< humor > If we don't think your post said what its title self-evidently means (to us), does that give us a license to accuse you of lying? < / humor >
As with newspapers, the headlines here aren't chosen by Scott, they're instead decided by editors who may have very different vibes intended than the actual content.
...would have been the funny addendum to >Please don’t have opinions based on the titles until you’ve read the posts!
I really hope that headlines for articles posted here are chosen by their authors, rather than anyone else. I particularly hope they aren't chosen by "editors" with goals unlikely to match those of the authors, never mind the readers.
There are more than enough sources of click bait headings out there already. The medium term (months) result is that people remember "links to this site are unreliable" and stop visiting the site. Scott really doesn't need editors driving his potential readers away.
It was intended as a joke, though I do notice Scott tends to have an unusually high proportion of sub-optimally titled posts relative to other Substackers. Others it's impossible to imagine anyone else doing the titling, because the titles are so...on brand. FdB's post titles are always just as rude as the content, so I can tell he writes those. (Said with approval, does-what-it-says-on-the-tin.)
Sometimes I think Scott just likes to/unintentionally subconsciously bury the lede a lot. Like this threepeat series where The Moral Of The Story is the difficulties of setting censorship bright lines. But that's not remotely obvious from any of the titles, and one must actually read the posts in full with a bit of attention to grasp the thesis <-> supporting arguments distinction. Thousands of comments debating the supporting arguments, not many grappling with the obvious thesis. Maybe because it's too banal, what with rationalists generally being opposed to censorship/"mis-disinformation" already? So we fight about what's more interesting instead.
("You can't blame Scott for writing these articles, he's just throwing red meat to his subscriber base...")
Honestly, the fact that so many people are having trouble grokking Scott's actual thesis is 100% his own fault. These posts remind me a lot of the frustration I get with Ayn Rand's use of the word "selfish" (or philosophers' habit more generally of taking ordinary words and giving them weird, idiosyncratic definitions that the reader has to dedicate half their mental energy to keeping straight): sure, I get what he's trying to say, but I don't understand the doubling and tripling down on a rhetorical strategy that clearly isn't working.
If "the virtue of selfishness" really means "there is nothing inherently moral about self-sacrifice; don't feel ashamed for living your life according to your own values," and you're not telling everyone to be anti-social pricks who lie, cheat, and steal their way through life, maybe it's counterproductive to keep using the phrase everyone assumes means the latter, and the time you're spending defending your slogan could be better used making actual, substantive arguments. (Although I get the impression Rand was being deliberately inflammatory for the attention, because she was at least as much a public figure/celebrity as she was a philosopher).
If I were to be similarly uncharitable to Scott, it almost seems like he's using deliberately inflammatory headlines to boost engagement. I'm sure he's not, as that doesn't jive with everything else I know about him from reading his blog over the years, but it's frustrating that he won't just admit that "yeah, it wasn't the best title. My definition of the word 'lie' is unusually narrow, and I probably should not have started a blog post intended to convince my progressive friends that it's not as easy to distinguish the kind of lying Infowars does from the kind of lying the NYT does as they think it is, with a headline implying you should be ~more~ credulous of the media."
I understand what he's saying, but he insists on saying it in a way that's almost guaranteed to confuse people.
I agree that these kinds of flare-ups are unseemly and unnecessary; it's sad when a blogger and their commentariat get into petty split-hair fights, and both sides have too much pride to admit they fucked up or went too far. Somewhat excusable for pundit-type bloggers whose true profession is Take-Slinger, less so for professed Rationalists. Principle of Charity doesn't extend to cover bullheadedness.
It's also still weird to me because...I use a similar truth taxonomy as Scott? A really shocking amount of Obvious Nonsense is, in fact, true[1]; it's challenging to litigate intent and ignorance if we aren't clear on whether we're judging what-the-text-actually-says, versus what-the-typical-takeaway-is. Not just in a legal-censorship way, but it's where so much of the difficulty with Bounded Distrust[2] comes from. It's like the Lizardman Constant: one must assume that everyone, of all persuasions, is trying to corral inconveniently-flexible facts into a narrative...that's the background level of noise in any news communication. One's priors will be weighted differently for favoured speakers, ingroups vs outgroups, certain topics and not others...but you still gotta Do The Math in each case, or the whole exercise falls apart.
Plus I think it shouldn't be as confusing for longtime readers, since Scott's been making more or less these same nitpicky points for years now. The old SSC posts "Against Lie Inflation" and the "Lies, Damned Lies, And Facebook" series, for example. I sometimes think there's been a bit too much Rally-Round-The-Schelling-Flag effect wrt NYT, so even though there's a very large deserved amount of incredulity, sometimes tribal defensiveness leads to excessive...I wouldn't say "paranoia" like Scott did, that's too combative. But like, confusion of the narrative-news ecosystem we actually have, and a Platonic Ideal of News Media which somehow turns a profit via telling The Truth, Only The Truth, And Nothing But The Truth. Businesses sell what people are willing to buy, and it's not unvarnished distortion-free facts.
I don't know. I just feel there is some real and important inferential distance between unreality-as-stated, and creating-the-illusion-of-reality-via-sophistry-and-misdirection. To me, the former is a much more egregious epistemic sin than advertising an inaccurate map. Scott didn't do a great job making that point clear in these posts, but it's an important point nonetheless. How does the apocryphal Soviet saying go..."Pravda is useful because it always lies, the NYT is less useful because it sometimes tells the truth"? I think MSM skeptics would rather it be Pravda all the way down. Sadly, things are not so convenient - so we must be *more* vigilant, not less. (I do fully agree the headline is a dud, misleading at best. Hence my joke. Scott doubling down by saying we didn't read the subtitle is...well...silly.)
[2] I liked Zvi's attempt at formalizing this solving-for-the-equilibrium. Specifically, if the obfuscations and distortions are the weak cliche ones you'd expect - that's simply the best they could afford: https://thezvi.substack.com/p/how-to-bounded-distrust
Great points. The annoying thing is that I also think I use a similar truth taxonomy as Scott. Zvi's method is actually remarkably close to how I already approach the news. I don't think I actually disagree with much of anything Scott's said in any of the three posts on this topic he's made. If I really zero in on what about the posts bugged me, it's that Scott didn't use sufficiently critical words to describe people he fully admits are doing bad things. It's kind of a strange and petty criticism when I lay it out like that, but it's interesting to see in myself (and many other commenters) just how much "I agree with you who the bad people are. Your feelings are valid" is something I'm looking for here. (I think this also explains some of the "you did not just draw an equivalence between the NYT and Infowars" people).
Scott may have accidentally stumbled across this community's scissor statement. The original post managed to call everyone dishonest without calling anyone a liar, which managed to piss off just about everyone, left, right, and center.
I think that I consider it qualitatively different to employ a word in a narrow, idiosyncratic way when you're trying to expand that definition beyond the original scope, rather than fighting to maintain a semblance of the original boundaries. It's very difficult, nowadays, to construct any argument for preserving the utility of certain words without meeting opposition that dismisses the effort as naïve (i.e. you can't fight the colloquial erosion of specificity, so why even bother?)
The major problem I have with this position is that it seems to completely miss the point. Influencing the colloquial evolution of this language is exactly what any thought-leader is trying to do when writing a post like this. Accusing them of needlessly discomfiting people is, in a backwards sort of way, acknowledging their success.
After all, this sort of discussion is a small part of how the vastly complex memetic landscape is shaped.
"maybe it's counterproductive to keep using the phrase everyone assumes means the latter"
I agree with this in principle, but sometimes it threatens to become arbitrarily hard to have a conversation about topics *without* trying to redefine a word.
Take "capitalism", which I've observed a lot of left-leaning folks to use as a synonym for "greed" (sometimes as "systemic greed"). I get why they're doing it, but there's a perfectly good existing word for it ("greed"), and I'd wish they'd back off from trying to give a different perfectly good existing word a permanent synonymous connotation. It makes conversation difficult.
Arguably, this is exactly what you're saying - we should avoid redefining words. But in this case, redefining it worked (for that tribe), and it bricked a perfectly useful word in that sphere of communication.
Which is to say a few things: (1) I think it's hard to guess when attempting to redefine a word will work and when it won't, (2) I think it can sometimes be a good idea to try and redefine words even if it seems likely to fail. (Although in my example, I think people have managed to partially evade the issue by using the term "free markets" instead.)
Re: Scott, I previously commented that (in these circles, at least) I think it's mostly a question of which definition of "lie" is useful for your model of the world. I share Scott's definition of "lie" and it's been useful to me and my interaction with people to have that distinction in my head, but I can easily imagine situations where that isn't true.
Do you really think that when the average person hears "the media is lying to you", they interpret it to mean "the media is presenting you with information that is technically accurate, to the best of their ability, but is leaving out important context in a biased manner"? I would predict with very high confidence that this is false, and that Scott's idea of what "the media lying" means is the normal one. I could *maybe* believe that people who talk about the NYT "lying" are usually applying a broader standard than Scott is here (although even then I'd place roughly 80/20 odds against), but I'm almost 100% confident that this isn't true in the context of "alternative" media like InfoWars.
Scott's idea of lying isn't even *Scott's own idea of lying*. At the end of this very post, he endorses sometimes calling "#6" "lying", and yet that directly contradicts the definition he bases the rest of his argument on.
I guess my issue with this series of posts is that the headline initially sounds like an interesting, debatable topic, but then the word "lie" is used in such a narrow sense the conclusion is meaningless. A post titled "The Media Very Rarely Completely Invents Facts Out Of a Whole Cloth With Zero Relationship Whatsoever With the Truth" would be a pretty boring and pointless post, and that seems to be the argument being made here. By this definition of "lie", even people like Trump or Andrew Tate or whoever are not lying, because they can point to some underlying shred of fact or say it's technically an opinion or they are passing on something they heard. That might be true, but who cares? If you define any term that narrowly it won't cover much.
Agreed. Scott's definition of lie is very narrow. He sees to be claiming that if an article offers even a smidgeon of support for an assertion, no matter how weak or irrelevant, then people shouldn’t call it a lie. But this clearly leads to absurd results. By this standard, even an article that says something like 'John is a murderer' and gives as its only evidence for this claim 'John seems like the kind of person who would be a murderer' would not be classified as a lie.
There's a point where an argument is so incredibly bad and misleading that it's equivalent to a lie.
But the whole point of this is that there are a lot of ways to convey an incorrect understanding of the world that aren't overt lies. And this is important because it tells you how to read the New York Times. If you know basically what narrative they're looking for, and you know that they will not lie but will also maybe leave out stuff or shove stuff to the end of the story to avoid messing up the narrative, this makes the NYT's reporting more valuable.
For example, consider: "Oh, wow, there was a really serious shooting but the article never mentions the race of the criminal and never shows any pictures even though he's in custody. I guess he's black." This wouldn't be a sensible inference except that most major US newspapers go out of their way not to mention the race of criminals or show their faces when they are black, apparently as an overt policy of trying to avoid supporting stereotypes.
Reading a story about a hate crime against a Jewish or Asian grandma in a big city where the race of the attacker is not mentioned, similarly, well, you can infer the race of the attacker by knowing the narrative the newspaper prefers. If the perp was a MAGA-hatted white guy, there would be a dozen pictures of him in his MAGA hat. If the perp was a crazy black homeless guy who beat up the grandma because the little voices told him to, the reporter would just accidentally somehow not ever mention his race and it would turn out that there was no room for a mugshot.
"If the perp was a crazy black homeless guy who beat up the grandma because the little voices told him to"
Speaking of ways to convey an incorrect understanding of the world:
In the US (with presumably similar numbers elsewhere), people with mental illnesses are FAR more likely (like an order of magnitude) to be the victim, rather than the perpetrator, of violent crime.
Iirc, they also make up a smaller percentage of perpetrators of violent crime than they do of the general population, ie mentally ill people are "underrepresented" among violent crime perps.
You could have easily given your example of race reporting in violent crime without bringing the issue of mental illness into it.
You recall incorrectly. People with schizophrenia are considerably (4 to 6x says the first study when Googling, and corroborated as approximately correct by subsequent studies.) more likely to commit violent crimes than the general population. The fact that they're more likely to be victimized by it than commit it is a sad, but irrelevant fact.
Of particular predictive power for violence is, in addition to psychosis, substance abuse and command hallucinations. So if you encounter a crazy black homeless guy who has voices telling him what to do, he's much more likely to attack you than the average, random person. Though most people, regardless of all those factors, are nonviolent.
Having an explicitly irrational actor as the hypothetical illustrates the point more effectively, if you ask me.
Also, if it matters to you, I write this comment with a person diagnosed with a psychotic illness lying in bed next to me. You don't need to ignore reality to have compassion.
I could have also given the example without incorporating race or homelessness, which work similarly--blacks and homeless people are overrepresented among both criminals and their victims. The whole point of my post is that media sources will omit a lot of possibly-relevant information (like the fact that the dude who beat up a Chinese grandma while screaming ethnic slurs was a crazy black homeless guy) in order to either support their preferred narrative, or to weaken some narratives that support stereotypes like that a lot of crime is committed by blacks, homeless people, and (some kinds of) crazy people.
But what I'm looking for in an information source isn't someone to sculpt my mental landscape in a socially-beneficial way or to fight back against harmful stereotypes by omitting relevant information in their stories. Instead, I'm looking for sources that will accurately inform me about the world. So if the Chinese grandma was bashed by a MAGA hat wearing Trump supporter, I want to know that, but if she was bashed by a black homeless dude who was in the middle of some kind of psychotic episode, I also want to know that.
I do not think completely missing the obvious subtitle in the first post and then ignoring the explicit mention of it in this article is more absurd than people reading a text that constantly repeats in slightly different ways "The media will very rarely lie in the sense of explicitly saying falsehoods, instead, they will convey misinformation by proxy, through quoting biased sources or signal-boosting some evidence while ignoring other" and then linking to a news post of the media doing just that, and claiming it somehow disproves the central point.
If people are lacking enough reading comprehesion to comment such things, I don't think it is unreasonable to imagine they also didn't see the obvious subtitle.
> By this standard, even an article that says something like 'John is a murderer' and gives as its only evidence for this claim 'John seems like the kind of person who would be a murderer' would not be classified as a lie.
> There's a point where an argument is so incredibly bad and misleading that it's equivalent to a lie.
Perhaps, but the standard you describe doesn't reach that point, or come close.
The problem with your example is that it's not misleading at all; when you say "John seems like the kind of person who would be a murderer, so he must actually be one", no one is going to believe you.
(1): strong affection for another arising out of kinship or personal ties
maternal love for a child
(2): attraction based on sexual desire : affection and tenderness felt by lovers
After all these years, they are still very much in love.
(3): affection based on admiration, benevolence, or common interests
love for his old schoolmates
If I were to say, and insistently repeat, "Parents should not express love for their children; it's not good for the children and if at all common is a sign of social degeneracy", I will probably cause a great deal of confusion and get more than a few people angry even though I am pedantically correct using *a* standard definition of "love" (#2).
If a thing is clearly true for one "standard definition" of a word, and clearly false for another standard definition of the same world, and clearly contentious for many people, then it might be best to chose a different word or phrase.
I agree that Scott's definition of a lie is too narrow. Experienced liars agree with Flashman: Suppressio veri is a useful servant, but Suggestio Falsi is a perilous master. Experienced liars are still knowingly telling lies of omission. But mass media is by definition a choke point, unlike social media. No news story can include everything. JD Bernal went so far as saying mass media is always censorship.
So we should not expect more from any mass media than we expect from advocates in court. If I'm on trial I don't want my defender to include every possible way I'm guilty. If I'm a D party loyalist I don't want the New York Times to include President Pedo Peter's bribes from Ukraine or anyone else, much less track down the underage naked girl Hunter tied up and photographed on his laptop to check if she's still alive. If I'm an R party loyalist I don't want Fox calling Arizona for D party before they have to, much less point out every last Trump stretcher.
Lying has a simple and precise definition: "Making an assertion which one does not think is true."
By this definition, the media rarely lies. It often deceives, however, and I would argue that that is sort of worse, because deception (lying using truths) is trickier to refute.
Trump espouses outright falsehoods so often and so strategically that I think it is more than reasonable to accuse him of lying. Now Andrew Tate may, in fact, be so stupid that he believes his claims—I genuinely can't tell. In any case, an assertion does not need to rise to the level of a lie to be harmful. In a certain sense, true believers in falsehoods are more dangerous than liars, because they spread untruth with a relatively clear conscience.
Wouldn't that precise definition be, "Making an assertion which one believes is not true"?
I ask this because you brought up the example of Trump. He generally seems to be simply indifferent about whether his factual assertions are true or not. If the assertion helped him pursue his objectives then it was worth saying, and if it didn't then it wasn't. That seems to be his only criteria.
Does _indifference_ to truthfulness count as lying? Seems like it should....but maybe that is actually a different category.
I always thought that the above book, due to its rather crude title, was on the intellectual level of those cringy self help books like "The Subtle Art of Not Giving a F*CK!!!"
But just at a glance, I think that Frankfurt's category of "bullshit" tracks exactly with what you're saying. The book seems pretty relevant to the whole discussion, actually.
The media doesn't lie. It either deceives, or more often simply pushes bullshit.
That's sort of isomorphic to John's "making an assertion which one does not think is true", which is precisely what Paul aimed to correct with "...one believes is NOT true."
I'm a software engineer. There's a usually a deep cultural rift, to put it mildly, between the engineering and the sales sides of every software company. The two groups don't understand each other and have quite visceral distaste for the other.
Trump is a salesman, period. And that's one of the many things, perhaps the main thing, that people who can't stand Trump can't stand about him. The fact that Trump, who was the President of the United Goddamn States, and therefore responsible in at least some way for what happened in the lives of ~300 million Americans, if not in the lives of ~7 billion humans on Planet Earth due to America's outsized influence in the world, just didn't give a shit if the things he said were true or not is beyond infuriating to the "engineering" mindset.
People for whom "true" and "false" are not important categories set my teeth on edge. And that's basically all PR people and most politicians, but Trump dials it up to 11.
Tied to the "salesman" concept is another important truth category that seems to be neglected here: statements that the speaker hopes will become true as a result of saying them. Self-fulfilling prophecies.
As someone in the finance world, the most obvious one to me is this: a bank CEO says "Our bank's financial position is fundamentally secure." The vast majority of the time bank CEOs say this, and the vast majority of the time it is correct.
Now, let's imagine one of those same CEOs were to instead say, "Our bank's financial position is highly dubious. I don't know if we'll survive another day. If you're exposed to any sort of counterparty risk or other financial risk from our bank's collapse, then you're about to get completely screwed over, sorry."
That statement, too, would very likely be correct as a very result of the bank CEO saying it.
CEOs at places like Lehman and Bear Sterns were accused of lying when they said their firms were secure, even though they were actually on the verge of collapse. Which, while I don't exactly have sympathy for those guys, I also don't think is entirely fair. They said the only thing they could say, out of HOPE that it would be true and knowing that if they didn't say it, they were certain to fail.
Back to a more proper salesman: imagine Steve Jobs. "This product is insanely great." People said he had a "reality distortion field". At least part of what made his products insanely great, they'd say, is the strongness of his assertions. Maybe, if he hadn't made such strong assertions about Apple's greatness, people would actually have rated their products lower. Maybe they would have derived less enjoyment from them, fewer endorphins released.
This is even clearer when you go beyond consumer electronics (which have both a fashion and function component) into pure fashion. There it's all just salesmanship, really. That dress, those blue jeans are the bee's knees because we have salesman'd our way into convincing you they're the bee's knees. If we hadn't convinced you, then we'd be liars.
It's intriguing how Peale was influenced by Christian Science, which was itself influenced by Hegel. In other words, there is a direct line of ideological descent from Hegel to Trump. That prussian is truly at the root of all modern philosophy and thus, all modern errors.
It's not like "Mexico" was/is actually "sending"¹ anyone, so yeah, clearly true!
And I don't think that's particularly odd: for known-to-be-false statements, they can point at the facts and say "see, this statement is false".
For the disliked true statements, sometimes the only thing his opponents could say was some sort of "hey, don't you know you're not supposed to say that out loud & in public?"
...which seems a lot more upsetting and frustrating as a position.
