a question normally that somebody has. So if it's a scientific study, it might be that the question is, is there a link between smoking and lung cancer, for example? And that will take many, many years to get an answer to. If it is a question about how many people are crossing the southern border, I mean, that's actually a really, really hard question to answer. That's something that colleagues of mine in the fact-checking department have really had a hard time grappling with because it involves dealing with a bunch of different data sets which don't really cohere with each other and which kind of seem to contradict with each other.

9:37And it also depends on, like, what does that question actually mean? okay, if you say how many people are crossing the southern border, there is a very kind of brute version of that question, which is literally how many individuals on a given day cross that line. But then you start to, it very quickly starts to get more complicated. I'm thinking about a piece that we ran a couple of years ago by John Blitzer, which dealt with exactly this question. And one of the real head scratches was, do we or do we not include in that count people who have crossed before and have been deported and then being readmitted?

10:17Also, do we count, are we counting people who are apprehended versus people who are not apprehended and estimated? It's like the difference between someone who was like coming over and trying to hide within the U.S. versus someone who was coming over and claiming asylum, essentially, and trying to go through that process. Right. Do we count people who are doing what? At least tens of thousands. You know, there's the fact. I was about to say millions, but I'm sure that's wrong. But many, many people do every day, which is commute over the border for work. Yeah. You know, presumably we're not because we are not asking that kind of a question.

10:54But you start to see that any given question that you ask has often multiple implications. And then when you're trying to get to the point of answering with data, that data itself has baked in a point of view. So you'll have a data set that tells you some section of the number of people who literally cross over the border on a given day. But that has baked into it a lot of different assumptions. And also it might have baked into it a fair amount of guesswork or estimation. So there's a problem of dealing with like active misinformation. And then there's also the problem of trying to check something that is just difficult to check because you're dealing with varying sources that, you know, are giving you a different picture and that themselves might be politicized.

11:40Like an example that immediately comes to mind is the question of how many people have been killed in Gaza. That that is not, even though it's a number that we're looking for, it is not at all a straightforward thing to check. No, I mean, I think that that is, obviously, everyone gets uncomfortable talking about this specific question of how many people have been killed in Gaza, because it is such an inflammatory topic for basically everyone. early in the war, I think there was a lot of discussion about whose numbers should we be reporting on. And there seems to have been a pretty general, as far as I encounter, drift towards taking the numbers from the Hamas-led, or Hamas-affiliated at least, Gaza Ministry of Health.

12:27And that's Hamas-led in the sense that Hamas is the government, or was the government in Gaza. So the Ministry of Health naturally is, in some sense, affiliated with Hamas. The Times uses those numbers. We use those numbers. The Times, I often, but I think not always, see them saying, you know, X number of people have now been killed in Gaza, according to the Ministry of Health, which is affiliated with Hamas, and which does not distinguish between civilians and combatants in giving its number. So that's one way to deal with this sort of problem. You really lay it all on the table for the reader.

13:07You say, this is where we are getting this number from. This is the political affiliation of this source. And this is a key bit of information about the methodology, which goes into this number. I think that that is often very much the right thing to do. It's not enough. You still have to, you really owe your readers the work of at least figuring out, should we be using this number at all? Should we be using this source at all? Because you could follow that procedure and say, this is where it's coming from. This is its methodology and this is its affiliations. And you would still be, you could still be amplifying something that's total nonsense, that has absolutely no confirmation to reality at all.

13:59So, yeah, you have to do your best to at least vet what you are handling. In the case of the number of casualties in Gaza, it's very, very hard to do any independent vetting because there have been no outside journalists allowed into the territory throughout the war. So that's when you end up in a situation where, you know, best you can do after a certain point is to say this is where the number is coming from this is the methodology this is the affiliations and so on and you look out for inconsistencies and you try to talk to people who tell you i mean i remember early on in the war we were talking to people who would give us a sense of how these numbers are being compiled where they're coming from give us some sense of reassurance that they are not conjured out of thin air we got to that point.

14:51Clearly, the Times got to that point. But we still try to, and the Times still tries to cite the source. So that's how we would go about trying to vet like a sort of a difficult figure. But I wonder about the pieces that hedge on kind of larger and more abstract claims, maybe even statements of opinion that kind of feel like fact. So, you know, on this podcast, for example, Andrew Marantz and I have had a series of conversations about the state of our democracy and whether we are heading toward authoritarianism. And so if someone were to write an article in which they claim that democracy is on the decline in the U.S., I think Andrew literally wrote that article, although he used a bunch of sources to back it up.