¹In the sense Trump meant here. Presumably the people Mexico actually *does* send to the US (eg their ambassador and embassy staff) are in a sufficiently high percentile for at least some relevant dimensions that they could reasonably be called "their best" (if not MIB "the best of the best of the best").
The implication here is that nations are no more than their governments, which I'd disagree with. There was still such a thing as Italy when Dante was writing (in fact, it's a word that he uses). Of course, governments can help forge nations -- there is probably more of an Italy today, some 150 years after unification, than in Dante's day. And if Italy were to split into separate states again, or be subsumed into a European superstate, there would probably be less of an Italy 150 years after that.
Mexico as a nation has been sending large numbers of people northward for decades, which the government in Mexico City has regarded with approval (mainly because of the economic boost provided by remittances) but hasn't done much to directly facilitate.
On the one hand, I have an engineering degree, and have a history of getting pushback from people above me because I insist on things being technically correct. I've had bosses complain when I've provided something that precisely met their specifications (but not what they actually wanted) and teachers punish me because I would insist on correcting their grade-appropriate but not technically true answers.
On the other hand, I grew up in the DC suburbs. News of politicians and political appointees say things which are not technically true is background noise here. Trump was not an outlier (or out-liar) in this respect. Our current president has a long history of telling fantasy stories, and if he's not deliberately lying when he does it now, the alternative is significantly scarier. We have "You can keep your health insurance", "I did not have sexual relations with that woman" and "I am not a crook!" from past presidents, plus countless examples from lesser figures.
The problem is that Trump spoke outside the overton window. I don't the window actually has a veracity condition, it is just about social acceptability. So Biden's lies are largely banal and within historical bounds. It isn't the lying, it is the social transgression.
Indifference to the truth is a category that Scott completely overlooked in his taxonomy at the end of the article, and it's really important. A huge amount of false information comes from this category, because for most people in most situations the truth is besides the point.
One thing that should concern us is that the cost of generating truth-indifferent information is dropping close to zero thanks to ChatGPT and other systems like it, because they are, by their nature, indifferent to the truth. They have no idea what's true or false, but they know how to generate text that sounds convincing. Expect the future to include dramatically more bullshit.
For more details on these ideas, listen to Ezra Klein's podcast from 5 days ago: A Skeptical Take on the A.I Revolution.
Agree – a good way to understand what is going on with a lot of media articles is to forget about true and false, and look instead at what the author feels, and what they want the reader to feel. That's a whole extra layer of analysis that Scott is neglecting in these posts.
I believe most of Trump's "lies" are hyperbole, which often bypasses detection in the same way sarcasm can, and thus is perceived to be a lie. Yes, there are cases when he seems to have lied, such as when he claimed during one of the debates that insulin could, in the future, be as cheap as water. But when he claimed, for example, that COVID was totally under control, that wasn't a lie, I believe, but the normal stuff said by all politicians.
I guess the point is that Trump's claims are not just misleading (which is typical for politicians), but positively false to a degree which is unprecedented in American politics. See here:
It has a lot of well-collated information. Some comments:
"In his 2016–2020 financial reports, he claimed that the Trump Hotel in the Old Post Office Building in Washington, D.C. had revenue of over $150 million. In 2021, the House Oversight and Reform Committee revealed that, to the contrary, the property had a net loss of $70 million during that period." I checked the source on this, and yes, that is all it says. Assuming the quote is entirely true, there is nothing that is a lie in it, if it also had $220 million in expenses.
"During a 2018 interview, television personality Billy Bush recounted a conversation he'd had with Trump years earlier in which he refuted Trump's repeated false claims that The Apprentice was the top-rated television program in America. Bush recalled Trump responding, 'Billy, look, you just tell them and they believe it. That's it: you just tell them and they believe. They just do.'" This sounds very typical Trump. Basically, when Trump says something, one should interpret it as if a politician said it, but more so.
"When Schwartz wrote The Art of the Deal, he created the phrase 'truthful hyperbole' as an 'artful euphemism' to describe Trump's 'loose relationship with the truth.'" Again, it takes a special interpretation. Once one gets used to speaking Trump, which ought to take perhaps five minutes, one can see at least what he's trying to say.
But don't expect promises to be kept the way they are stated. If he says he guarantees something, he means he emphasizes its importance. If he says something will make money, even something like "huge amounts of money, tons of cash", it doesn't mean it will be profitable. If that's lying in your book, then so be it, and I won't argue that.
Obama promised "hope and change" without specifying whether the change would be good or bad. Here in Michigan, Jennifer Granholm promised we would be "blown away" by her second term if she were reelected. She was right. I was astonished people reelected her with a line like that. And now Gretchen Whitmer was reelected after promising to "fix the damn roads" and not doing it. Trump may be more blatant, but I don't see him doing anything really out of line with other politicians.
That's always been my problem with the "Trump is a liar" remarks. The proper comparison set is *other politicians* and that set is the absolute worst set to look at and claim Trump is an outlier. I think Trump lies a lot (or has a very loose relationship with truth to the point of indifference), but so does Biden, and both Clintons, and so on. If you put Trump up against Jimmy Carter and call Trump a liar, I might believe it and take it seriously. You put him up against "Wipe it, you mean with a cloth?" Hillary Clinton, I just roll my eyes at both.
Trump is I think an outlier in that he tells *blatantly obvious* lies, demanding that his followers accept them as proof of their loyalty and ingroup status, and does not defend himself against the obvious outraged rebuttals except with repetition of the original lie. Other politicians mostly tell lies with the intent to deceive and with some plausible-deniability level excuse to hide behind.
I take issue with the claim that Trump is only somewhat worse along the scale of normal politician dishonesty.
Obama promised "hope and change," which isn't a statement whose truth value one can easily validate in the first place. This is typical of politicians. Gretchen Whitmer promised to "fix the damn roads." This is normal for politicians, but they do not usually have the power to unilaterally get things done while balancing other priorities, and this sort of electioneering essentially amounts to overselling an offer to make this a priority in office. This is behavior that could reasonably be described as dishonest, and is a good example of why people usually have low opinions of the honesty of politicians.
But before Trump, most politicians have been very wary of making statements that are non-weasel-wordily, unequivocally false, at the moment of their making them, in a way which could be positively proven to be the case. People interpret that very weasel-wordiness, which allows politicians who make misleading statements to backtrack or justify deeply misleading statements as not having been lies if they're caught out, as a sign of dishonesty, because they're so avoidant of clear and unambiguous assertions. This was an advantage for Trump, many of whose supporters saw him as more honest, because he was prepared to make clear, unambiguous assertions which were flatly false.
Saying that he had closed his Chinese bank account when he had not, or that he would publicly release his tax returns after election and then not only not doing so voluntarily but spending his whole term in office fighting against being compelled to do so in court, are dishonesty of a sort that politicians before Trump have rarely exhibited. Just as the media rarely "lies," according to a strict, narrow sense of lie, politicians also rarely "lie." Trump is an exception to that norm, and took advantage of the public's expectations of bounded dishonesty among politicians.
Yea, this. Trump is much more specific with his lies than the professional politicians and also prolific, often weirdly so. And then once he's done it, if challenged he invariably doubles down. For years I thought that Bill Clinton was the most dishonest person to get in or close to the Oval Office during my adult lifetime, but Trump makes Slick Willy's deceitfulness seem almost quaint.
Most politicians lie or exaggerate or spin towards some specific purpose, to cover something up or change how something is perceived or harm their opponents or whatever. And Trump certainly does that stuff. But then he also just literally makes shit up, without being asked, and spews it out. Have you ever attended one of his rallies? Friends of mine have and one of them once made me sit and watch like 40 minutes of live video that she'd recorded -- Trump stood up there at the microphone for that whole time and literally every other sentence included a flat-out lie. About all sorts of random stuff besides big issues -- personal stuff, and even just random topics.
I think the giant lie about the stolen election is far more egregious than the distortion most politicians have hitherto peddled. That suffices for me.
<i>Lying has a simple and precise definition: "Making an assertion which one does not think is true."</i>
I'd add something about having an intent to deceive, otherwise actors, fiction authors, and people telling jokes would all count as liars, which seems implausible.
I think the reason that the extremely narrow point is not actually vacuous is one the article makes a strong case for, it's just not a case that's compressed into the headline.
If you have a goal of censoring false and misleading news, it's extremely salient to that goal if very little news meets strict and uncontentious criteria of "false," and which news one regards as "misleading" depends overwhelmingly on one's political affiliation.
There's news that pretty much everyone, liberal, conservative, or occupying some position outside the left-right divide entirely, can accept is simply factually false, and our information landscape is not impoverished by rejecting it. Unfortunately, once you remove all of it from the equation, most of the actually contentious news is still left, and if you resolve to censor "fake news," you're necessarily going to have a big fight about what it is. If any platform chooses to censor "fake news" according to its perception of what qualifies, that perception is almost certainly going to reflect their own biases, because in most cases, determining what is or isn't "fake" relies on heavily value-laden judgment calls.
By more expansive definitions of "lying," I think that both left and right-aligned news media "lie" fairly often, but as a society, I don't think we have very good means to impartially arbitrate when they're doing that.
YES! This is how I interpreted Scott’s posts from the beginning, and I felt like I was taking crazy pills that no one in the comments seemed to get this. I’m glad I’m not alone.
Right. Also, censoring all news media that engage in sloppy or motivated reasoning, omit relevant facts to tell the story they want to tell, imply far more than they can prove, etc., will mean censoring basically all the news media, most politicians' statements, a large fraction of official government statements, etc.
Very few comments here would survive, either. It's not constitutionally possible to censor the fuck out of "news media" but leave ordinary blog commenters untouched. The First Amendment does not distinguish between the New York Times and Joe Sixpack keyboard warrier banging out a righteous takedown of whatever the guy above him has said. If it's illegal for the first to engage in sloppy or motivated reasoning in print, then it's illegal for the second, and for everybody, large or small. Conversely, if it's protected speech, it's protected for everybody.
Nit: it's constitutionally possible to establish a censorship office that doesn't have the resources to touch more than a tiny percentage of blog commenters and whose internal incentives are best satisfied by using whatever resources it does have to go after the biggest, juiciest targets. I wouldn't recommend it; even if one would consider that end result desirable, there are too many ways for the implementation to go badly wrong. But it is possible.
You're arguing prosecutorial discretion is constitutional? I don't disagree, but you still have to criminalize what you don't intend to prosecute, you can't differentiate in the statute itself. Exempli gratia, the fact that the 55 MPH speed limit was never seriously enforced in Montana doesn't mean it was therefore not illegal. And if you had the bad luck to be one of the few drivers who got a ticket, it would do no good at all to argue to the court that so-and-so blew past a trooper doing 85 and the cop merely waved hello.
Looks to me as just another instance of the decoupling/contextualizing divide. To some people the narrow point that the NYT and InfoWars can be described as committing the same sort of sin is irrelevant, because they and their allies consider themselves capable of reliably telling the crucial difference anyway.
All this discussion about "people want to censor X" would be greatly improved if people would specificy *who* they imagine is doing the censorship in their scenario.
Like, the government needs to be extremely cautious about punishing all but the most obvious and disprovable lies, because when the government acts it's threatening someone with fines and jail. But if I'm a Reddit moderator, I can be a bit more free about saying "I will take down vaccine misinformation, using my own judgement about how misleading your posts are" because the harm of being unfairly banned from a subreddit is far less.
In recent context, I would say that this is Twitter (or other social media) specific at the least. That Twitter weighed in on the side of certain viewpoints (generally towards the things the federal government was pressuring them to support) and against alternatives - including alternatives that were plausible at the time and have since been proven true.
I thought something similar at first, but then I realized that in the context of whether or not we should censor “lies“, it’s actually a very good definition to use.
It's the *only* definition that is useful to the larger argument. Scott was not originally trying to argue in the first two posts that people have the wrong definition of lies, but rather, that we are going to speak specifically about things whose veracity can be objectively determined. It was probably a mistake to call those "lies" given what he was trying to argue.
However, at the end of *this* post, Scott seems to be arguing for exactly what everyone has been disagreeing with him about for the other two posts, so maybe they were onto something all along.
No it's not, because if it's the definition you use, the censorship stops being useful, precisely because it's not screening out much - and certainly not the stuff the would-be censors *want* to apply it towards.
Yes, I think that is the point. Those who wish to censor want to have (and imagine they will have) a much larger effect. I believe Scott's point is that anything approaching "objective" censorship will have a very small effect, and thus, to get the larger effect most censors seem to want, one must of necessity enter the realm of interpretation, with the inherent cost being the censors' priors significantly influencing what counts as "lying" and that those in power have increased incentives to further expand the definition of "lying" because it will serve their ends.
If you want to sign on to a censorship regime it makes very much difference what counts as a lie and what does not. If you include much more than Lie Type 7, it becomes very hard to justify any of your censorship.
"A post titled "The Media Very Rarely Completely Invents Facts Out Of a Whole Cloth With Zero Relationship Whatsoever With the Truth" would be a pretty boring and pointless post..."
I think this line is the *observation* more than the argument of the posts. The argument appears to be that it would be hard to arrive at an objective model for censoring "misinformation / disinformation" that does any actual censoring, which seems considerably more interesting.
I'd say if the post has made you turn away in disgust from debating about on whose forehead we get to gleefully slap the incendiary and pejorative label LIAR! -- and we need, instead, to take the much more tedious slog to rhetorical victory of digging carefully into the complete picture of the thinking represented by a statement and slotting it carefuly into one of Scott's 7 categories -- why, this is a quite useful result. Conducive to a more civilized public discourse.
Victorian gentlemen use to resort to pistols at dawn when called a liar to one's face. I doubt very much Victorians lied any less (or more) than we do, so I do not think it had any salutory influence on public honesty -- but having that odd firewall in public discourse did mean they did not sling unforgiveable personal insults at each other in the heat of the debate nearly as often or as casually as we do. Might be worth considering.
It's worth making the argument, because there's a sizable number of people who would disagree, and state that the media actually does invent facts from thin air.
It would be nice if we never had to state the completely obvious, but we don't live in that world.
The Dominion vs Fox case is getting some quotes in depositions that get pretty close to 7. The issue is booking a guest who you believe is advancing a false theory, and then giving her a pretty uncritical slot to make her case.
>Two years later, [Hannity] was asked about Powell’s theory in a seven-hour deposition that was reportedly shared during a court hearing earlier this week as part of the Dominion lawsuit: “I did not believe it for one second,” he said under oath. Powell also walked back her theorizing in 2021, with her lawyers stating “no reasonable person would conclude that the statements [Powell made] were truly statements of fact.”
I think the motive here is that Fox is worried about losing market share to programs like OAN, and needs to put some stories they believe to be false on air in order to keep credibility with their audience, who believes them to be true.
I would count that as lying. At a more polished, fact-checking place like the NYT, I think the behavior would less often be a lie of commission, and more often killing a story that the paper thinks holds up, but they think their readers won't believe. (This is the claim made by Nellie Bowles about her experience at the Times).
In your list, you clearly forgot 4.5: Not reasoning about the truth/falsity of the question at all, but instead answering with some mix of:
- what you believe you are supposed to answer
- what you believe people like you are supposed to believe
- what some higher status than you person said about the question, or someone whose reasoning ability and data sources you respect
- what you expect will make the person asking the question happiest
- what you think the person answering the question expects (i.e. how can you avoid seeming "weird" to them)
Arguably some of this counts as reasoning, but except perhaps for the strategy of "I don't understand this, so I'll trust an expert", none of it is reasoning about the topic at hand.
Came here to say exactly this. Bullshitting is an important part of the falsehood ecosystem, and it doesn't really fit on the list - maybe it's close to (4), or maybe it's actually close to (7).
Even more typically than DinoNerds' bullet points, bullshit is often motivated by
- what you think you can say to get what you want.
Much of this is how we learn truthful reasoning, though, if we're lucky enough to be in a community where social rewards are tied to learning truthful answers.
When you don't already know how to reason about the topic at hand, and you're still learning, you shoot yourself in the foot if you're not willing to "stoop" to some "fake it till you make it". (I learned this after shooting my own feet an embarrassing amount.) "I'm not sure how to reason about this problem on the problem set, but I bet my teacher is good enough to have given me a relevant example, maybe even dropped a hint about which example to use," is reasoning about the social context, not the problem at hand. It's "faking it", to some extent. But it gives you resources to do the problem at hand in order to develop eventual understanding.
Creating a social context that channels the desire to give a pleasing, "normal" answer into finding an actually-correct answer seems crucial to learning from preschool through university. Deep understanding often takes a bunch of shallow conventions for granted, like the order of the alphabet (the ABC song trains children in the order people "like them" are supposed to believe). And, when you don't yet fully understand a problem, going through the motions of how you've socially inferred "it's supposed to be" solved can help you understand. Answering as you're "supposed to" develops the capacity to reason about the topic.
I learned a lot of things in school that weren't in fact true, particularly in ELHI. I learned even more things that weren't true from religious adults and religious leadership.
Children tend to absorb what those around them claim, without too much critical thinking, and that's a good strategy, both for learning useful information and for avoiding doing things that will get them beaten up or worse. They may or may not eventually go beyond parroting "God is good", "Canadians are better than Americans", "Columbus discovered America", "Eating fat is harmful; sugar is great", etc. etc.
I'm not sure that they actually learn how to go beyond it anywhere in the school system, as commonly experienced - or even whether, as children, they have the cognitive capacity to cope with their teachers being wrong, without experiencing it as betrayal and disillusionment, rather than simply the normal human condition.
"I'm not sure that they actually learn how to go beyond it anywhere in the school system"
Some learn to go beyond it, and those who do usually first have gone through it. That is, attributing the success of these "some" solely to individual superior aptitude, rather than the conventional, sometimes-wrong, education which gave them the scaffolding to exercise their aptitude, takes the scaffolding for granted, and it can't be. Most people who contribute to the sum of human understanding won't be savants from some isolated village with little formal education: there are some of those, but not enough; education, for all its flaws and betrayals, really does help.
It would be strange to say someone who plays piano well doesn't really understand playing piano, or someone fluent in Portuguese, or at constructing correct proofs in geometry and spotting errors in incorrect ones, doesn't understand Portuguese or geometry, respectively. Someone successful at fixing cars understands *something* about them, even if not enough to accurately explain the science of how they work. For much understanding, doing something well enough is adequate evidence that you've gained understanding about the subject itself, not just BSing.
Then there are other subjects, where it really is much harder to tell...
To say that (almost) every genius first went to school, so school must be helping, doesn't really follow when schooling is compulsory. Might as well say that paying taxes (the act itself, not anything the taxes are spent on) is essential for motivating scientific advancement, or something.
Modern schooling does an awful lot of teaching "guess the teacher's password", and regrettably little of teaching "find what is true"
I'm not claiming the education we were forced to attend must be of the best kind – because, look, most geniuses have gone through it, too!
I'm saying even the very talented and intelligent aren't born knowing how to reason about the subject they develop outstanding aptitude in, and typically benefit from social prompts to learn. I'm saying that being in a community where true reasoning and aptitude are socially rewarded matters to developing that aptitude, and that development often includes (even for geniuses) a component of "going through the motions" before really understanding what you're doing and why you're doing it.
I'm also saying that the community can be far from perfect at rewarding true aptitude and be better (even for geniuses) than lacking community – but saying even a flawed thing is better than nothing isn't praising the flaws themselves. Even so, intelligent individuals good at spotting the flaws might underestimate how much their schooling did prepare them to learn true things.
Kids who learn reading, writing, and arithmetic, and that libraries are wonderful places where you can learn stuff for yourself, from their teachers (including parents and private teachers) have been schooled in something truly useful for finding out truth, even if that schooling came with a fair amount of guff. That doesn't make the guff good, or prove that their schooling couldn't easily have been improved on. But the less literate and numerate you are, the harder it is for you to even tell when your teachers are wrong: An education that leaves you literate, numerate – and interested in truthfulness – enough to resent what your teachers got wrong is an education that has partially succeeded – including at teaching you to value truthfulness.
Something like this issue came up back in the bad old days, when good upstanding anglo protestants were sure there were Jesuits around every corner, and each one had an imprimatur from the Pope itself to lie their heretical little heads off.
John Henry Newman actually reminds me of you in that he is sometimes possessed of such moral rectitude that it can be painful to observe him from the outside... pearls before swine is a phrase which comes to mind. Anyway, he discusses a similar issue (lying vs. deception—that is, what exactly constitutes a lie, & the exact limits of the morality of lying) here:
To me this all seems quite simple. You're absolutely right. *Lying* (presenting false propositions) is not *deception* (presenting true propositions in a such a way that, taken at face value, they seem to imply false ones). Whether intentionally or not (whether culpably or not), the media is very deceptive (1-6)—yet very rarely do they go so far as to outright lie (rarely do they get all the way to 7).
The problem is that, as you note, people sloppily equivocate between lying, mere falsehood, deception, and a mere disregard for the truth.
The only way to get out of this, in my opinion, will be to return to moral realism—and ultimately, to natural law. People only began to lie because they did not observe the 9th commandment, which obviously proscribes both lying and deception (which latter thing all ought to agree is wrong, excepting the edge cases Newman discusses above).
To this you might reply that, even without the ten commandments, all men agree that deception is wrong. But in fact, on consequentalist ethical systems, it simply isn't.
I've long been puzzled by the hatred many people have for consequentialist ethics. Your comment is helpful--I agree that a naive, zero-step lookahead consequentialism would allow more deception than would be good for society--according to consequentialist ethics.
Try thinking of consequentialist ethics as ethical epistemology, and virtue or duty ethics as pragmatic implementations of ethics that you can give to computation-limited people in order to implement the results of consequentialist conclusions in the real world.
By ethical epistemology, I mean that consequentialism is the right method of deciding what's good and bad for someone with omniscience and infinite computational power. A consequentialist with lots of computational power would look ahead and see that telling everyone that it's okay to lie if that seems like them to lead to the best outcome, would lead to most people lying a lot, a breakdown of trust, and failure to cooperate or compromise.
Consequentialism gives an approximation to ethical judgement that gets more accurate as you give it more information and computational power. The approximation can be very bad if you use a greedy consequentialism with no lookahead.
A consequentialist who understands human bias would also realize that most people are overconfident in their opinions and computations, and that most people are too dumb or ignorant to act very ethically if they think for themselves. People with different degrees of intelligence, or different kinds of expertise or jobs, should have different ethical heuristics. The whole system may, if conditions are right, evolve into an ecosystem of different ethical systems which together implement an approximation of consequentialism.
(I don't believe anybody has ever successfully *designed* a distributed virtue ethics to implement a consequentialist ethics, nor that any attempt to do so is likely to succeed.)
Attempts to give everybody exactly the same virtue ethics can also work, but it should be realized that this is a pragmatic approximation, not adherence to some eternal code of behavior dictated by a great sky-being. It will break down with people who have to make very high-stakes decisions that need to be correct, like soldiers, doctors, and the rulers of nations. Such people must have the leeway to do things that society at large, with its simpler ethical heuristics, calls evil because they are discouraged from thinking consequentially.
So in building your society, you promote virtues like truthfulness. But these aren't absolute, eternal values; they're heuristics, things easy to grasp which work well most of the time. Only when society is faced with a long-term, complex, high-stakes moral problem do you break out the big guns and apply consequentialist reasoning, slowly, carefully, and preferably collectively, because only the "ultimate consequences" determine whether some action is good or bad.
("The ultimate consequences" may not exist. Total utility, even time-discounted, isn't guaranteed to converge as you look farther and farther ahead. I don't think this is a problem in most situations. You do the best you can with the resources you have, and call it ethical.)
We teach children virtue ethics because it's the foundation. Lying is bad is simple and easy to comprehend, but what about when you are older and ready for more complexity? If you are hiding Jews and the Nazis come to your door, is lying bad? The consequentialist question becomes "when is lying bad?" Phil makes that point well.
>People only began to lie because they did not observe the 9th commandment
Suppose I think that the Bible is false, what reason do I have then for following the 9th commandment and/or the natural law? The fatal problem with religion in the modern age, that many philisophers have observed, is that it's moral authority is inextricably tied to plausibility of its ontology and factual claims, and those look ever more untenable as our understanding of how the world actually works grows. The God is dead, we killed him, and he's not coming back, so maybe it's time to seriously consider alternatives?
The idea lying is wrong doesn't depend on belief in revelation. I think that most people would agree with it intuitively, but you can argue for it on purely natural grounds.
On a personal note, I resorted to God because I felt that I had exhausted all alternatives. In retrospect it seems like a living example of reductio ad absurdum.
Sure, aversion to lying is arguably both an evolutionary adaptation and could be justified on consequentialist/game-theoretical grounds. In practice people don't derive normative rules from first principles (and are incapable of doing so reliably), but instead use time-tested templates provided by culture, which is where deontology and virtue ethics ultimately come from.
I've heard from a few people that they find religion a more tolerable alternative to despair of nihilism, but I just can't understand it. I can't make myself believe something in spite of overwhelming disconfirming evidence, and I probably wouldn't want to even if I could.
The idea is that their intuitions depend on truths, even if these truths are implicit to most people (and even if most people are far from living up to what they say they believe).
A brief argument to justify that lying is really morally wrong: If we hold to the golden rule, then we ought to reject lying. Because we don't wish to be deceived (or as you said: even if it is possible to wish for that, it shouldn't be), we shouldn't deceive others.
On consequentialism one ought to become religious, provided that it can be shown that doing so would make one happier! But I don't hold to that logic—it's not that I became religious because it was more comforting (I agree that it's evil to try to believe in something untrue), but I turned to God because the alternative premise had lead to contradiction and absurdity.
I should specify that on the religious picture, conversion happens because of grace, not just free will.
I think that we basically agree regarding lying, just frame it in different terms/coming from different aesthetics.
In terms of absurdity, it seems to me that both accepting and denying the existence of a (deist) God leads to absurdities, like there being an uncaused cause or there having passed an infinite amount of time prior to the current moment.
However, even if you accept a deist God, I've never seen a good argument for further subscription to any particular dogma, e.g. a Catholic, Muslim or Buddhist one. They all seem strictly epistemologically inferior in positing a vast edifice of outlandish unsubstantiated claims.
You can't distinguish between someone who tells false claims and doesn't care if it's true or not (4) and someone who tells false claims intentionally (7) you just incentivize people like Trump who always fall back on a type of "insanity" defense. Trump tells false claims all the time but he (and apparently you) defends himself by saying that he didn't know it was false, even though anyone reasonable presented with the same facts as he was would know that it was false.
At that point, it is impossible to identify anyone as lying according to your definition. Even though Trump is a clear liar, who has lead fraudulent enterprises on multiple occasions, one could never criticize him for it. And that is a loss, because his reckless disregard for the truth greatly harmed our societal ability to have any conversations. In addition to the immediate effect of many people believing the false claims that he made (even though you won't allow me to use the word "lie" to describe it).
In theory, it can be very difficult to distinguish 4 and 7 as you say. In practice, people in 7 send each other messages saying "haha the suckers fell for it" in a slack channel called "liar liar pants on fire".
Except that Alex Jones did that and Scott still questioned whether that was actually what he felt.
Additionally, unless everyone (including Scott) agrees that Trump is in category 7, which I don't think everyone agrees on, then he is taking advantage of the loophole that Scott's "useful idiot" stance creates.
All that stuff about how paranoid your readers are speaks to me, I’m also often shocked by the paranoia of my comments section, often relating to the exact same issues and in the exact same ways. Do comments sections just draw paranoid people or are we naive dopes?
I am one of the people who was/is very hard on the media.
But my guess would simply be that people like you and Scott who are trying to create positive information transfer have a greater appreciation for how difficult it is than people like myself on the outside who can just get riled up about all the failures and make angry japes to vent our disappointment.
I do think the core desire for strong condemning language around this issue is the vast gulf between how much imprimatur as arbiters of truth the journalists/outlets/acolytes of the MSM allocate to themselves, and how much they really deserve.
The focus isn't on the NYT/NPR as a bunch of flawed humans doing the best they can. And it isn't on that because the NYT/NPR presents and postures itself as THE PLACE the go to get information about what is really going on in the world. And it manifestly both fails at this regularly, and often isn't even trying to do this.
And the frustration is so large people just want to bring out the largest rhetorical guns to blast them they possibly can justify. Which of course is ironic. Humans are dumb, especially me.
I have a hearty dislike of the media also, but it has little to do with how much they lie, bullshit, or deceive, because I don't think they do so to any significantly greater degree than good ol' boy Walter Cronkite or fast-talking Ed Murrow back in the day.
What annoys me is the saccharine unction that drenches everything, like they're all college sophomore Sir Galahads rescuing damsels from Sir Mordred all day long, instead of just hacks pulling down a salary by shipping out a dog's breakfast of what seems to be roughly true and approximately interesting at fast-food speeds.
If they re-adopted the world-weary cynical hard-drinking edge of the Raymond Chandler reporter ("Ah...I'm just a hack trying to make a livin'. Didja know about this bum politician's shenanigans? Whaddya think...?") I'd probably be fine with them, even if they were just as often wrong, and pushing some establishment narrative or other just as hard.
It's like going to the hardware store for a screw to fix the microwave, and the sales help says "A...screw? Nonsense, you poor ignorant peasant, what you want is a fastener, and they come in a wide variety of types and sizes, which passeth your understanding I am sure, but I majored in Truth and Wisdom, so if you stick with me for the next hour I'll begin with basic metallurgy and Roman techniques of construction, and I daresay we'll uncover the real root of your personal problems -- tell me, have you ever kicked the dog because the microwave won't start? Tsk! -- and how you can get therapy to learn to just accept the microwave as it chooses to be in the moment."
The urge to clop them upside the head is strong. Just give me the fucking screw. Just tell me whether Backhmut was overrrun, the price of gas went up or down, and whether Amazon is hiring again, then shut up and play some Cheap Trick or Mozart. If I want to be scourged I'll go to church.
Your account of Raymond Chandler amused me, because that's pretty much the impression I got from Trump. I could easily imagine him saying "I'm just a guy trying to MAGA. Here's this problem happening at the border. Whaddya think?". I (not a Trump supporter, but just opposed largely on account of his lack of experience) naturally felt I couldn't believe anything he said unless I'd checked it some other way. However, I also felt like that was *obvious*. Like a guy selling knockoffs at a street corner. I could see it coming.
Counterintuitively, this worked in Trump's favor. Why? Because he was basically competing against the clean-cut guy in the department store... who was *also* selling knockoffs.
Who writes the most memorable and negative reviews on Yelp? People of rigid patterns of thought who get hung up on one small negative and just go nuclear over it.
Someone who reads a long nuanced post and thinks "eh...there's a couple points here where I think this wavers from Gospel truth...some footnotes/caveats/elaborations/softening adjectives that I would definitely add..." is not going to be sufficiently driven to wrestle the Substack comment system into submission in order to make a long nuanced comment of weak partial disagreement -- and even if he does, it won't stand out as much to you, the author.
I think it's sort of the parallel to the Twiiterverse being far more shrill, dogmatic, and Manichean than actual people in their actual lives.
Comment sections absolutely attract the paranoid, as well as a few other categories of unusual people. (I’m not sure which categories I fall into, but I recall that traditional comment sections represent a fraction of a percent of readers, and even Substack comment sections represent only about 10-20% of readers, who are already self-selected in all sorts of weird ways.)
Yeah reminds me of when I was a volunteer developer on an extremely popular open source game, and all the developers and friends of the developers did mostly MP, and all the "regular" players on the forums etc. did at least some MP.
So imagine our surprise when the data showed that as far as total hours played SP was ~97% and MP ~3%.
I saw an example from World of Tanks, where apparently if you post on their official forums you have to log into your account and it shows some basic stats about you. Some guy was complaining a lot about the game or some update. He had played about 100 matches and had over 9,000 posts on the forums.
My guess (and I do not claim a lot of confidence here): The combination of the internet and the financial collapse of a lot of media has left us with tons of very visible examples of lies and deception and general dysfunction by all kinds of high-prestige, formerly-trusted institutions. The result is that everyone starts out kinda defaulting to skepticism or wondering if they're being lied to all the time.
I'd compare this with how the replication crisis has affected my reaction to reading about new results in the social sciences--like I should start out with high skepticism and a cynical "this was probably p-hacked or something" assumption, because I know how much BS was published in the social sciences in the past, and it's hard to believe that there isn't still plenty being published now.
Similarly, you can now access rebuttals to assertions in the major media, which was very difficult to do pre-internet. Some random guy with the time and inclination can point out the falsehoods or misleading information in a major news network's reporting, which I think has resulted in a lot of us noticing how frequently the news does this. It may have been similarly common in the 1970s (to pick a random pre-internet date), but very rarely known.
I think Dan Rather's scandal was the point of change, where it became obvious that the public could actually affect things by fact checking the major media.
Speaking of which, I haven't seen Rather's name in this discussion. While I'm pretty sure he didn't technically lie in the initial report, his continuing to maintain his belief in the documents gets to the point where it's impossible to distinguish from deliberate ignorance.
OTOH, but 60 minutes did that misleading edit of their interview and was caught on it in (I think) the 80s, and seemed to keep their prestigious trusted place in American society. Maybe the existence of alternatives matters here?
60 Minutes did have some prior scandals, but to my knowledge they were opposed by companies with the ability to hire lawyers to challenge them in civil court, and it took time for the courts to work. Rather was exposed, not by the target of the segment lawyering up and taking the time to go through the court system, but by a blogger, and the false material was exposed as false in a matter of days.
And 60 Minutes (and even Rather himself) still haven't completely lost their trust.
Hah! If you and Scott are naive dopes, honestly, I am, too. ...I also rarely comment, in general, so I suppose I'm not helping the statistic. I think Carl Pham's comment here is spot-on - people with stronger opinions are just more likely to post, and paranoia breeds strong opinions.
By the time you get to "Reasoning badly, because you are biased, and on some more-or-less subconscious level not even trying to reason well.", you're full blown trying to deceive people in ways which you would not if you felt accountable for what you were saying. This is the *central thing* that accusations of lying are for, and so the term "Lying!" absolutely fits when someone makes real sure to not reason carefully because they anticipate that reasoning carefully would cause them to know that the things they are saying aren't true. After all, there's conservation of expected evidence, and if they know to avoid reasoning carefully they damn well know enough to conclude that what they're saying isn't likely to be true.
Scott's category there is not people doing it about a specific issue, but that that sloppiness is part of their character. These people are not "anticipating that reasoning carefully would cause them to know that the things they are saying aren't true", they just don't reason carefully as a matter of course, perhaps because they consider feelings to be more important.
I'd say most people fall under that category, and it doesn't seem useful to say most people are liars.
No, they definitely are aware that careful reasoning would lead them to conclude that they're speaking falsehoods.
The test is to point them at the flaws in their thinking and see what they do. If they say "Oh shit, thanks for pointing it out!" (i.e. show genuine surprise), then they weren't aware. If they don't even falter in their BS and keep on as if you hadn't said anything to challenge them, then it's possible that they genuinely don't care enough to even anticipate and you get no data (though they likely do anyway). Generally though, these people will make active attempts to avoid looking at the thing you're pointing at, and you can only do that if you already know that looking there will do something you don't want.
Popularity of lying cannot make it not a lie. You just have to be careful with how you apply your limited attempts to shame people for things, since lying is too ubiquitous to just knee jerk it away by using the word "liar" as a shaming tactic. Christianity knows this distinction well. "Everyone is a sinner", "Love the sinner, hate the sin", etc.
I think most journalists are consistently ignorant about pretty much every technical subject. This is part of the point of "Gell-Mann Amnesia" - every time I read an article in a field where I have technical knowledge, it's either wrong or so garbled that it's not even wrong.
Garrett complained about the media's problems with reporting on firearms, but the problems there are not particularly distinctive. I've just come to mostly accept that articles in the general press will make egregious errors on everything other than easily-available, easily-understood public information. This isn't primarily because of bias, and it's certainly not lying. It's just people being dumb.
I would say the reporting related to guns is particularly bad because the particular tribe journalists tend to come from is substantially less likely to contain firearms enthusiasts. So there are a lot of newsrooms where nobody has a good bullshit detector for gun related information, because they don’t live in a social circle where a working knowledge of guns is in the water supply and where your buddies will laugh at you if you do something like call a folding stock “the part that goes up”.
Journalists are rarely “experts” on technical subjects, but there are definitely areas where they are more or less ignorant. You don’t need to be an expert to have sufficient working or casual knowledge to be able to sniff out likely bullshit.
This is one area where I think increasing diversity of backgrounds could be legitimately valuable.
Also, its not hard to find a competent gun expert, the Times, Post, etc just actively avoid it. This is not like employing a full time Biomedical Engineering expert for the 3 annual stories you are going to run about new developments in artificial organs. You are running hundreds of stories annually on the subject of guns, and you don't have anyone on staff who knows the difference between semi-automatic and automatic firing.
Right. There's a point where intentional ignorance becomes no different from knowing falsehood, at least morally. The standard for 7 should be "knows or should reasonably know that the statement is false". Reckless disregard for the truth is morally indistinguishable from a knowing lie IMO. It certainly is for libel standards (the former is what's know as "actual malice", in one of those "legal terms that don't mean what you think they mean" instances.)
Sometimes I do think it’s “actively avoiding it”. But other times I think it’s not quite that. It’s more like “so completely ignorant of the subject and unexposed to anyone who isn’t that you don’t even realize that a distinction might exist, or if it did, that anyone would care”.
Now I do agree that at some point if you’re going to write about a subject a lot you have a responsibility to cast a wide net and at least become familiar with a range of experts and opinions. So I don’t want to imply that I’m excusing ignorance or laziness.
But if you and literally everyone you interact with on a frequent basis all share the same blind spot, how do you know the right question to ask, who you should ask for help finding the right questions, or even whether there are any questions worth asking?
Because people have repeatedly told them that they're wrong. And experts have schooled them. And they've just closed their ears and said "nah nah nah, I can't hear you!"
Have “people told them” or have “people they don’t know yelled at/about them on Facebook or in media they don’t consume”?
I think we should be careful to avoid ascribing to malice what can be explained by bubble effects. Again, not an excuse - but maybe part of the reason. And I’m making a point that’s more expansive than just the gun issue explicitly.
Editorials in their own papers. Reports in academic journals. Court documents. All the things they're *supposed* to research as part of their jobs. Heck, even experts *they themselves talked to*. Many of whom have explicitly said that they told them differently, but they ignored the experts or misquoted them to say just the opposite and then quoted 'anonymous' sources which said utter obvious garbage.
On the gun issue in specific (and in many others in general), the kinds of crap they spout is of the order of "1 + 1 = banana". Not even *plausibly* wrong. And requiring willful blindness to reality to say without malicious lies.
Edit: And at some point, I don't believe there's a difference between actual malice (the literal meaning) and a suitably reckless disregard for truth or falsity. At least not in practice, as far as anyone else is concerned.
But this still explains too little. Liz Cheney is now kind of a NYT darling for being anti-Trump and probably has slipped pretty far into being purely anti-Republican. But I'm pretty sure she's been hunting and knows the difference between an AR-15 and a M-16. There are lots of people like her. Now, most aren't Vanderbilt heirs who can afford 3 years of unpaid internships followed by 10 years earning $40k/year in New York or DC, but that is their own weird problem.
Not just firearms. I am from Eastern Europe and have closely watched coverage about war in Ukraine. Journalists are extremely bad if talking about military matters, military weapons, weapon systems, etc. Making mistakes, not understading the subject and creating bizarre interpretations.
Historically, I think there was also a culture in progressive spaces where firearms knowledge was considered morally bad. Lots of people seemed fairly proud of their complete lack of knowledge, which confused me greatly, and I didn't witness that same effect with other technical subjects.
In Sunday school I was taught about lying being divided into "sins of commission" (category 7) and "sins of omission" (category 5 & 6). The point being that leaving out key points to deceive someone is still lying, just another kind.
"Commission" and "omission" may be a useful start on your names for every category.
"I think a more likely explanation is that many people are choosing to attribute deaths to the vaccine that are not actually from the vaccine. For example, let's say Person A gets vaccinated and dies shortly after of some completely unrelated cause. And let's say Person B, the loved one being polled, has priors about vaccines or the medical establishment or whatever that cause them to be convinced it was actually the vaccine that killed Person A.