15:37I guess like if you were to if someone were to write like an opinion, I guess if someone were to write a piece that basically said, you know, democracy is dying in the US. Is there even really a way to fact check that or to verify that? Would you treat that as a statement of opinion or would you, you know, kind of give the writer the counter arguments for why democracy is fine? Like how would you deal with something like that? OK, so you asked, is there a way to fact check that? And you asked, is there a way to verify that? I would say that the answers to those two questions are yes and no. I think that you can't verify that because it's not actually a claim that is verifiable because it inherently includes a certain amount of opinion.

16:19You know, democracy is a, you know, to get kind of stupid about it, democracy as a concept, it doesn't die, right? So there's a built-in metaphor that you can't actually verify that. But that tells you something, right? It tells you that you are dealing with something rhetorical. Okay, so you're dealing with something rhetorical. Does that mean that you can't fact check it, that you just throw up your hands and say, oh, just the writer's opinion? No, definitely not. I think that this should be perfectly obvious to anyone who's ever had an argument with anyone about anything. But opinions are based on information, on ideas, on prejudices, on all sorts of things.

17:00and those things are susceptible to what is included under the general banner of fact-checking. So first of all, you know, is democracy dying in the U.S.? Well, you might look at things such as voter turnout in elections. You might look at things such as difference between the partisan vote and how that plays out in the number of seats that a certain party gets. Those are things that are quantifiable, that are measurable. Those are things that you can actually like provide some data for, or hopefully you can provide some data for if you can find some good sources for it. And if you, you know, you might have a writer who said that and you come to them and say, well, actually, you know, being the nitpicker, pedantic, fact checker, there's all of these data points which seem to suggest that democracy in a certain sense is thriving.

17:50Okay, so you could bring them that. They might still say to you, that's not how I meant it. So then you do need to make sure that you are talking about the same thing, right? Beyond data, then there are the impressions of experts in the field that we could get into. So if you have a series of very prominent scholars who are willing to talk to you, they tend to be aware of the counter arguments because they have to defend their own work against them. So if they are a good, worthwhile scholar, someone who you should be talking to, who it's worth you talking to, then yeah, hopefully they'll tell you about that.

18:28And then something that I think is really important for fact checkers to think about, and honestly, again, anyone who's ever had an argument with anyone about anything, is to think, what is the person on the other side of this argument trying to convince me of? What are they going to be saying? What are the arguments that they'll be using? What are the holes that they'll poke in my argument? This is something where the fact checker is acting like almost a sort of proxy for the reader, maybe a proxy for a hostile reader saying, this is where I would attack your argument if I could. It's a sort of peer review process.

19:01So that's definitely something that we can get to in fact checking. It's sort of does your logic work? Are you ignoring something? Are you being really prejudiced? Are you rolling your eyes when I present you with a counter argument? You know, doing that is worth it because it means that we don't have obvious mistakes. But it's also really worth it because it proves and tempers our work and makes it stronger. When we come back from the break, we're going to talk about Trump's relationship with facts. And we're also going to talk about what Fergus has described as the latest phase in Trump's war on data, which is the headline of a recent piece that he wrote about Trump's firing of the Bureau of Labor Statistics commissioner in August.

19:39More from the political scene from The New Yorker in just a moment.

20:14Thank you. Vogue, Chloe Mal, and head of editorial content at British Vogue, Chomanati, as they explore style and culture through the lens of fashion with guests like Martha Stewart, Kamala Harris, and Tracee Ellis Ross. The Run Through with Vogue, new episodes every Tuesday and Thursday, wherever you get your podcasts.

20:46So, Fergus, you recently wrote that when it comes to Trump, facts shift depending on Trump's personalist view of the world and on his rhetorical needs. And you also wrote that Trump's fluid relationship to facts doesn't mean that they don't matter to him. They matter a great deal so long as they say what he wants, and if they don't, he's happy to ask for new ones. This is definitely true. And so, you know, the occasion for you writing this piece was that Trump had just fired the Bureau of Labor Statistics Commissioner after he claimed without evidence that a weak jobs report was rigged. And so he got rid of the commissioner and, you know, vowed to replace her with someone who would be, you know, better and I guess just less politically in the tank for Biden since that was kind of the implication of what Trump was saying.