"In hypothetical reality, Person A lived a rather unhealthy lifestyle, had lots of risk factors for a heart attack, and would have died from a heart attack, regardless of whether they'd gotten the vaccine. Then, when Person A does, indeed, die of a heart attack, and by sheer coincidence had recently gotten vaccinated, Person B blames the COVID vaccine when polled, but it wasn't really the vaccine that killed their loved one."
This seems fair, since this is how Covid deaths were counted. Once enough tests were available, anyone who died, was tested for covid; if they tested positive, they were reported as a "covid-related death". News outlets reported covid-related deaths as Covid deaths (which was, BTW, a very common media lie).
For one example, an employee of my brother-in-law was supposed to get a liver transplant. He was tested for Covid before being admitted, tested positive, and so was taken off the list of eligible liver recipients, and died soon after of liver failure. This was counted as a Covid-related death. Which it was, in a sense.
What's confusing is the fact that, in the face of unreality, people claiming to propagate The Truth actually propagate untruths, hence fascist 'anti-fascists', racist 'anti-racists', the proposal for a national Office of Disinformation -- presumably to identify which disinformation the government approves of.
Regressives declare that norms are obsolete and oppressive, then strain credulity to 'normalize' the bizarre. It all makes perfect sense if one accepts chaos. But if you're building your regressive paradise around chaos, you can't turn about and fault others for not recognizing your new, improved 'norms'. You just told us norms are obsolete. Oh! Wait, I get it. Our norms are obsolete. Yours, on the other hand, are golden. It's a pretty juvenile style of rebellion.
Brilliant, best piece I've read in awhile that exposes the sinews of the fundamental disagreements that hover over and stalk almost every expression on social media. And REALLY depressing. Thanks anyway.
This is a really straightforward case of "Taboo Your Words". It literally does not matter whether the media is "lying" or not. Scott's original claim was that the media rarely does number 7, and that this is important /specifically because/ that means it is very, very difficult to gain consensus that a given media outlet has done something worthy of censorship. Scott's point is that "reasonable people can disagree" about almost anything, which is proven in how many reasonable people are disagreeing about every one of these issues in the comments.
Given this wide swath of reasonable disagreement, what kinds of statements can be responsibly prohibited or punished? A vanishingly small number. Almost none.
That is the point. That is Scott's claim. That is what you have to engage with. To say "No, I think this behavior should be called lying" is to change the subject. To say "Actually the New York Times is better than/just as bad as/worse than" InfoWars is to change the subject. Your task is to find an article so egregiously false that almost no one in the world would take issue with you trying to get it banned. Until you can do that, Scott's point stands as almost trivially true and impossible to deny.
Yes well said and I thought this was also the whole point of Scott raising this topic, so I was a little surprised that he didn't underline that in this post and seemed focused himself on parsing words. I felt like if I were Scott I'd be shouting "you're not hearing me!" but maybe he feels that way a lot and so this discussion didn't feel more egregiously that way to him.
Censoring is very difficult to start with for the stated reason, and even more difficult when there is uncertainty involved about an alleged future outcome that is unprovable. Many of the most emotionally entangled subjects up for censorship fall under this category.
I do think this argument goes the other way and we should try to build a consensus for censorship that goes beyond category 7. The hard part is figuring out how far to go.
>Given this wide swath of reasonable disagreement, what kinds of statements can be responsibly prohibited or punished? A vanishingly small number. Almost none.
Which is absurd, because this media furiously demands that people not literally lying need to be censored.
But most calls for censorship are not about truthiness, but specifically and carefully couched in terms of "harm". In other words, they would not care too much about deceptions that they think reduce overall harm, like claiming that vaccines prevent transmission which incentivized people to get vaccinated, just like they *would* care about scientific truths that they think will cause harm, like acknowledging differences between genders/races/whatever-group.
I can find dozens of articles that are egregiously false. Scott will refuse to put them in category 7 because he will blow smoke about how they may have been honestly mistaken. Moreover, even if they fell in category 7, many would disagree that they should be censored. So this categorization doesn't actually help in the censorship conversation.
How about separating the factual claim of whether people actually lie from the conversation about what appropriate censorship is, which may or may not have anything to do with lying at all!
I agree this is the argument Scott is trying to have, but I think a lot of the comments here are not as orthogonal to Scotts claim as you seem to think. Specifically that because a media outlet rarely does 7, it follows that it should be hard to censor 'misinformation'.
However this is totally contingent on accepting the idea that only 7 constitutes real misinformation. Arguing about what is or is not a lie, is directly arguing with a core assumption of Scotts arguments. If somebody holds that anything beyond 2 is misinformation and should be subject to censorship, that is a consistent principled position that they can maintain. If Scott can not defend the idea that only 7 is misinformation that can be fairly censored then his 'true' argument is necessarily unconvincing.
It is not enough for one person to believe that anything above 2 is misinformation to collapse Scott's argument: it would require that everyone else agrees on that that is the correct tier, and to agree on to what tier any given claim belongs.
Scott is focusing on tier 7 because people often say, when defending censoring misinformation, that outright fabrications of that kind are a) uncontroversially bad, b) easy to identify, and c) common enough to be a problem. Scott was trying to disprove C, but ended up disproving B as well.
The goal of the censorship scheme is to appear as fair and unbiased as possible, and Scott is arguing that claims from tiers 3-6 are so fraught with emotional and political valence (or sometimes just genuine complexity or uncertainty) that any misinformation filter that applies to them will necessarily appear unfair and biased to large groups of people.
The claim you have to defend is not that "anything beyond 2 is misinformation", consistent and principled as that position may be, but rather that /everyone agrees/ that anything beyond 2 is misinformation, and /everyone agrees/ which articles fall into the category and which do not. Otherwise the misinformation filter is going to start removing stories you believe are true.
I was responding specifically to your point that arguing about what is or is not a lie is orthogonal to Scotts position, 'changing the subject', when in fact I think it is directly relevant and a central part of the argument. Scott must defend his focus on 7 and if he fails to do so his argument fails to be convincing. There are tons of largely unsupported and unargued assumption about censorship and what the public believes, how they do/will behave. You think the comment section shows people being unable to understand Scotts position, I think it shows that Scott's model of the problem and actors is wrong, such that drawing conclusions from his model is likely to produce error.
Honestly, I don't think retreating to 7 produces the results that Scott seems to think it would, I think a very large cohort of the relevant population specifically believes that some 'facts' are more false than true, and for example, reporting 'factual' race crime statistics is actually reporting something that is false, because it fails to grapple with the broader implications and context of the issue. 7 is IMO a false category when it comes to how groups of people view reality and truth and far from something that society can/is coordinated around.
These kinds of fundamental disagreements with Scott's model of reality are what the comment section is reflecting.
I don't believe the comment section shows that people are unable to understand his position, I believe it proves his position. If there is any substantive disagreement about what constitutes a lie---which this comments section undeniably demonstrates---a censorship standard based on "lying" is worthless.
I think there is a really fundamental disconnect here, I saw Scott as making the point, to quote you,
"Scott's original claim was that the media rarely does number 7, and that this is important /specifically because/ that means it is very, very difficult to gain consensus that a given media outlet has done something worthy of censorship."
which I think is wrong as a comment about the world that we live in. There is significant consensus on the kinds of things that people want censored that do not meet the specific definition of 7. In fact I think there is a pretty strong consensus here in the comments that 7 specifically is not a good standard for censorship/labeling something misinformation. We currently live in a world where such standards are being practices everyday, and presumably the people who practice them find them to be valuable and are able to act because they have the support of their local consensus.
I think Scott is making a point that works assuming you accept his priors, and a lot of people don't, and these last two posts are basically Scott trying to defend his priors. I think that conversation is directly relevant to Scott's point, and you seemed to disagree.
The only consensus I have seen in the comments is the number of people who think Scott is wrong. They don't agree on what Scott is wrong about, or with each other on what is right. I think that's a very big point in favor of Scott's argument.
> We currently live in a world where such standards are being practices everyday, and presumably the people who practice them find them to be valuable and are able to act because they have the support of their local consensus.
If this were true, we wouldn't even be having this conversation. We would already be living in a misinformation-free world.
In the second article, Scott responded to several examples of things commenters claimed were obvious misinformation worthy of being censored. Do you endorse the claim that all of these were, in fact, valid examples of misinformation? Do you believe everyone but Scott endorses that claim? If not, your proposed consensus does not exist.
The American legal system has built out a robust framework for classifying and evaluating different levels of “being misleading” (ranging from intentional, unambiguous, damaging misrepresentation to unintentional, slight, inconsequential carelessness with respect to accuracy). Securities fraud case law is one example of a large body of reference material. Just noting that in case helpful to anybody doing a deep dive into this sort of discussion, no need to reinvent the wheel.
Other legal systems have done so too, I’m sure, but I’m unfamiliar.
How about "Reasoning well, and just having an interest in pretty (ostensibly, to most of us!) interesting things". The laptop, sure, but we've run over that ground enough. Compare it with Amy Carter, whose habit of reading during state dinners was grounds for a dozen stories. Or the Bush twins, whose famous underage margaritas will follow them to the grave.
The incuriosity of the media, even when it seems like they could "sell papers", has amazed me most. (Crime no longer automatically interests them - when they must know that their readers, or at least their female readers, will never tire of the agony column.) Reporters used to like to dig up ... weird stuff. Maybe that was a job best done away from a keyboard. Maybe it was not very edifying. Are we edifying now?
Because it was mentioned, I thought of the Zimmermann/Trayvon story. Now, I have no opinion one way or another about whether the "white frat boy" Zimmermann should or should not have been convicted. Actually, I don't even know what happened with that. But I one time listened to Loury and McWhorter on the former's podcast; and they presented some really strange info about the prosecution. Like, strong evidence that the star witness was not even who the prosecution claimed she was. The two of them had already smoked this, based on their shared and immediate suspicion about the incongruity of a "cute" boy like Trayvon, and the girl put forth as his girlfriend, whose testimony involved what he said to her on the phone that night, as best I recall.
This seemed pretty crazy!
Well, the world moved on from this case, so maybe it's not so strange that none of these discrepancies pointing to a rather shady prosecution, made any news. Still, I rather expected that when two well-respected academics made these claims and inferences on their podcast, that some news organization or other would take an interest.
As an attorney I can’t lie to a court. But I absolutely CAN selectively choose which facts to highlight, in service of the position I’m advocating. I would have been ethically allowed, had someone hired me to do so, to advance InfoWars birth certificate theory in court (standing problems aside). There are a few things considered lies of omission, but they’re much more black/white than this, e.g. if a case I’m citing was later overturned and I know that.
So the bar association agrees with Scott.
It all goes back to people’s desire that newspapers, at least the ones on the righteous enlightened side of issues where their readers surely reside, be thought of as motivated by a purer motive than lawyers. But lawyers and journalists are the same on this: we’re all just saying the best version of what our customers want us to say. Scott’s definition of lie is fine for both.
Here's another corner case: What about when someone does something 2-5, then is made aware of their error and chooses not to correct their statement and issue a retraction, because doing so would make them look bad or go against their desired narrative. Is it reasonable to say that their statement is now, retroactively, lying?
The original statement is not a lie, just a bad take. If you ask them directly and they repeat it, the follow-up is a lie.. But not going out of your way to bring the topic up again doesn't make it a lie,
So then Infowars lied, because they were corrected on every issue and continued to disseminate false information. But Scott insists that they might have continued to believe the false claims even after being corrected. There's no possible way to move into category 7 according to Scott!
Being misleading is not the same thing as lying. But it is bad. Lawyers, for example, are not supposed to mislead the court. That includes more than just not lying to the court.
People being misleading little sneaks is something that's perfectly valid to call out. They're just not technically liars.
I understand your point that there is legitimately no way to draw a bright golden line between awful Infowars style reporting and honest, excellent reporting that uses true facts to erroneously report a false conclusion.
There's also no bright golden line between negligence and failed due diligence, or reckless disregard for life and acceptable risk, or good actions and bad actions.
Using similar criteria we'd struggle to ever punish wrongdoing in any field. I agree that speech demands considerable more care than other things and would warn folks off of government control of speech for this reason. But to suggest that we avoid any line we can't draw in gold marker is to suggest no lines at all.
This is the fundamental flaw of Scott's argument, and I wish he'd grappled with it, rather than using needlessly inflammatory language and then getting caught up in arguments about it.
1. You may not be able to fact check a scientist yourself, but you can talk to other experts in the field. As Beowulf notes, an actual epidemiologist could likely have told the LA times that ocean aerosols weren't an area of concern. But they didn't ask, either out of laziness or because they were concerned their sensational headline would be endangered.
2. Usually when I criticize someone in media for acting uncritically, it's because they're citing a source with a known poor track record. Repeating something a white house press secretary says without comment is flatly irresponsible. Same with anything involving "anonymous officials within the intelligence community".
Much worse that not asking out of laziness is not asking out of ignorance. Like "this is a scientist, therefore we have no need to speak to another scientist".
A lot of journalists will speak to an expert on any subject and then that's sufficient: they have spoken to an expert, the expert has given them an expert opinion, there is no need to speak to a second expert.
This has been one of the fundamental problems with all science coverage: they don't know that they are doing it wrong, because "asking an expert" is doing it right by journalistic standards.
One really good example of 6 is a technique used by certain (tabloid) newspapers in England. They will report that "a source close to Downing Street" said something or other. Normally, that sentence would refer to someone who works in Number Ten, ie a politician or a senior political staffer.
But sometimes, they just guess what is going on (a smart journalist with good background sources usually has a pretty good idea even if no-one will actually tell them), so they send a junior staffer to go and stand in the street outside the building, call into the office on their mobile phone, and repeat whatever quote they decided they wanted to print. The person was close to Downing Street - he was standing right outside! - so it isn't a literally false fact.
Most of the time, the good political journalist has correctly assessed the situation and the result is that everyone is convinced that they have incredibly good sources and no-one can work out who those sources are. Occasionally, they guess wrong and they would look like idiots ... except everyone just thinks they were misled by a source who wasn't in the loop on that occasion.
I think this highlights a problem with the categorisation: if your distortion of the facts is in aid of a substantively correct conclusion, then that is a much less problematic situation than distorting the facts in aid of a substantively incorrect conclusion. That's not particularly correlated with the question of whether you are lying though.
I think there's lots of "mood affiliation" here: if someone is saying something they believe and which the reader doesn't believe, then they will think that they are distorting the facts to prove something that is wrong, and that's much more like lying than distorting the facts to prove something that is correct. But the definition of "wrong" and "correct" is, all too often, "what the reader agrees with".
Your categories of things-which-might-be-lying is missing "reasoning poorly because you don't care about the truth".
It sounds like you want number 4 to encompass that, but it really is a different thing than being biased. And it's different from number 7 unless you include "A implies B" implications as "facts" which can be made up.
And I think we should call this thing lying, even if we wouldn't call number 4 lying, and that this is what InfoWars does at least some of the time when it makes physical-world claims which are false.
---
But regardless of whether you call this lying, I think that it is worth at least noting that InfoWars is doing a _different thing_ than the rest of the media here. The NYT mostly manages to hedge its physical-world claims enough to avoid them being outright false. InfoWars does not. So defending InfoWars against a charge of lying requires us to give them the benefit of the doubt in their motivations in saying false things, whereas we can defend the NYT on the basis that they didn't say false things in the first place. This really is a difference of _kind_, not merely of degree.
---
Kind of a sidebar, but - I'm fine with giving people the benefit of the doubt, but that means we don't say "they are lying", not that we say "they're not lying". To say "when [establishment papers] do err, it’s by committing a more venial version of the same sin Infowars commits", that requires us to actively claim that InfoWars is not lying, which is different from just giving them the benefit of the doubt. The most we can say is that, lacking access to their motivations, they _might_ be committing the same sins.
On the Jeremy Goldberg Inflation comment: In high school I took AP Economics. The teach was a history teacher assigned to the class despite no economics background, but he'd been teaching it for many years, so I'd expect him to at least be able to parrot the textbook. In any case, he told us he could not understand what the difference was between deflation and disinflation. Note that I'm not saying he glossed over it or explained it badly, I'm saying he literally told the class he didn't see how there was a difference between these two things. I only talked to a few of the other students about this, and most of them didn't think his confusion was problematic, even though they were taking calculus at the time and so were definitely taught about the difference between functions and their first and second derivatives. So personally I find it *very* easy to believe that the entire newsroom at the Washington Post didn't notice any problem with what was being published.
Does “disinflation” actually mean a decrease in inflation rates while they remain positive? Or does it actually mean a (brief) period of decrease in prices that may not be long enough to drive year-over-year rates negative?
Disinflation is a negative second derivative of the price level, typically paired with a positive first derivative; deflation is a negative first derivative.
Both derivatives being negative would probably be described as "accelerating deflation" or similar.
Honestly, I think keeping track in your head of something like "first and second derivative of price level" is hard for most people. Hell, otherwise reasonably smart people get confused about comparing stocks and flows (approximately x and dx) all the damn time.
Were there any takes on your core conclusion (that stopping "lying" is an invalid argument for censorship enthusiasts) worth highlighting? I found myself frustrated that the comments I read were all focused on the nitpicky definition that was clearly highlighted as nitpicky, and not on what you were actually arguing *for*. If your nitpicky argument is invalid, does this mean that censorship standards can be invoked as long as journalists are avoiding points 6&7? All 7 points? What regulations would be put in place to address these?
> I found myself frustrated that the comments I read were all focused on the nitpicky definition that was clearly highlighted as nitpicky, and not on what you were actually arguing *for*.
On a couple of occasions I've tried to start making a point by establishing some common background material.
It's never worked. If you start with background material, you can't reach your actual point. You have to start there.
Okay, so now that we've discussed the problem for three posts, is it finally time to propose solutions? Because I'd *really* like to be able to trust at least some sources of media to not outright mislead me to false conclusions, and I'd like the ones doing that to be disincentivized from continuing.
The only way I think this can work in general is if we have media sources who have:
a. An incentive to get things right and convey them to you honestly.
b. A way to find out when they're doing that well/poorly.
c. A reason to care when (b) says they're not doing a good job.
The classic version of this is Vegas oddsmakers. They have a genuine incentive to get the odds right, they'll learn quickly if they're doing badly at it, and they won't keep their job if they keep making bad probability estimates.
My sense is that journalists are mostly not in the learning-reality-and-conveying-it-honestly business. That is, the best journalists care about those things because of an internal ethic or integrity (thus, the NYT tending to keep the contrary-to-narrative facts in the story but de-emphasize them), but I think the actual thing they're being paid for is writing stuff that brings in revenue--either from ads or from subscriptions. From the perspective of the NYT's management, a sequence of factually-accurate and honestly-conveyed stories that piss off a bunch of subscribers and advertisers is much worse than a sequence of factually-shaky and rather deceptive stories that make the subscribers and advertisers happy.
That is, I think a lot of our collective frustration with journalism is that we're mad it's not doing some other job we wish it would do. I would like NPR, the Wall Street Journal, the Washington Post, the New York Times, the Economist, etc., to make a genuine effort to understand what they are reporting and then convey it in a clear and accurate way, including links to raw data, unedited interviews, etc., so I can follow up. But I'm in a very small minority, even for those generally high-end news sources. I think a careful and accurate series of reports on what urban crime looks like might actually be financially bad for NPR, because it would upset the donors. They're probably better off running ideologically comfortable stories about historical racism and how it echoes through to today. (And indeed, I'm sure there are places where historical racism hurts blacks now, but I'm like 99% sure that the impact of historical racism on current-day blacks is a tiny fraction of the impact of crime on blacks.)
I think some people keep their audience by being careful with facts, because that's their audience. Nate Silver would probably lose his audience by becoming another writer of fuzzy-headed ideologically-reassuring stories. But I think they are the exceptions.
Think about it in terms of Silver's home turf, sports reporting. Some people want an accurate assessment of how likely the home team is to end up winning the playoffs, but most people probably really want to hear a nice happy story about the strengths, virtues, and heartwarming story of the home team, how they deserve to win the playoffs and will surely win because of their many great qualities. Telling the fans who want that kind of coverage that the odds are 4:1 against the home team is bad for business...unless you've built up an audience that actually wants the hard-headed stuff instead of the heartwarming stuff.
A lot of people go into journalism not to debunk bad ideas but to use their verbal skills to promote what seem to them at first glance, working under time pressure, as good ideas. Malcolm Gladwell is a striking example of this.
I wrote a number of critiques of Gladwell when his reputation was sky-high in 2005-2010, and no doubt played a role in the decline of his reputation.
But, my conclusion was that Malcolm's faults as an explicator of the real world were due not to anything contemptible about him, but instead due to likable aspects of his personality: he really admired college professors and their pet theories, so when he heard a clever new idea from one of them, rather than wanting to see if he could debunk it (as I tend to do), his natural inclination was to get enthusiastic thinking about how he could use his writing ability to persuasively frame this new idea so that more people would understand it and appreciate it.
Gladwell didn't have strong critical thinking skills or much of an urge to debunk, but then most people don't either.
I think that, in addition to media sources have your a, b and c, we will also need *readers* who have:
a. The ability to quickly visualize what claim(s) an article is making that the author believes are most important.
b. The ability to view what material in that article supports any given claim it's making.
c. The ability to check that material against other sources.
This is basically a first order logic theorem prover. And it's not necessarily a pie-in-the-sky contraption. (I spent years working on the precursor to such a thing.)
Part of the idea here is that different readers know different things, and such a technology could starkly illuminate the difference between what this new article is claiming, and every other article that reader has read. So, a "what's new" section that will report things like "the vaccine offered prior to December 8 is believed by the CDC to *not* be effective against variant 92.33.1411D" or whatever. It would tease out every referent, and even paraphrase the summary multiple ways. And the reader could even ask it questions like "does this mean I have to get another shot?" and it will answer either "yes, given these conditions" or even ask the reader reasonably simple questions, playing the role of an expert acting in that reader's interests.
Unlike GPT, such an "auto-reader" would be employing discrete symbolic reasoning logic, not a neural net or some Markov chain method. That discrete approach would have many other applications as well.
> you can fill your articles with sentences about how you’re not claiming X, and people will still find ways to accuse you of lying because you said X.
I think this is because if an article spends 5,000 words seeming to say one thing, and 50 words saying the opposite, people figure the 5,000 words outweigh the 50.
It's like the disclaimers on car commercials - the whole commercial is designed to say emphatically "drive fast! drive offroad! drift! buying this car will make people want to have sex with you! you will be cool if you own this car!" and then tiny text says that's not true. The ad is leaving people with a pretty clear impression, and it's not what's in the fine print.
Okay but in this case it’s more like “Scott spent 50 words saying 1 controversial thing (“the media rarely lies”) then spends 5000 words explaining the very technically precise way in which that’s true and why it matters” and everybody just wants to argue about the 50 words.
I think at some point, you have to accept that if the reader couldn't be arsed to read beyond the headline/title and first paragraph of an article, their opinion on the article probably isn't worth much. I mean, if Scott were being dishonest in the titles, that would be bad, but as it is, he's just making a precise statement that won't fit in the title.
If you "read" an academic paper by only reading the title and the first paragraph of the introduction, I'm not actually very interested in your opinion of the paper and its results.
I don't think the 5000-word controversial thing is the title, although that's controversial too, it's the clearly implied equivalence between infowars (total trash, never believe them), and the nyt (generally credible so long as you understand their neoliberal bias).
I don’t think it is charitable or reasonable to call something a “clearly implied equivalence” when Scott explicitly and repeatedly denied that he was implying that equivalence.
I'm trying to explain why people reacted how they did. A lot of people seemed to think the post spent a lot of time implying equivalence.
The way language and human beings work is if you spend a lot of time/space seeming to say one thing, and then briefly explicitly deny that thing, people will come away with the impression that you think and meant the thing you spent a long time seeming to imply.
Most of the post implies an equivalence between InfoWars and the NYT in the sense that it denies any qualitative difference between them. Briefly explicitly denying that doesn't undo in people's minds all the time they just spent seeming to hear the opposite.
You don't have to like it, but that is how people work, as evidenced by the comments.
You’re still being uncharitable because you’re saying Scott is actually contradicting himself with his denial. He’s not - the idea that he’s implying equivalence, that the point of his article is to say “InfoWars and NYT are the same” is a facile and superficial reading of what he wrote. The explicit denial should only be necessary as a prompt to the most uncharitable readers, but they ignore even that.
“NYT and InfoWars are different, but mostly in degree not kind” is the obvious intent, and that’s not “equivalence”
It's interesting that you're doing the exact thing reading my comment that I'm saying people did reading Scott's original post - you're responding to the emotional tenor of the bulk of my post, and ignoring the little caveats sprinkled throughout. E.g. you glide right over "seeming to say", or "equivalence ... in the sense that it denies any qualitative difference" and just respond to what it feels like my overall point is.
That's a very normal thing to do. It's how human beings read unless they're being very very careful. Which is my whole point. The original post was, presumably unintentionally, written in a way that led a great many people to misunderstand it.
It spent a lot of time *seeming* to say one thing, then briefly denied saying that. People are going to latch onto lots of "seeming" over brief explicit statements. Proof: all of the many many commenters who clearly did exactly that.
I'm literally just describing why, in my opinion, that happened. I'm not being charitable or uncharitable.
Prior belief e.g., "I have a prior belief that the entire staff of Infowars are idiots and/or insane so take every single word on their website with heaping tablespoons of salt"
It’s a Bayesian prior. Basically it’s how likely you are to believe X before some new piece of evidence about X updates your beliefs. You can look up “Bayes’ Theorem” for the mathematical context the term comes from.
The biases you had going in. I can read a perfect argument for why the world is flat, but my prior for "the world is round" is strong enough that I'm still not going to believe it even when I can't find flaws in it; I'm going to conclude it's more likely I'm too dumb to catch the flaws, than that there aren't any flaws.
You see a lot of arguments around here about "falling back on priors", mostly when an argument is not obviously wrong but still kind of weak.
Prior beliefs. This is used in roughly the Bayesian reasoning sense, which has a math definition, but approximately comes down to the idea that you started with an understanding of the world, and then saw some new evidence--your resulting beliefs at the end of that will be a mix of the starting ones and the new evidence, and will depend on the strength of your prior beliefs and the new evidence.
Suppose I have a lifelong friend I know to be a gentle and peaceful person. A local crazy lady claims to have seen him murder someone. My ending assessment of the probability that he murdered someone is probably still really low--the quality of the evidence is low, and my prior beliefs are strong.
By contrast, suppose the police then find the body of his ex-wife shot with bullets identified as coming from a gun registered to him, and a surveillance camera at a local convenience store appears to show him hauling a dead body in the back of his pickup truck the night she disappeared. At the end of that, my assessment of the probability that he committed the murder is going to be much higher.
Most journalism is hearsay (reporters telling readers what somebody else told them), but people still want it be admissible as evidence in the court of public opinion. I think that's behind people's dissatisfaction with the argument "this article didn't say X was true, it just said that so-and-so said X was true, therefore it's not lying".
Side note: I help fund a YouTube channel whose mission is producing documentaries about historical firearms. The main guy behind it is really, really serious about getting things right, but obviously he wasn't alive or there when e.g. Sam Colt was developing his first revolver designs, so he has to rely on books. To his horror, he's come to find that most books that claim to be authoritative on the subject, and that people treat as authoritative on the subject, don't cite any sources, have inconsistencies, etc. People have been repeating the same "facts" for decades as one book cites another book that cites another book going back decades upon decades and when he finally traces it all back, finds that the original source just said things without any kind of primary source documentation and presented them as fact. Or that the primary sources themselves are contradicted by physical evidence. It's got to be exhausting. The idea that anybody can really know what happened hundreds of years ago honestly seems ludicrous. (Edit: link to a video summarizing some of these woes here: https://www.youtube.com/watch?v=whYgOjAkC2g)
I long ago gave up on the idea of ever being able to actually know what happened about any remotely controversial event, e.g. Rittenhouse or whether COVID was a lab leak, because by the time it's become controversial enough for me to care what the truth is about it, there's already a zillion articles out there written by people with their own biases and agendas. Frankly, the world might be a better place if more people simply accepted that if they weren't there, they can't know what really happened. All they can really know is what people say happened. And even if you know what the people who were actually *there* say happened...I mean, would you trust anybody who was inside the Capitol on January 6th, 2021 to give a completely unbiased and factual account of what happened there? Do you even trust their memory of what happened, given that memory is imperfect? And yet whether the events of that day was an Trump-organized attempt at a coup d'etat against the very foundations of our democracy, or a just a bunch of idiotic protesters larping as revolutionary war heroes, seems really, really important to a lot of people. And so they want to be able to point to stories in the the news and say "see, this NYT article says X, so I'm right!"
The lab leak thing is complicated by the fact that everyone from some hapless lab tech who stuck himself with a needle by mistake and came down with a weird flu a week later all the way up to Chairman Xi has a huge incentive *not* to allow any evidence of a lab leak to come out, and at this point they have probably had the time to make sure that most such evidence is gone. That doesn't mean that it was a lab leak, just that it is very unlikely we will ever see evidence that it was a lab leak if it really was.
Yes. I meant to include some words along the lines of how it's quite common for the folks closest to the truth to have the least to gain for the truth to be known, but I'd already spent enough time working on my comment at that point.
[INFO WARS not shown, off the left hand side of the page in the abyss]
You aren't going to be able to reconcile those three models even if you're upfront about expected disagreements and load your article with caveats and very precise subtitles. I wonder how much of the technical points about the finer details of the denotative interpretation of the word "lying" is rationalization over a gestalt model of where the players fall on the truthiness spectrum.
>I almost always see people erring in the direction of accusing people too quickly, rather than too slowly.
On this note the devil is called “the accuser of our brethren” in Revelation 12:10. There’s definitely a Christian notion that eagerness to accuse is a major human evil.
Something that I was surprised to not see addressed is the distinction between media and written articles. Written articles are the easiest to analyze, but "the media" includes other, well, media. InfoWars is most famous not for the website but for the radio show. Fox News is most famous not for the website but for the TV channel. It may well be that the same analysis holds equally across all forms of media, or roughly equally across most forms of media, or it may vary widely. But I don't recall seeing this addressed anywhere in the various pieces by Scott or the comment sections, and I found the omission worth noting.
I’m in camp “This is a really important point, but framing it as a semantic question about lying is distracting people from the real argument being made.”
You aren’t actually making a nitpicky technical point, you’re saying that *for the purposes of censorship* any objective measure of truth would either be over-broad (and thus inconsistent in practice), or else so narrow to be useless. I think people are just getting distracted because it’s been framed as something much more pedantic.
This is something I'm not sure about. There are places where Scott seems to be saying this and there are places where he seems to be trying to make a broader point. I think that more clarity on the scope of the claim would help.
Now we have this taxonomy of dishonesty we can look at it from the other side. A taxonomy of levels of belief correlated with 1-7. I think someone who wishes for people to believe x may not have to resort to 7 or even 6 very often or at all. 4 and 5 can funnel someone into believing just as effectively in most cases. It’s a different question though. From the perspective of designing standards of truthiness, it might have to be requiring (like some states’ ballots) an argument, a counterargument and two rebuttals. I don’t know.
He is strongly suggestive that the point of this is about censorship. But I think that is not actually the central point of any of these articles - just a corollary he hopes we think follows from the claims here. (I actually think we should just be more comfortable with censorship of some things that aren’t literal category 7 lies.)
from my perspective, if you focus on censorship, then his definition is reasonable and useful, even if a little distracting.
If you look at it towards any other point, then it's just nitpicky and pedantic. This might be an example of the very common phenomenon where someone starts with a very specific point, but then drifts elsewhere while responding piecemeal to the specific comments of others.
But Scott framed the point about the media never lying as merely a lemma needed to reject a certain facile argument for bans on misinformation.
If he was a dead author then maybe we should reinterpret him but given he's an active participant in the dialog I think we should assume that's the point he wants to make and why he finds it interesting until he indicates otherwise. Besides, I think good Gricean norms mean we should assume that the speaker thinks the interest of his claim will be apparent to his audience. Without the initial framing it's unclear why we should care about the fact the media never lies under this definition. And some writers are obscure like that but Scott pretty much never is.
Why should we be more comfortable with that kind of censorship? Seems to me that the long history of censorship by powerful entities suggests we should have a strong rebuttable presumption against censorship of things that seem intuitively 'bad'.
I don't see how we could possibly have the kind of evidence needed to rebutt that presumption. Since the harms from censorship are usually of a sort that isn't recognized at the time we'd need a long time period in which we were comfortable with what you suggest that we still judged was desireable long after the fact.
Even at just the object level it's not even clear that the kind of censorship you suggest is even an effective way to decrease the bad beliefs in question. It's not even clear that the European laws about Nazi propoganda have increased of decreased the appeal of the far right (being forbidden adds cachet).
Not to mention the fact that the mere knowledge that certain views are being suppressed risks lending credence to all the worst kind of beliefs since they can (more credibly) just insist the reason you aren't hearing about it from reputable sources is the censorship.
I'm not particularly relieved to know after decades of reading that the NYT excels at "ethically" misrepresenting the world on certain subjects. For most things that don't touch the culture wars the NYT can write some exceptional articles where fact checkers do their job in a professional unbiased manner.
Otherwise ... they paint a not representative picture of the world primarily through selection bias and framing bias. Selection bias is rather obvious, excessive highlighting of the right's culture war flaws and the near dismissal of the left's culture war flaws. One is forced to read multiple news sources, you just can't know what is missing.
Framing bias is another art form that any skilled reader can easily spot lately. Which subjects are given the news analysis treatment and which subjects are allowed to have opinions from biased experts quoted liberally throughout the piece virtually unopposed. Expert laundering has become pervasive across the news industry partially leading to the decline of trust in experts. Some subjects are heavily investigated and others are ignored.
Subjects that don't fit a preferred narrative that cannot be ignored are usually given a "just the facts" treatment with a strict line on what are facts and what are not, as well as zero curiosity to ponder obvious questions out loud. No opining experts for context here. These storylines typically last no longer than a day or two.
Random example of common framing bias today, subhead on front page:
"The revelation is sure to intensify Republican attacks on President Biden, who has called former President Trump irresponsible for hoarding sensitive files.
It is not clear where or when they were recovered. But aides have been scouring various places since November, when a handful of classified files was found."
The story is apparently Republicans Pounce and the Biden team earnestly working hard to find all the missing files. Reverse the identity groups and these headlines will change. It has become very tiresome.
Sometimes group identity information is reported, sometimes it is not. When it is included in the headline or first paragraph it tends to be a preferred identity narrative, otherwise it is relegated to the end or simply not reported. For most news sources you can simply look at which anecdotal stories they choose to cover and that will highlight the preferred narratives with near certainty.
Knowing exactly where the industry's current ethical lines are and adhering to those slavishly doesn't stop one from being able to cleverly misrepresent the facts.
I guess people like accusing others of lying because there's a universal(?) human norm against lying so it's a quick and clear way to discredit someone. Saying someone “is grossly negligent in their reasoning and is happy to not investigate beliefs any further that are emotionally satisfying” isn't quite as pithy and there's not as strong a norm against this behavior. But it would be good if we had words for this.
Should we have a contest to propose words for these concepts?
If the target audience gets just as convinced by 2/3/4/5/6, they don’t have to attempt 7 very often (or at all).
If most journalists in a given newsroom would top out at 4, I think all you need is one editor or even a colleague who does 6 to start seeing effects in the writing. I think it’s slightly contagious, higher IQ is often (not always) more resistant to it, and the more it happens, the more the higher IQ people leave the profession. But I appreciate this taxonomy of dishonesty, it’s a helpful scale.
I'm sympathetic to your views on this overall, but I do think I'd categorically consider 6 to be "lying". At the end of the day, when we talk about why lying is bad, what matters is really the intent to deceive. "Lying by omission" is a standard phrase, after all.
On the other hand, intent is nearly impossible to discern with certainty, so I largely avoid trying to make any judgement calls on it. Accuracy rates and targeted bias are measurable enough that I can assume no one is actively trying to lie without it affecting my opinion on how I should respond to any given statement.
I disagree with the line-drawings on the Infowars claim in section 2. Infowars saying "the document is a shoddily contrived hoax" (if that's a direct quote from Infowars) is doing something different than what most media outlets do.
Most media outlets generally try to avoid explicitly saying false things. They are willing to do various other things that are misleading, like presenting information in distorted ways which suggest a false conclusion, or abridging a quotation and altering its context, or citing a false claim that someone else made in a way that suggests that it's true. And they sometimes make mistakes and accidentally say false things, e.g. when they're confused (often in an ideologically convenient direction) or when they get sloppy and fail to find artful phrasings that avoid making explicit claims. But they'd try pretty hard to avoid calling an accurate document a "hoax".
If this discussion is trying to offer models of how the media does and doesn't distort the truth, then IMO there is a joint here where things should be carved, putting this Infowars stuff on one side and most of what you'd see in the Washington Post or whatever on the other. And that seems more relevant than line-drawing around the word "lie", and separate from trying to weigh in on how much we're bothered by different sorts of distortions.
The defense of including "the document is a shoddily contrived hoax" in the same category as everything else retreated to the idea that whether something is a lie depends on the mental state of the person saying it, and we can't know for sure what the author believes.
But that's steering the discussion away from trying to actually model what the media is doing. Plenty of people believe that Epstein didn't kill himself, and it would not be at all surprising to see a reddit post that asserts "Epstein didn't kill himself." But it would be surprising to see a Washington Post article that asserts "Epstein didn't kill himself."
Even though there are probably reporters and editors at the Washington Post who believe that Epstein didn't kill himself. Because the Post, and most major media outlets, are doing something different than what Redditors do. By asserting a claim about the world, like "Epstein didn't kill himself", the media outlet is engaged in *reporting*, and that carries with it a process of 'trying not to report inaccurately' by checking if the claim is true / what it's backed by. And 'the author believes it's true' or 'there exists an argument for it' are insufficient.
Possibly dropping the word "lie" from this discussion altogether would be an improvement, because the question of whether the author might have believed what they were saying is more relevant for arguing over whether it technically counts as a lie than it is for talking about how the media usually works and out-of-the-ordinary an example like this is.
I’m not convinced that the 9%/7% inflation thing is actually false. Reported inflation rates are usually not point-in-time estimates of the current annualized rate of change of prices, but are usually instead reports of the year-over-year change from 12 months ago. If the yoy change in one month is 9% and the yoy change in the next month is 7%, it seems quite likely that prices actually *did* come down in the intervening month. But it would take a bit more digging to be sure.
Off the top of my head, there were already talks of dangerously accelerating inflation last December (at least in Europe), so the argument that the Dec 2021 baseline is about the same as the Nov 2021 is at least questionable. I know for Germany that prices did indeed come down, but only by -0.4% MoM (give or take) in the face of a -2.x% YoY decrease, suggesting a fairly high MoM increase from November to December 2021.
I think the miscommunication here is that when talking about whether a statement "A => B" is true, for some people that means "given known and fixed truth values for A and B in the real world, is the statement true", and you make the point that it's true because B is true (and A, too, but that doesn't even matter).
However, other people might treat it as true only if it is universally true for all A and B because B inherently logically follows from A. And under this interpretation, the statement is fal.. not universally true, because a price decrease cannot be inferred from shrinking YoY inflation rates alone.
I agree however that it's fair to expect media to be in the business of reporting facts in the real world rather than mathematical reasoning on an abstract level.
It is literally true that prices declined (not just increased at a slower rate) in December, as measured by e.g. the CPI index - Federal reserve series here: https://fred.stlouisfed.org/series/CPIAUCSL . So the commenter who claims it is a lie is instead mistaken. This is of course another vector by which people might think the media lies more than it does.
EDIT: Looking back the commenter's datestamp was 12/29/2022. Not sure what article they were looking at, and not sure what month's data was being reported. That said "judge is incorrect" (like me, in this comment judging another comment, maybe!) still stands as perhaps additional way the media might be perceived to lie.