21:29Allegedly. And yeah, exactly. Allegedly. Without evidence, he was just making these claims. So we've seen Trump fire a lot of people for disagreeing with him over the years. And, you know, Nixon and other presidents have gone after jobs reports in the past. And so why do you think that this particular dismissal was so alarming? Like kind of what did it signal to you in terms of kind of the current state of Trump's relationship with facts? Because I feel like it went from Trump dismissing facts to Trump kind of just suggesting that he was going to create new ones that worked for him. And we haven't necessarily seen that, and we can talk a little bit about that later.

22:04But, like, why was that firing a big deal for people who don't care about the Bureau of Labor Statistics? Guilty. I was not previously a huge devotee of the Bureau of Labor Statistics. But it definitely caught my eye, as it did many other people. Why was it so important? Why did it seem so striking to people? I think it is partly because of the actually, it turns out, really, really crucial nature of the information that is put out by the BLS. The jobs report being one of their big products. It is one of the prime indicators of the health of the U.S. economy. It's one of the things that the Federal Reserve is really looking at when they decide whether or not to lower interest rates, which is something that Trump wants them to do.

22:53So partly it's that, it's that it's really important. But he's fired important people in the past, for sure. I think what it really exemplified is something that is new in the second term under Trump, which is that in his first term, he was obviously in control of his cabinet. It was kind of a revolving door, people in and out all the time. The federal bureaucracy was kind of mostly left alone. They kept on operating in much the same way that they had before, in the way that Steve Bannon would call the deep state. In the second term, as has been written about kind of ad nauseum and really well demonstrated, Trump came in with a very well formulated plan for how to deal with that very strong federal bureaucracy that exists, that has a lot of independence, that has a huge number of rules around how it operates, that are in many cases supposed to insulate it from political interference.

23:52And this firing of Erica McIntuffer, the BLS commissioner, was a just super high profile example of Trump cutting straight through that independence, that insulation that is in place, that is supposed to make it so that the data being produced by the federal government and the decisions that it is making are related, at least, to reality. which is not to say that they are not coming from a politically partisan point of view that there is a particular party in control of the presidency at any given time. And the bureaucracy has a lot of politics kind of built into it, but there's at least the appearance.

24:34And I think in many ways, the actual, the actuality of independence. Of course, you know, in the weeks since then, there's been the firing of the CDC director for kind of similar reasons. I mean, there's dispute around why she was fired. It wasn't quite as explicit as Trump, who just said, I'm firing the BLS commissioner because I don't like the numbers. but the CDC director wrote a piece in the Wall Street Journal saying that she had been fired because she had refused to rubber stamp the recommendations of a vaccine safety panel that had been stacked with people who were handpicked by RFK you know that's another example of the same kind of thing happening that is you know just as alarming to me it's a bit muddier it's a bit less obvious and it's um it's not the president doing it himself it's not Trump doing what he often does, which is just really saying the quiet part out loud, having no pretense of hiding behind norms and previous standards.

25:36Yeah, let's talk a little bit about RFK Jr. because I feel like he presents both the public and journalists with a kind of interesting quandary, which is that, you know, I feel like in the first Trump administration, a lot of what journalists did was that they would take some baseless claim from Trump and then show that it was wrong by putting up data that pointed to the contrary. But now we're in the situation where we are more often than not having to figure out the difference between good data and bad data. So you have RFK Jr. who will point to certain scientific studies that then other people in the scientific community will say are just total BS.

26:11And, you know, that there are these other studies that, you know, contradict those studies. And it just kind of makes me wonder how normal people are supposed to figure out which side to believe. I mean, obviously, if we were doing a piece on something that RFK Jr. has said. I mean, we have done a bunch of pieces on him. You know, we spend a lot of time looking at like the sample size in a study and like the conditions under which, you know, and what journal published it and was it peer reviewed? But like the normal American just like watching TV isn't going to do that. And so, yeah, I mean, do you think that in some ways that presents just like an even harder problem than what we've seen from other members of the Trump administration in the past, which is just kind of like, oh, we're just kind of doing this because we want to and we're not even going to bother to try to back it up with something.

26:54I do think that presents a really, really difficult problem. I mean, you know, you just gestured to one of the things that I think makes the media essential in, you know, I guess kind of any complex society, certainly in the society that we live in now, just there's so much to think about at any one given time. You're trying to figure out what actually are the figures around how many people died during COVID-19, for example, how many people died during the pandemic. Yeah, there are a bunch of studies about it. Some of them are high quality, some of them are low quality. Figuring that out is really hard.