I think some of the confusion might be between "not lying" and "we don't know if they're lying or not, because we don't know what they really believe." We begin in ignorance of other people's beliefs and often don't learn enough to be sure either way. But if you say "not lying" then it sounds like you're making a claim about their state of mind, when often we don't know that.
Maybe it would be clearer to say that proving strangers to be lying is hard.
The story about Jones and the radioactivity to me isn't all that clear-cut, maybe actually supporting Scott's point.
If Jones just wanted to push some false claims about radioactivity levels, why go through all the effort? Why send a team to California, why use real Geiger counters (somebody must make fake ones for movies, right?), why not use some legal, naturally radioactive substance to deliberately create a fake measuring spike (in relative terms, even if not harmful), why take the risk of live reporting? Given that they were absolutely not set up to produce fake reporting, Jones screaming at their lack of finding something suggests that he seriously expected another result and irrationally didn't want to update, resisting it quite ferociously. Otherwise, why not accept that there's nothing to then instruct them to just make something false up, rather than screaming and telling them to merely *not* say that they didn't find anything (which, for all he knows, might well be because of $other_conspiracy_theory).
Did the reporters eventually report something which they knew to be a lie? The quoted excerpt actually sounds like they *didn't* because they "couldn't stop [saying they weren't finding anything]", but I'm not familiar with the original story. Either way, it sounds like a reporter not being willing to claim evidence they didn't find, and his boss being furious at him for not finding evidence he is certain should be discoverable.
Jones has certainly deliberately pushed lies in other contexts like Sandy Hook, but if the above story really is the best known example, I remain unconvinced of the prevalence of that being exorbitantly high.
I recall Fukushima fairly well because I am a night owl who was up late as the news broke from Japan. Mainstream news media headlines tended to run something like this from Day 1 to Day 4 (note, these are just semi-comic reconstructions of my memories, not quotes):
1. Fukushima Nuclear Power Plant Swamped, but Experts Say No Radiation Danger
2. Building Blows Up at Fukushima, But Danger Minimal
3. Second Building at Fukushima Explodes, But Risk Is Limited
4. Blast Obliterates Third Fukushima Building
As it turned out, California wasn't drenched in radiation from exploding buildings, but events were not reassuring at the time. You didn't have to be an Alex Jones fan to be worried by this progression of events.
From the anecdote, it sounds like Alex Jones spent real money in the hope of getting a real scoop, which speaks better of him than I would have guessed.
It sure seems like a news organization that cared could have gotten together people with the right expertise to make pretty accurate predictions about the consequences. My assumption here is that they didn't because they didn't care that much--panic was a better response for them in terms of ratings, and they had no incentive to make accurate and boring statements like "yeah, this is probably pretty unhealthy for the people trying to stop the crisis, and I wouldn't go fishing right outside the plant and eat what I caught just now, but it's almost certainly not going to affect anyone in California."
Making accurate predictions is not their job, and is hard and takes work and knowledge they mostly don't have, so they don't do it.
The mainstream media kept implying that there was little to worry about at Fukushima, and then, three days in a row, buildings blew up. I was surprised that the MSM was downplaying the risk from a nuclear power plant disaster, but that was the general tone of coverage.
Personally, I found this pattern worrisome at the time.
Fortunately, after awhile, the nuclear power plant finally stopped blowing up.
I posted the excerpt (and Scott linked to the full NYT story). You make some great points.
For background, I think IW consists of a spectrum from true believers to psychopaths. The latter prefers to keep their lies deniable. Pointing to a specific statement as a lie basically requires a glimpse behind the scenes (which aren't exorbitantly common).
Here's my model of the situation:
TL;DR Jones sent out true believers to California. He wanted them to lie, but they didn't understand (or at least, would not lie without direct orders).
In terms of procuring legal radioactive substances or fake Geiger counters- I think that would be legitimately difficult. There were significant time constraints, and these aren't clever or well-connected people even if someone did think of it. (And I suspect that an InfoWars employee attempting to procure radioactive substances "ASAP" would be denied and flagged regardless of quantity.)
In terms of why the employees were sent out live in the first place: I think the strategy has been successful before- it's easy to convince an audience that the person in front of your camera is a baddie, even if it's live. But this time two true believer-ish employees were sent out to check a machine's reading of the air. I think what happened is "Mistakes were made"- Jones and upper management simply didn't consider that perhaps the camera crew wouldn't have the means or willingness to fake a dangerous situation.
As to Jones' behavior- I don't think he breaks keyfabe very often. He wanted them to figure out how to fake the result, or if nothing else, pretend to be run off the road by feds. (The employee in the quote later mentions Jones doing the latter.) But usually this is communicated via "Alex is mad at you, make him happy." With two (true-believer) employees in another state who didn't realize what their task was, he wasn't willing to risk directly ordering them to lie. Doing so would mean realizing he's a fraud when he isn't there to manage the situation.
On the “fake news” thing - I do recall seeing a few “articles” on sites like weather737.com, claiming “meteorologists forecast largest blizzard in history to hit North Carolina in 12 days”. Totally outrageous, beyond any reasonable forecast, but coming soon enough after a polar vortex event and about a date far enough in the future that it will have lost social media currency by the time it is checked. I don’t exactly know why these briefly proliferated (probably some time in the 2014-2016 era?) but then disappeared, but it was a fascinating phenomenon, and not only about political news.
I worry this definition of a lie is so narrow and so unprovable that it's a bit of motte and bailey when taken back to the original justification (that censorship attempts are ill-advised since the media rarely lies) and weakens that argument. The call is to censor misinformation or disinformation or 'fake news' and that is not equivalent to censoring lying if you're going to use a such narrow definition and is likely much more straightforward (if imho still ill-advised).
In particular, your definition of lying leans heavily on intent which is obviously pretty hard to conclusively determine. Short of leaked chat logs saying 'ha, ha I'm totally making this up' from the 'We Are Lying Liars' chat room it's going to be difficult to prove they are lying even if they are (though if you follow Matt Levine this seems to totally be a regular thing in finance so perhaps the absence of regular leaks of this nature should say something on earnestness / self-delusion of the journalist cast). I think the a better title for the first piece is 'The Media Can Vary Rarely Be Proved To Be Lying' rather than 'The Media Vary Rarely Lies' which is in some sense is unknowable for many cases using this very narrow definition of lying and probably works just as well for your argument.
But back to the censorship point, the steel man case probably doesn't take intent into consideration at all. But rather seeks to ban factually incorrect statements (whether by malice, incompetence, bias or otherwise) or mandate the presence of certain context on particular issues. And that seems potentially possible and orthogonal to whether the media lies to you in some very strict sense of lies. We have (flawed) social mechanisms to determine is something is proven to various degrees of certainty and use them in critical places, courts for one example.
So, there's been a huge amount of discussion, what the word "lie" means, which is very hard to resolve because it's a fairly vague term and even more to the point, it's an emotionally loaded word with accusation and anger as part of its meaning. Not only that the discussion was focused around some material that's already very emotionally charged, like Alex Jones vile attacks on families whose children were murdered. I think it's kind of impressive that the discussion has been relatively sensible given this context.
With regard to your conclusion, Scott, I think it's still a bit superficial. You seem to assume that finding some other source for what you're saying makes it a different issue than outright making things up. Tell that to an AI that turns racist reflecting its input. And we humans often make up our minds about things for non-rational reasons then construct rational sounding arguments to support what we're saying, especially when talking politics, especially the sort that's basically an emotional minefield.
You seem to dismiss the idea that Jones decided it would popularize him to come up with an attack on the Sandy Hook victims, then looked around to find some flimsy stuff that could be used as his "source", so he couldn't be accused of making up what he was saying. How is that different than making it up, except to give him a way to claim he didn't make it up. It's very easy to "support" pretty much anything you want to say this way It didn't take Musk long to find a specious source of really nasty stuff to retweet about Paul Pelosi, and was that different in any meaningful way than his made-up "pedo boy" crap?
I don't see how making distinctions between between misleading statements supported by the flimsiest of sources and making up facts is especially meaningful, in a context where you can find a tweet or a posting to support most any position you want to take.
But also, a story composed of made up facts can contain truths that can't be told as effectively any other way. (I guess that's why you make up your allegories.) A factual account is always going to be subject to bias and error. It's how humans communicate. People communicate to help each other understand or to mislead each other, among other reasons. To me, distinctions between made-up facts and misleading interpretations ultimately don't really make anything much clearer, especially when we're talking about communications with content that's basically emotional rather than factual.
I've agreed with all your posts on these topics. I find myself not motivated to post when I agree.
7 is lying
5 and 6 are being deceptive
3 and 4 are being incompetent
2 is forgiveable
I find 3 and above bad enough, that once we get to 7 I just don't care very much. My starting point is that the news media I want to consume should be trying to convey the truth, and anything past 3 is a failure to even *pursue* that mission.
I think some of the frustration in comments is that it seems like your standard for the media is "they aren't doing a really bad thing", but many people's standard is "are they doing a a good thing".
And if Scott had written an article called "Doctors very rarely murder people", commenters would be missing the point if they replied with lots of examples of doctors failing to cure people.
More importantly, it's not the case that Scott is being overly lenient or complimentary towards the media. His point is that even when the media do these clearly bad things, they're not technically lying, therefore it would be extremely hard to come up with a set of laws or censorship algorithms that prevent the bad things. He's not disagreeing with you that the bad things are bad.
The question is around intent. Doctors very often kill people (medical mistakes kills 10s maybe even 100s of thousands each year). But do they murder them? That requires intent and definitions vary, is operating drunk murder? While ill and sleep deprived? After a bad breakup? Civil and criminal courts and judicial codes have a range of intents and responsibilities and muddle through. Scott's using a First Degree Premeditated Murder as his standard of lying, and sure fine. But I don't really care if my Dr. murders me or just kills me tripping on LSD or something and the more defensible misinformation censoring is more akin to mandating that Dr.s not be high and pilots have 8 hours of rest than outlawing murder.
You as the dead person may not care whether you were murdered or accidentally killed, but the rest of us care very much. A person who kills by accident will probably feel remorse and will take care not to repeat the act, since he has no reason to do so and probably good reasons not to. But a person who murders on the other hand is much more likely to repeat the act -- since he obviously found a reason to do so that was not sufficiently restrained by conscience and strong social disapproval.
Such a person is a far greater danger to the rest of us than an accidental killer, even though for you (one victim) the difference is academic. That's why the law puts a very, very heavy stress on intent. Intent tells us how likely this person is to be a continuing or greater problem.
Banned for this comment. I think I might have posted a link to an article about why this was false, but even if not, accusing me here just seems weirdly aggressive.
Why should we care if the media doesn't lie under your definition of lie? How does that relate to the original argument claiming to discredit the facile argument for blocking misinformation?
I thought you wanted to argue that you can't say that such bans only bar clearly valueless content because they would inevitably have to ban content that isn't lying.
But with this narrow definition of lying why should we assume that you can't only ban clearly valueless content (say the weekly world news) just because it's not lying under your definition?
I ultimately agree there are problems with such bans but I feel like the debate has gotten a bit too focused on whether the media lies to the point of losing sight of the original reason this claim mattered.
First, to be clear, I meant why should we care in the context of the argument Scott was making.
But on the general point I suspect you care about how the media behaves not whether the meaning of the word lie is such that it accurately describes their behavior.
If you don't give any special weight to the word lie then there isn't really anything illuminating about media behavior in Scott's claim. Scott could have conveyed the same content about media behavior merely by saying: the media almost never make assertions in their own voice and even the most misleading claims are usually written as a literally true statement that some other entity made some claim. Very few people would have doubted that claim (whether or not that's literally never outside of disgraced reporters or virtually never doesn't seem very important)
The only reason it seems like some substantial claim was made is because of our prior associations with the word lie. Once you simply take it as a stipulative definition then Scott's claim doesn't really have much substance.
> I’m kind of sensitive to this because for almost every article I write, people in the comments accuse me of lying, or “pretending” I don’t know why the statements I made are wrong, or some other offense which I plead innocent to.
You asked for that treatment. Once you've made statements like this:
(B) Huh, it sure is weird that all the research on racial differences in intelligence says one thing while the official orthodoxy says something completely different. Oh well, what are you going to do? (Early ACX; not sure how to find it)
Not to put too fine a point on it, but if Steve Sailer is persistently right about some important issues that most media sources are persistently unable or unwilling to report well on, why does that lead you to not cite him?
It's a difficult conundrum for Scott. He wants to tell the truth and thus do good for humanity, but he knows that likewise I've been telling the truth for decades in order to try to do good for humanity. But I mostly inspire mindless rage and increasing ignorance and stupidity. So Scott tries to subtly obscure and bury the turth under a torrent of words so that only the bright will understand while the less intelligent will assume that, while they can't quite follow him, obviously he can't be like that horrible hateful Sailer.
> I asked “Why? A woman said she had been raped, and Rolling Stone believed her. The woman was making it up, but Rolling Stone wasn't” and Deepa commented “Isn't it the job of a reporter to investigate? And be good at it?” I don’t want to pick on Deepa, but this is what happens when you have an overly expansive definition of “lie”!
On a psychological level, it's plausible this is an "honest mistake" on Rolling Stone's part.
On a practical level, 1. it led to them writing a widely-disseminated article that made very many people believe a straight-up falsehood, and 2. it is pretty obvious that this failure of due diligence was in service to advancing their ideological agenda. (Compare how Rolling Stone would treat e.g. a story about a trans woman assaulting someone in a women's bathroom. In this counterfactual, despite being very similar on the object level, it's obvious they would do their very best to either debunk the story or bury it; there is no world where they credulously run a story like the Jackie one.)
You say:
> When someone says “Joe is a liar”, I don’t want to have to ask every time “Do you mean you have some actual evidence for this, or that they said something you disagree with and you instantly leapt to ‘this is reckless disregard for the truth because nobody could ever be so dumb as to honestly disagree with me’?” I think if we let people use the word “lie” this way, then the overwhelming majority of accusations of lying would be false. Why would we want to define a word in a way that dooms it to constantly be used incorrectly to mislead people?
But this is fundamentally missing the point. The main inference people are usually trying to convey by saying "X is a liar" is something like "X is an utterly unreliable source of information and should not be given any credence". This is a statement solely about X's claims and their relationship to the truth; it need not rely on any inference about X's state of mind.
It may be psychologically interesting to dive into the question of whether X is knowingly making false statements, or knowingly deceiving, or merely speaking with reckless disregard for the truth. But from the point of view of the information consumer, these distinctions don't matter. All the reader needs to know is that X is grievously unreliable.
In practice, almost all media outlets pass the bar of being grievously unreliable as information sources. If you're the priest or the therapist of an NYT reporter, you should make the careful distinctions about the exact psychological impetus behind their deceptions. But if you're merely a reader, the important thing to take away is the fact of their unreliability.
An audio tape of Rolling Stone reporter Sabrina Rubin Erdely's conversation with UVa coed Jackie Coakley was released during the libel trial. The funny thing is that the student is clearly the alpha female of the pair, dominating the conversation with a steady flow of BS, while the 40ish journalist chips in with a few nerdy true believer factoids about how studies show that rape on campus etc etc that Jackie then runs with.
A really interesting historical work could be written about the collaboration amongst activists, journalists, and the Obama Administration in the years up to 2014 that laid the groundwork for the notorious Rolling Stone fiasco. It had been pretty clear to me from 2013 onward that the New York Times and the White House were conspiring to make Rape on Campus into a huge crisis. But nobody seems very interested in documenting afterwards how these type of Democratic Coalition campaigns are organized.
It seems like there are two things that went wildly wrong with this story (though note that I didn't follow it beyond reading the original story and some online discussion as it unraveled):
a. Erdely should have been a more skeptical observer of the story. There were some obviously implausible bits (broken glass), and it seems like even minimal checking was enough to discover big holes in it. So why didn't Erdely chase down those details and realize there was a problem?
b. Rolling Stone's fact checkers and editors should have been a second line of defense against RS publishing a story that would end up blowing up in an embarrassing way that damaged their reputation and got them sued.
In Erdely's case, maybe she was just totally convinced by Coakley's story and mannerism and believed her. You can imagine how a lot of cultural messages about believing women and rape culture on campus and such might have contributed to this effect.
Alternatively, perhaps she thought that the details of the story were unimportant because it was conveying an important truth--that campus rapes are a serious problem, that frat boys are bad, that UVA is full of a bunch of dirty outgroupers who deserve bad things, etc. Certainly, a huge amount of the support the story got in media and online before it fell apart hit those points pretty hard. How dare you doubt that this friendless old lady I just grabbed is a witch--are *you* a witch, too?
But it sure seems likely that she basically knew this story was full of holes, but thought that nobody would ever check deeply enough to disprove it, and that she'd covered her ass sufficiently by having a source she could say she'd accurately quoted. That is, she may have been lying in one of the ways Scott says media do sometimes lie--by quoting a source saying something that's pretty implausible and full of holes, but then not engaging in any further questioning or thought that might mess up the story. I think Megan McArdle used the term "too good to check" when describing the story initially.
The first motivation doesn't apply to the editors/fact checkers, but the last two might. Perhaps they thought "this is Socially Important" and "one does not doubt the word of a woman who claims to have been rape, that's just Not Done." But also, perhaps they thought "this is going to be a very big story, we'll win awards--so maybe it's a little shaky, but who's ever going to dig enough to find out?"
Did enough information come out in the lawsuits and such later to untangle which of these is more likely? If Erdely and her editor/fact checkers were reasoning along the lines of "this story is probably made-up but who's ever gonna know?", that implies that probably a lot more big stories along these lines were similarly shaky, only nobody every investigated further and so the full-of-holes story just became accepted as truth everywhere. That certainly has happened in a number of high-profile stories over the years, but I have no good intuition for how common it is.
The Wikipedia entry on Erdely is also interesting, highlighting two other news stories about rape victims, one of which also appeared to have major problems.
An important thing to remember is that the Rolling Stone "Rape on Campus" story was publicly admired by many leading journalists, such as Jeffery Goldberg of The Atlantic, who went out of their way to tweet congratulations to Sabrina Rubin Erdely. It was the Spirit of the Age. Hadn't we been running articles for 18 months in support of the Obama Administration's "Dear Colleagues" letter to college administrators about the frat boy rape threat?
Five days after it was published, Richard Bradley, an editor in the financial press, became the first professional journalist to doubt it publicly, and I was the second four days after that. That seemed to open up the floodgates for skeptics. But that was a long time before more than two journalists doubted this celebrated story.
Most of the journalists who retweeted the article appear to have reasoned: Of course blond fraternity boys at the U. of Virginia are smashing glass in their rapine and hitting the poor source on the head with a thrown beer bootle. That's what blond boys do: they hold kristallnachts.
Ironically, the Rolling Stone hoax led to an actual kristallnacht on campus with Woke students smashing the windows of a frat house.
Here's my 2014 analysis of Erdely's absurd article.
The main thing I've changed my mind about is that in 2014 I gallantly assumed Jackie Coakley was some poor girl in need of protection from exploitation by Rolling Stone. But my impression now is more that she's a George Santos-type who likes concocting tall tales about herself and that she exploited poor, naive Rolling Stone with her ridiculous story lifted in part from "Dawson's Creek."
Heavens, redefining "liar" to mean "source of unreliable information" is quite a big jump. The Internet would become a very quiet place if we were all to think that way.
Why would you say that? Quiet requires that the speakers stop speaking, not that the listeners stop listening (or believing).
In any case, the admonition used to be to "not believe anything you read on the internet"[1], and I'm not at all convinced that it does not continue to be sound advice. Insofar as my question remains "whether I should update my model of the world based on this information?", the sort of distinctions being discussed here are a pointless distraction.
It does not matter whether someone is technically a "liar" under some stupidly pedantic definition of the term, and never has. What matters is "can I expect the received information and proffered conclusions to be more reflective of the truth than not?", and a taxonomy of the various ways of answering "no" to the foregoing is more of a hinderance than a help.
It's not even particularly helpful to considering the issue of censorship, because the key issue there is "can I trust you, who will be doing the censoring, to only censor the bad stuff and not the good stuff?" Which can instantly be answered in the negative for all values of "you, who will be doing the censoring", "bad stuff", and "good stuff".
If you *really* want, the matter can be reasoned out in greater detail, but I find that "Fuck no!" is a perfectly adequate argument in this case.
---
[1] We're talking about the later '90s, when the various cranks and conspiracy theorists had already hit the net big time, but mainstream media still mostly functioned in the physical realm
I'm pretty convinced that if the listeners stop listening, the speakers will stop speaking. Very few people will put the energy into speaking if the evidence suggests nobody is listening.
Also, I suggest you are completely wrong for the general population (i.e. those other than yourself) about the importance of whether someone can be called a "liar" to general approbation or not. As evidence I would adduce the heated conversations which you can see have formed right here on the subject -- clearly most people *do* care about this, a lot.
The reason they care is because "liar" is highly pejorative. It's the difference between saying a woman is "sexually confident" and saying she's a "slut." One will get you punched in the nose, the other won't. So when people debate "When is it OK to call so-and-so a liar?" they are debating (in addition to the named topic) "When is it OK to say so-and-so is an Untermensch, an undesirable, someone we are justified in socially shunning, mocking, subjecting to both informal and formal social penalties?" It's a big deal. The distinction between saying "He was wrong/misinformed/bullshitting" and "He's a liar" is very strong.
Consequently, people who permit the word "liar" to be applied to people who are merely mistaken, or reasoning poorly, or even bullshitting, will find themselves increasingly socially isolated, since that's not what everybody else means by it, and it would be as offensive and likely to end badly (for the speaker) as replacing "sexually confident" with "slutty" in college classroom conversation.
When I took an intro semantics linguistics class in collage, I was told, bluntly, that there is not and cannot be a single satisfying all encompassing definition of either "a lie" or "lying," and that the best we can do is jargon that covers restricted cases (e.g. perjury, defamation). Otherwise, we're stuck with the unsatisfying conclusion that the concept of lies and lying is a broad category with a clear center, but fuzzy boundaries similar to the concept of a heap or flatness. This dogma also came with a whole host of interesting edge cases, for example:
A man lives in a home on a suburban cul-de-sac, his neighbor asks to borrow his snow blower. The man, knowing that his neighbor had destroyed his possessions in the past and not wanting to loan out his snow blower replies "I no longer own a snow blower." This has everything you could want in a case of lying: 1) a statement that is false, 2) intent to deceive, and 3) creative genesis of a narrative (the "making" part of "making stuff up"). You can change the story to remove each element: what if he thinks he has a snowblower, but his wife sold it earlier in the week without telling him? What if he just hallucinated that he didn't have a snow blower? What if he transferred ownership of his snowblower to his daughter? Any of these would be enough to definitively make it "not perjury," at least one of them would not be enough to avoid "defamation," but the charge of "lying" isn't so easily or cleanly evaded.
I'm not familiar enough with the state of the field to know whether this is indeed the dogma of the field, let alone to know if it's justified. But I do know enough to know that a lot of good work has been done on this project and you don't need to re-invent all of it yourselves if you don't want to.
So if we stretch out the time frame to be the past two years we get 8% implying that every single person who died died of the vaccine! We could have stoppered death forever!
Okay but this also makes it implausible that they just incorrectly attributed post-vaccine deaths to the vaccine - there weren't enough post-vaccine deaths!
If your cousin tells you on New Year's Day that her cousin told her on Christmas that her cousin's cousin told her on Thanksgiving that her cousin's cousin's cousin died from the COVID vaccine, how many relatives are we talking about?
As a rule of thumb, 1% of Americans die each year. (That' why our life expectancy is ~100).
If each of them has 4 relatives, then we get 4% of the population having a dead relative. That's probably about right if you count children, siblings, and perhaps(?) grandchildren, but nothing else.
As soon as you count uncles and cousins and perhaps even more, the number will be a lot higher. Especially because people who are old now often have/had a lots of siblings.
People very often believe false things after they read news stories. This is by design. Naive readers think that news stories are credible without having a good understanding of what it means to be credible, and willfully ignorant readers don't care if they believe false things. The act of "staying informed" has positive emotional valence for these people, but they would believe fewer false things if they stopped reading the news.
Many people are angry about the status quo, and they're trying to convince everyone that the news is not credible, whatever that means. They want "staying informed" to have negative emotional valence. They interpret this series of blog posts as a defense of the status quo, which should not be defended. They don't believe that careful caveats about nitpicky technical linguistic questions can change the directionally incorrect impact the blog posts will have on emotional valence.
This is a bravery debate. Scott is technically correct about everything, but nobody cares about that. They care about emotional valence and downstream consequences. They care about how common it is to believe false things.
(This is just one perspective, and it's oversimplified, but I think people need to be more aware of this dynamic.)
You just explained my emotional response to these posts that I've been having trouble expressing. "Lying" is worse than "not lying," and so a blog post centered around the idea that "the media rarely lies" feels like an undeserved defense of a frequently deceptive media, regardless of what Scott's intentions actually were. And his defensive follow-up posts sound a lot like "what? I said that I didn't come to praise Caesar and that Brutus is an honorable man, I don't understand what you're complaining about."
I agree with the broader point Scott is trying to make, and I can't really find technical fault with anything he wrote, but it's framed in a way that seems to be making excuses for dishonest people. Even if he's not, really. He's still providing a level of respect and decorum that I don't think the media really deserves (they may not be 100% forthright all the time, but at least they're not liars). I know I'm probably being unfair to Scott, but his continued insistence that "the media very rarely lies" isn't a terrible headline almost guaranteed to be misconstrued is frustrating. If you're going to use a misleading clickbait headline for your article, then you don't get to complain when people ignore your buried lede and spend their time reacting to the clickbait headline. You reap what you sow.
I think one of the problems with this discussion-- which I'm sure came up in the comments, but I haven't looked through those-- is that the title used the word 'rarely', which means that by definition Scott and the people in the comments are going to be talking past each other when people in the comments bring up examples of what they think are 'lies'.
Yeah, I think most of your readers have agreed with you all along. Your critics are what make you stronger. Anyway, I want to carve out a category:
2.5) Reasons well but is dumb. Dumb in this use means uninformed. I don't know the true facts. That would be my category. Oh and after dumb, whatever bias I have takes over.
I've stopped consuming the news. People at work will tell me if something important happens. And I just focus on local stuff, work, my kids, the house, cars, neighbors, friends...Oh and the neighborhood here online at ACX. It's great, thanks. (Oh I follow weather and local sports on the news, and getting politics into sports (and weather) is something I hate.)
I have found the reactions to this series of posts bewildering. You're stating an extremely simple premise, and for some reason you have comments full of people attacking...something else. I would expect readers of this blog to share a level of clarity of thinking and basic literacy, and yet you have numerous comments throwing out counterexamples that are unrelated to your central point.
I'm not trying to say people commenting are dumb and wrong, because that seems unlikely. But reading these comments I genuinely don't understand what is going on in the people's heads. (I'm thinking specifically of comments that are like "oh yeah but look the media once reported that person X said Y and this wasn't true so the media is lying". Or even "You're saying the NYT is as bad as info-wards.) Again, non-judgemental, I'm just clearly missing something about how people's brains work that is causing this behavior.
I think most of the people commenting in that vein go into this discussion with very firm beliefs that they have caught the media lying, and not infrequently so. I therefore would just call the phenomenon you observe "update resistance", which is how people's brains work most of the time, as I learned from reading ACX and SSC.
Of course one must be different to not blanketly chalk any disagreement up to update resistance - some of the cited complains do convince me to some extent that Scott's definition is so pedantic that it's borderline useless, many others however don't seem to be convincing at all because they sometimes even inadvertently prove Scott's point (see my comment on the Alex Jones thing). So I definitely see your point.
I think this is definitely part of it. What confuses me most though is the fact that the counterexamples don't pass the bar Scott established. Maybe they already classified those things are lies by their own definition, and didn't recompute the label with Scott's when they posted them? I think that could explain some of it.
Unrelated, I think Scott's is sort of my default day-to-day definition of lying--it would never occur to me to call any of these other things lies. So whether or not its useful, I would say it's not uncommon in my experience.
He's not directly saying the NYT is as bad as Infowars, but he's putting them in the same category for reasons that are not really clear. And people are reasonably upset about the implication and suspecting he has an axe to grind.
(Like, this article is literally about how someone with an agenda can make technically true but misleading statements that imply your desired conclusion, are you really shocked that people will be looking at what the article implies rather than just what it literally says?)
"He's not directly saying the NYT is as bad as Infowars, but he's putting them in the same category for reasons that are not really clear. And people are reasonably upset about the implication and suspecting he has an axe to grind."
I don't think that's reasonable! This seems like a decoupling thing, right? It's fine if they're in the same category or share traits or whatever, the same way that it's fine if you talk about the fact that Hitler loved his dog. It doesn't mean you like or support Hitler, it's just a true thing about the world. I'd expect people who don't think-about-thinking much to get upset if you say nice things about Hitler, but I wouldn't have expected that from this blog's readers.
If someone was specifically trying to draw attention to how both Adolf Hitler and Joe Random own dogs, I think it would be reasonable to suspect they have a problem with Mr. Random. Sure, there's nothing wrong with owning a dog, but you still have to wonder why they think it's important for us to draw a comparison between Mr. Random and Hitler.
(The fancy name for this is Gricean implication - bringing something up in a conversation inherently implies it's relevant.)
In this case, the reason Scott said he's putting Infowars and the NYT was to argue that there was no principled way to distinguish "the sort of lying the NYT does" and "the sort of lying Alex Jones does." But many people (including me) disagree. Even without a 100% objective measure of the truth there are many different axes on which you can differentiate those types of lies, and saying that they're both the same type of lying is ignoring those distinctions.
I think this is a good point, and I agree with the first two paragraphs (though not your third). But what is weird to me that people argued against it so _badly_, like bringing up examples that obviously played into Scott's hands.
What I'm settling on as the process is this:
* People read the thing
* The social implications of the thing but not the object level thing make them uncomfortable
* They feel they need to provide counterarguments, but aren't exactly in touch with what is prompting this, so they come up with arguments that are just sort of a mess
This is not to say that's 100% of what happened, but it does explain some of the more bizarre counterexamples.
My prior on Scott having an axe to grind with the NYT is not exactly zero, but I don't necessarily think it's fair to suspect this based on the post in question. What he's doing is perfectly in line with his philosophy from even the SSC era, before the NYT thing.
Most people have a gut reaction along the lines of 'Clearly, Infowars is waaaay worse than the NYT', but gut reaction isn't good enough from a rational standpoint, so let's analyze why we think that is - *without* relying on the conclusion that Infowars is made by despicable humans and only read by deplorables.
(My memory is really bad so I don't remember the title of the SCC post, but there's this one about 'other people [not] getting "the thing", where "the thing" is some understanding that the person of the opposing faction is willing to fight just as hard and is just as convinced about their positions as you are about yours' - possibly butchered it. But I think that's where he's coming from when he consciously de-emphasizes judging the messaging/narrative of the respective publication.)
And then we see that, yes, maybe NYT has better fact-checking, yes, maybe NYT has a higher bar of who they are willing to regard as a 'credible source' (didn't work too well in the Russian collusion story, but that was not just NYT). But most of this is quantitative - we'll have a hard time making a qualitative distinction in the sense of "Infowars does X (which is egregious), and NYT has never done and would never do X".
Some people may say that the quantitative difference is so large that it becomes qualitative. Others might say that the distinction comes from the NYT having good values and Infowars having obviously bad values. These distinctions may work for you, but from my perspective it's fully in-character for Scott to dismiss them as too arbitrary, I don't see how this is picking on the NYT in particular.
I am also absolutely amazed by the response. I can't believe so many people would like things they believe to be lies by some definition to be censored - what are we, back in the Soviet Union? I also can't believe most people think they can easily tell whether things are true or false by whatever their definitions are.
Not only I wouldn't want a job censoring things under any circumstances, I also wouldn't want a job checking facts for being outright false unless I was paid really, really well, because it's a very difficult job. Most stories of uncovering academic fraud involve crazy amounts of difficult detective work - this stuff is not easy! Same goes for many other kinds of fraud, the kinds you can go to jail for.
So, do you think that there shouldn't be any laws against libel or slander? Because that's a fairly common case of the government punishing things that it believes to be false and harmful to someone, which actually did happen to Alex Jones.
Oh no, these laws are fine. When a case goes to court, it goes through the process of establishing the facts and the intent. The process is public, and the accused party gets to speak to defend itself. This might go wrong, but this is pretty much as fair as it goes.
That's a far cry from "lets just prevent everyone who thinks something we deem to be a lie from publicly speaking about it".
EDIT: The legal process is a non-negligible burden on the accuser. This makes sure that it's only used on rare occasions, rather than every time someone says something the accuser doesn't like.
I've had this relevant Dr. Seuss song playing in my head since morning:
"With a flick of my thumb, with the greatest of ease, if I wish I can stop all the buzzing of bees." That's exactly what Twitter, Google, etc. were doing - and I would think this would bother almost anyone who gives it a minute of thought, but apparently that's not so.
I have definitely detected an increase in people who think censoring is both necessary to prevent various forms of harm and not particularly difficult to implement. This group usually likes to put square quotes around the term free speech.
What I haven't seen a single incidence of is a person who believes in such a censorship regime that doesn't additionally believe the current holders of cultural power are on their side. I kind of get whiplash between arguments that censorship laws are a net good and that they believe their opponents are malicious authoritarians in hiding.
Okay, but how do you get from the extremely narrow and pedantic point that deception rarely involves literally inventing facts whole-cloth to the point you were originally trying to make about censorship?
That was the original reason you argued that this distinction matters, at the start of this sequence - you said that "people" (unspecified) want to censor Infowars but there's no way to do that in a principled way without also censoring the NYT for doing the same sort of thing.
But the legal system *already* knows how to make these distinctions. Judges know and apply all sorts of nitpicky categories like "reckless disregard for the truth" or "matters of public concern" to identify exactly how much slack we want to cut someone for telling a lie that isn't an outright fabrication. And this system has demonstrably managed to punish Alex Jones for his most egregious type-6 lies while not punishing the more minor lies of the NYT.
People don't use these sorts of distinctions in everyday speech, sure, but when it comes to actual potential censorship by the government these distinctions are already well-known. So what exactly are you arguing should be different?
The US government isn't ever going to be doing this type of censorship directly due to the 1st Amendment. The legal system uses a non-criminal civil process to potentially punish a single instance, it doesn't impose lifetime bans of speaking. The censorship regimes at social media companies do their work in hiding with little transparency. There is a significant burden on those who wish to prove a lie legally so the process is not used frivolously. The FBI sends private emails to "trust and safety" personnel. The legal system is accountable through the voters.
The vast majority of things censorship supporters want to ban will not pass legal muster used in the current court system. Perhaps this is the point you are trying to make but the counterpoint is that this is not the type of process these regimes would support because it doesn't fulfill their goals. Thus. back to the original point of you aren't going to be able to articulate a neutral law that achieves your goals of silencing the wrong-thinkers in practice without figuratively sweeping up both InfoWars and the NYT.
> I asked “Why? A woman said she had been raped, and Rolling Stone believed her. The woman was making it up, but Rolling Stone wasn't” and Deepa commented “Isn't it the job of a reporter to investigate? And be good at it?” I don’t want to pick on Deepa, but this is what happens when you have an overly expansive definition of “lie”!
Doesn't this open you up to manipulation by bad actors? What if I tell you that I have a 100% guaranteed method to make you money. Many people are suckered in. I collect fees from them and become quite wealthy. And it turns out its all a scam. They call me a liar and I say "Well, my friend Bob told me about it. So I wasn't LYING technically." If lying is only saying something which the speaker knows to be untrue then the concept effectively does not exist.
Fwiw, in defamation cases and the like you have to prove you did sufficient investigation and all that. Which Rolling Stone did not. Which is why they settled. They were going to lose badly.
I think your example might be unfair, because in that example the scammer has independent reasons to know he's running a scam, other than Bob's story. He's running the scam! Collecting money and getting rich.
A fairer example would be if Bob is running the scam, and you're not part of it, but you hear from Bob that it's a fabulous get rich quick scheme, totally legit. You pass the story onto a friend of yours, enthusiastically, and your friend invests with Bob and loses everything. The question then becomes: can your friend sue you for fraud in the same way he can sue Bob? Shouldn't you have done due diligence[1] before passing on the recommendation? Aren't you just as much a liar as Bob himself? I suspect most people would have at least a little problem with this.
---------------------
[1] I mean, you absolutely should have, morally and legally, if you're acting as your friend's financial counselor, and he pays you a fee for that purpose. But a newspaper is not in a fiduciary relationship with its readership. Nobody has created a contract, explicitly or even implicitly, in which you pay money for a product of a certain level of guaranteed accuracy. The news is sold by newspapers and media on an "as-is" caveat emptor basis. It's up to you to read the prospectus carefully before investing.
No, no Carl. That's what happens when you have an overly expansive definition of lie! You see, yes it was untrue. And yes I was benefitting from it. And yes I could have discovered it if I'd done a little investigation. But I wasn't making it up. Bob was. I was just credulously repeating what Bob said in good faith.
Your example is a good example of the same phenomenon. But I was purposefully pushing it to an absurd point with mine.
As to the whole newspapers are caveat emptor: They're not actually. The story did meet the legal definition of defamation. Yes, truth is a defense to defamation and untruth is not automatically defamation and the law is not the exact remit of morality. But it doesn't change that Scott's definition here is so strict he's required to believe many convicted fraudsters never lied.
Yes and no. Defamation is not unique to newspapers, a blog commenter can be guilty of defamation as easily as the New York Times. Furthermore, I don't think there is any implicit contract between a newspaper and a reader *other than the defamed person* that the newspaper will not defame anybody.
Indeed, if anything the existence of tabloid journalism (or InfoWars for that matter) shows that the public actually likes and wants newspapers that come as close to defaming people[1] as is legally possible.
I'm comfortable with concluding some fraudsters are not liars. There are many ways to mislead people other than telling outright lies. That's why we have a separate word "fraud" instead of just merging "fraudster" and "liar" because they are in one-to-one congruence.
But if we go the other way, if we expand "lying" to include a host of much more minor human foibles that are intended to influence -- perhaps even influence a bit beyond what objective fact strictly merits -- we are going to make the word as useless as the word "racism" has become[2].
We need strong, toxic words, words that indicate stuff that is so far beyond the pale nobody will defend it, and we all recoil from it. Their existence, and occasional use, serves to set clear boundaries on civilized behavior that stiffen the spines of those otherwise too weak-minded to resist momentary temptation. ("Do this and people will call you a liar. It's not worth it.")
But to preserve those clear markers we also need to be stingy in their application. They really need to be applied where even very generous or hihgly biased souls would (perhaps reluctantly) mostly agree. For example, we should orobably only call "lies" the things Donald Trump or Joe Biden say that most people who voted for Trump/Biden would agree are "lies," so we can achieve a 80-90% social consensus on who is a liar and who is...something else, maybe something undesirable, but not over the line with the Nazis and molesters.
Preserving the social consensus around what a toxic word means is far more important, in the long run, than winning any momentary social battle. Or rather, failing to preserve the social consensus means winning today's battle for your faction but losing the long-term war to preserve general social cohesion. A poor tradeoff for which our great-grandchildren will hate us.
------------
[1] Although by "people" here I obviously mean the half-human half-orc out-group bastards who are sapping and impurifying our precious bodily fluids.
[2] Not to mention it will kill salesmanship, politicking, and most Internet commentary dead. We'd all be reduced to the bland anodynes of the Camazotzian press, approving mildly to strongly of the latest revised Five Year Plan.
> Defamation is not unique to newspapers, a blog commenter can be guilty of defamation as easily as the New York Times. Furthermore, I don't think there is any implicit contract between a newspaper and a reader *other than the defamed person* that the newspaper will not defame anybody.
I don't see the relevance. Journalists are not a special class. They have no more rights than a random citizen. So the standards put on the general public are the standards we expect from journalists as well. There's no implicit contract. There's instead an explicit rule about what anyone, including a newspaper, is allowed to do. And unless you're going to argue we should expect criminality then we should expect them to stay on the right side of that line.
> Indeed, if anything the existence of tabloid journalism (or InfoWars for that matter) shows that the public actually likes and wants newspapers that come as close to defaming people[1] as is legally possible.
And the existence of the Rolling Stone case and Sandy Hook case show that the public actually likes people who commit actual defamation. I again don't see the relevance. People like defaming people is a true statement. But so is that people like stealing.
> We need strong, toxic words, words that indicate stuff that is so far beyond the pale nobody will defend it, and we all recoil from it. Their existence, and occasional use, serves to set clear boundaries on civilized behavior that stiffen the spines of those otherwise too weak-minded to resist momentary temptation. ("Do this and people will call you a liar. It's not worth it.")
Yes, I agree. In fact I think we should go beyond having the words be toxic. They should be legally enforced through things like defamation or fraud statutes. And Rolling Stone crossed that line. Scott is arguing, however, that this isn't a case of them being a liar. He's implicitly setting up a definition of liar that is more strict than defamation statutes. That's what I disagree with.
> But to preserve those clear markers we also need to be stingy in their application. They really need to be applied where even very generous or hihgly biased souls would (perhaps reluctantly) mostly agree. For example, we should orobably only call "lies" the things Donald Trump or Joe Biden say that most people who voted for Trump/Biden would agree are "lies," so we can achieve a 80-90% social consensus on who is a liar and who is...something else, maybe something undesirable, but not over the line with the Nazis and molesters.
I agree we need to be parsimonious with their application. But this is neither the legal standard or the practical standard as actually applied.
> Preserving the social consensus around what a toxic word means is far more important, in the long run, than winning any momentary social battle. Or rather, failing to preserve the social consensus means winning today's battle for your faction but losing the long-term war to preserve general social cohesion. A poor tradeoff for which our great-grandchildren will hate us.
No, I'm not sure. I'm sure I disagree with Scott's point that the thing didn't rise to the point of being a liar. I'm not sure if you agree with him or me though.
I agree with the central thesis, and should probably be a bit more careful about the use of the word "lie". That said, is the "(make|leaves) other people confused" in 5/6 not overly specific? Did you mean, rather, to say something like "intended to achieve a specific goal other than relaying useful information about the world" (implant an idea, attract readers, sabotage reasonable discussion, etc.)?
You keep ignoring the difference between "a lie" and "a liar".
I believe that the main reason your posts seem so off to many people is simply that: judging the sincerity and the integrity of any given report in isolation is indeed a tricky and delicate business, as you so articulately demonstrate. I agree with you completely on this. But judging the sincerity and the integrity of a given news-source is much more robust, and orders of magnitude simpler. And it's the later that is relevant to your original point, about fighting fake news.
There's a standard argument in cognitive therapy (at least that's where I know it from) against using terms like "liar": they are either overly strict or arbitrary labels.
I'm sure you have lied before, even in the narrow sense of claiming something that you knew to be false (no offense). Does this make you a liar? No, because you don't do it all the time, possibly even rarely. But what is the frequency at which somebody has to lie that it's justifiable to call him a liar? Once a month, week, day? If more than 50% of your statements are lies? So you're good if you truthfully comment on the weather between two lies? Statements of importance? Important to whom?
Maybe you want to use the "I know it when I see it" escape hatch, but this seems pointless in an argument with someone who very apparently sees the same things and comes to a different conclusion.
Your argument seems to me so general to render it useless (is anyone really tall?).
But anyway, we can and should avoid these problems altogether: we're talking about news-sources, not private people. Roughly estimating the trustworthiness of news agencies and labeling those below some threshold as "fake-news sources" does not suffer from all the psychological, sociological and ethical issues that may arise when labeling specific people as "liars".
It does suffer from being easily subvertable by publishing more true but irrelevant (relative to the goal that is likely being pursued by pushing fake news, they can still be globally relevant) articles if you define the threshold in relative terms. And absolute terms don't really make sense either. One would have to measure it as a fraction of volume of information conveyed, where information that is exclusive/reported by fewer other outlets is weighted higher (this would also mean a completely fabricated story would be given a very high weight).
It gets very technical very quickly, and maybe "I know it when I see it" sometimes just is the best we can do. But I would also say that this is decidedly different from saying that someone is tall, because height in an upright position w/o non-body-accessories like shoes can be easily and objectively measured and compared. Nobody is ever absolutely tall (compared to, idk, an elephant), but if you know someone's height, you can tell their percentile within a certain population of humans and make a well-founded claim that they're tall relative to that population. This isn't possible with "trustworthiness" of media outlets.
I don't claim that "I know it when I see it". I straight out claim that it's a tractable problem to formalize and solve. I think that the objections you give and the difficulties you point out are valid, but can all be satisfactorily addressed.
How? We may start by quoting Scott himself: "P(A|B) = [P(A)*P(B|A)]/P(B), all the rest is commentary". But I'm aware that it's not obvious and there's a lot to be said. However I'd like to avoid delving into this (technical, as you rightly say) discussion right now since it's worth doing so only after there's an agreement over the following points: (1) the epistemological issues that worry Scott looks very different on the level of "news-items" and the level of "news-agencies", and (2) it's the later that practically matters, and it's an easier problem to solve.
To clarify: it's conceivable that the problem still remains difficult, but I claim it's clearly much easier, and that most of Scott's arguments are at the very least dramatically weakened when applied to it, if they applicable at all.
For anyone who has published and has accused someone of lying, the consequences, if wrong, can be disastrous. Some people care about the consequences more than others. “They should have known” is not a proper defense most of the time. In general, both writing and research with the intent to publish are painstakingly tedious, so even the best of journalists can err in way that elicits others to call them a liar. The best of us should accustom ourselves, out of habit, to reserve this accusation in public as much as possible. Better to say it is misleading or deceptive whenever practicable? Or if we're in the accusatory mood, we can say someone is trying to mislead/deceive? Doing this can keep the focus on debate rather than defending why something is not a lie.
Furthermore, everyone believes things that are false, and for most of us, a magical revelation of some of those misbeliefs, if we magically accepted them, would be a shock. Many of us want to know if we’re wrong even if it will ruin our day/week/life, but many others don’t want to know, at least on a subconscious level. I wonder if any correlation can be made between people who want to avoid the truth, intentionally or instinctively, and those who want to “wirehead” themselves to a perpetual happiness device. I wouldn’t accuse anyone of fitting this mold, but it would be an interesting study.
> Looking at these people's other responses, they did not show a consistent tendencies to make things up or say outrageous things (except for one who listed their religion as “Satanist”).
This is a legitimate, if niche, thing. There's a Church of Satan which has real religious beliefs, and there's an unrelated Satanic Temple which is atheist, but for all I know some of its members call themselves Satanists.
Paul makes a good point about proving too much. If you define "lie" in such a way that "The National Enquirer Rarely Lies" is a true statement, then who the hell cares. Ultimately, people are interested in how often NYT or Infowars does "That Thing The National Enquirer Does" when writing a story, whether or not you want to call it lying.
This problem would be reduced if the news media had a convention of annotating each headline to indicate whether the editors think the proposition is probably true (PT) or probably false (PF)?
'BIGFOOT FOUND' (PF)
And for question titles, probably (P) and probably not (PN) instead:
'WILL BIGFOOT BE FOUND?' (PN)
'SOURCE CLAIMS BIGFOOT FOUND' (PT) defeats the purpose. So the convention would only be meaningful if it was strictly applied to the inner proposition.
However, I expect that this annotation would reduce readership (by making articles less interesting to read), and that it will therefore never happen.
Scott, I feel like the big issue here is that you're conflating two different things - both of which can be described as "technically true, but misleading/deceptive," but which are nonetheless very different in both degree and kind.
The first category includes your example of FOX News saying "more Whites are killed by police than Blacks" without mentioning that Blacks are killed at a proportionally higher rate. It also includes your example of the New York Times saying "Blacks are killed by police at a higher rate than Whites" without mentioning that police are called to Black neighborhoods much more often. In both cases, the fact is true, and arguably even genuinely useful to know, but there's some critical piece of context missing which leads people to draw flawed conclusions. This isn't quite lying, exactly, but it's not telling the whole truth either. And if the missing context was purposefully left out in a deliberate effort to make people draw false conclusions, then I'd consider it a lie of omission. On your 1-7 scale, these would be ranked 5 or 6.
The second category is when someone just tells a straight-up bald faced lie, but argues that it's "true" in some abstract sense through some opaque trick of semantics. On your 1-7 scale, these would still be ranked 7, or at least 6.9999999! But I think you're making the mistake of grouping them together with the 5's and 6's. Some examples of what I mean:
-Redefining words to mean something entirely different from the common usage without letting anyone know that you're a nonstandard definition
-Taking advantage of homonyms to substitute a word for a different but identical-sounding one (e.g. if your wife asks you to deposit a check for her, and you say "I promise I'll stop by the bank today," but then go fishing at the riverbank instead)
-Speaking in metaphorical terms, despite all context making your statement sound literal (one famous example is Obi-Wan saying "Darth Vader killed your father")
-The practice of mental reservation (i.e. saying "I don't have any money" out loud while mentally appending "in the Bank of Zimbabwe" to the statement, thus giving the version of the statement in your head a positive truth value)
-Hearing someone make a statement that you know to be false, then sharing it with other people without letting them know that it's not true (for instance, if Bob says "I saw Frank's girlfriend Martha making out with Sam at lunch today," and you know for a fact that Martha was alone studying in the library during lunch period, but you still go around telling people "Bob said he saw Martha making out with Sam" and letting them assume it's probably true)
To even call these "lies of omission" would be overly generous. They're just outright, blatant, unambiguous lies, simply worded in such a way that the speaker can retroactively *claim* (however implausibly) that they were being honest. And InfoWars, unlike the New York Times, frequently resorts to these sorts of tricks. ("I didn't say "[false statement]," I just said "X said [false statement]," that's totally different" is a pretty common defense of theirs.)
>Redefining words to mean something entirely different from the common usage without letting anyone know that you're a nonstandard definition
I've been seeing this a lot lately. I believe it was the New York Times that was stating this week that we never had any lockdowns in the US, simply "intermittent state by state mitigation measures" or some other such euphemism. Oh goody, I'll have to go inform the proprietors of the restaurants that were forced out of business in my town in 2020 that we never had any lockdowns, I'm sure they'll be thrilled.
When was the like button removed? Or am just not seeing it?
Anyways, thanks Scott for discussing that at length with anyone. That's exactly how I feel about the phrase "lie" but would have given up arguing by now.
Just going to add that I was quoted in an article in The Times of London (on the occasion of exam results day) as saying something like, "Anyone who says marking standards are falling is belittling our achievements". Didn't say it! But it did support the narrative of the article.
"I'm skeptical that 10 out of 500 people were unfortunate enough to have 2 household members die: one from COVID and one from the vaccine. (Especially because these are not large households; 4 of these 10 report that they have 1 other household member [...])"
Those 4 are obviously protagonists/narrators in hackneyed "they were dead all along"-twist horror stories.
and I think it's generally individual reporters doing this freelance, so to speak. I can't think of any fabrication scandals where editors issued an order to make things up. Maybe there were some back in the "yellow press" days?
Could it be a helpful contrast to say that Moscow Kreml (or indeed any nation at war, Kiev as well but less in-your-face) lies in the point 7 sense continuously 24/7? It's part of the war effort to spread misinformation.
Most media are NOT participants of an actual physical war and do NOT deliberately spread lies, though they still serve interests that compel them to do all things in points 2 through 6. Contrasted with a clear example of point 7, and thus a bit detached from the U.S. bi-partisan echo-chamber circle-wankery, perhaps the claim would sound a lot less controversial.
He distinguishes it from lying, defining it as "speech intended to persuade without regard for truth". A key summary from the wikipedia article:
>Both people who are lying and people who are telling the truth are focused on the truth. The liar wants to steer people away from discovering the truth and the person telling the truth wants to present the truth. The bullshitter differs from both liars and people presenting the truth with their disregard of the truth.
Now, you might say that bullshitting is just a variant of "egregious disregard for the truth", so it's already been analysed above. But I think that we rationalists have a bit of a blind spot where bullshit is concerned; we're so concerned with the truth that we assume that others are equally so concerned (interestingly, many conspiracy theorists are similarly very concerned about the truth). It didn't appear in Scott's list in section 8 (point 4 was the closest, but "more-or-less subconsciously not trying to reason well" doesn't really cover "entirely unconcerned with reasoning well, and not seeing why reasoning well would even be an issue").
So I think we want to round bullshit to something more understandable. Alex Jones, for example, is mainly a bullshitter. The truth or not of his published statements is not a concern of his. Is he lying? As Scott pointed out, he's not. Is he telling the truth as he sees it? No, he's not doing that either. So I think Scott is reasoning "he's not lying" while others are reasoning "he's not telling the truth" and then get into definition arguments. But if we add the third category of bullshit, then it seems easier to classify Alex Jones.
So, armed with bullshit as a category, what can we say about the other media sources? Weekly World News? Bullshit. New York Times? Going off the descriptions in this article and the original one, not bullshitting (or rarely doing so) - we might need to add another category for what they're doing, but it's not bullshitting. Infowars in general? A high proportion of bullshit, though not all of it.
Anyway, just wanted to remind people that some people really don't care about the truth, one way or the other. They're not lying, they literally don't care about the truth, and probably have difficulty imagining that other people do.
One might think that you are expending an inordinate amount of time and energy on an undeserving subject, but not me. The actual subject of this investigation is:
On issue 3 and media believing their own lies, among many obvious differences between Infowars and NYT I see one that is particularly relevant.
Imagine that journalist Alice works for a newspaper and covers space exploration. She has opinions on the matter, reads many sources, does her best to decide which stories are true and which are not, and honestly tries to inform her readers as well as possible. There are people out there who disagree with Alice on some topics, sometimes pretty emotionally, but these people do not control most of her sources and do not have the means to manipulate her. When Alice finds a story about Saturn rings of some given relevance and novelty to her readers, Alice would try to establish the degree of truth in this story and will not publish it unless she was 95% confident in it.
Another journalist, Bob, works for another news outlet, covering domestic politics, NSA surveillance, and conspiracy theories. He has the same motivations as Alice and also does his best to decide which stories are true and which are not and tries to honestly inform his readers. There are more people out there who disagree with Bob. However, for some weird reason, Bob honestly believes some of the outlandish conspiracy theories, such that there is a secret email list of journalists that covertly coordinates domestic policy coverage in main news outlets to favour one party, or that Twitter is running a shadow banning program against people who express beliefs similar to Bob's, or that most of media fact checker outlets are coordinated and funded by the same people who fund one of the political parties.
When Bob sees a story about a certain conspiracy theory of the same novelty and relevance to his readers as was Saturn rings story to Alice' readers, he would do his best to establish degree of truth of this story and would not publish it unless this degree was rather high. Would it be rational for Bob to set this threshold at the same 95% as Alice? I do not think so. Bob honestly believes the conspiracy theories about Twitter and factcheckers, so he thinks that his and his readers news flows are heavily biased against this kind of stories. A lot of good and true stories simply do not propagate well enough to reach Bob and his readers because of suppression. So the stories that do get to Bob are more valuable, and Bob would set his threshold at 80% and not at 95%. This does not mean that Bob is less honest of more gullible than Alice. He just believes that he lives in a world with powerful forces that manipulate the news flows and sees his duty to try to compensate.
Nicely argued. I especially like your last series of comments. One problem you hint at but don't really answer is what are we supposed to say/do about the many instances when grossly deceptive information is disseminated without actually crossing the threshold of genuine lying?
There is a huge ingroup/outgroup component to this. When your side is just wrong, you can safely ignore it for a while, then come out with retractions/corrections if you want (e.g. NYT on Russiagate, Hunter Biden laptop, etc). When the other side is a member of your outgroup, it's all "lies" down the line, even if the "lies" are mistaken yet honestly help beliefs. The reasoning here is pretty simple, calling someone a "liar" is a moral attack that us useful to deploy on your enemies.
Contrast Alex Jones to a group that says lots of false things all the time but we don't call a bunch of liars: flat-earthers. For a variety of reasons that may include a mix of genuine belief / trolling / performance, they make even more unlikely claim that the Earth is flat. I highly doubt that the NYT will be condemning them as a bunch of liars anytime soon. Nor do we condemn professional wrestlers, or members of small religious movements even if we think their beliefs are mistaken. You have to be an outgroup member to deserve this treatment.
Also, Jones never said he lied as far as I can tell. Rather, he said that the shooting actually happened which the media ran with as "Alex Jones admits he lied". But if you look at what he actually said at the time of the shooting, it was mostly just his conspiratorial take on the situation that things look fishy. The NYT helpfully compiled a list of Alex Jones "lies" around Sandy Hook. It is here:
Ironically, if he has said at the time "I think the shootings definitely happened, nothing to see here" , then *that* would have been a straight-up lie as he clearly did not think that. This just highlights the alchemy of converting "my opponent is wrong about something" to "my opponent is a liar"
What about media that are propaganda amplifiers, like Russia Today? Sure, they still prefer 6 to 7 (if only it's generally considered to be more effective), but when the same source claims four contradictory versions in two-to-three days (I think I'm not even exaggerating the time - thinking of the Bucha disaster) with the only thing in common "we did nothing wrong there", one has to wonder whether this is even possible (quoting gwern: "“Possibility” in ordinary language doesn’t include astronomically small odds.") to do on pure 6, without 7.
It is so nice to see discussions of moral theology on here re: "what constitutes a lie? is lying always wrong? is silence, equivocation, or mental reservation the same as lying?" 😁
"In a country with 100 million conservatives, is it really that hard to find a handful of them capable of writing news articles?"
Regarding the liberal bias in newsrooms, this is an old story. Even the media itself (at least back several years) recognised this. *Why* it happens is something that is much debated.
The Public Editor of the New York Times wrote about this in 2016:
"Over the years, talking to fellow conservatives about media bias, it has usually been my place, as one who worked in mainstream media, to tell conservatives that they’re wrong in some significant way about media bias — not its existence, but the way it works. Most reporters and editors, in my 20 years of experience, do not set out to slant stories, and in fact try to be fair. The bias that creeps into their coverage is typically the result of a newsroom monoculture, in which they don’t see the bias because everybody, or nearly everybody, within that culture agrees on so much. In the case of gay rights and the marriage debate, though, they don’t even make an effort to be fair. I have heard some version of the “error has no rights” claim for years now. They honestly believe they are morally absolved from having to treat the views of about half the country with basic fairness in reporting. And they are shocked — believe me, they really are — that these people view them and the work they do with suspicion, even contempt."
So where are all the conservative reporters? Probably working for newspapers you never heard of because they're local or regional. Why aren't they getting jobs in the NYT or the Post? It's that thing about "cultural fit" in interviews - sorry but we just don't think you'd fit in to the company culture here.
Remember back in 2020 the staff protesting they felt unsafe because the NYT ran an op-ed? Now just imagine the protesting about expecting them to work side-by-side, in the same room with, a bigot and reactionary!
The curious thing is that this is a relatively recent development. The left-right division in the media used to be roughly similar to that in the general population. For example, take a look at this fascinating graph of newspaper endorsements for President versus the popular vote:
In 1984 the fraction of newspapers endorsing Ronald Reagan for re-election was pretty close to the fraction of the popular vote that Reagan won. But the situation skewed increasingly left after that, so that today newspapers are wildly out of step with the general voting population.
So why has newspaper culture become so weirdly exclusive in the last 50 years?
"So why has newspaper culture become so weirdly exclusive in the last 50 years?"
I think it's a combination of things. American journalism liked to present itself as being 'just the facts' fair, impartial, unbiased reporting unlike European advocacy journalism (where every country has its version of "A is the right-wing paper supporting Party D and B is the left-wing paper supporting Party E"), but that was relatively modern. Good old muck-raking and "follow the proprietor's political views" was plenty common in American papers.
So there was indeed an underlying slant, even if the staff did genuinely think they were being fair and impartial. Secondly, there were big conservative as well as liberal papers, and regional/local papers were as influential in their own markets. The New York Times might like to think of itself as 'the newspaper of record' but an awful lot of people around the country were not reading the Times, they were reading the big regional paper and/or their own local papers.
Thirdly, the liberal reporters became senior and moved up to be editors, as per this opinion piece in the NYT itself from 2015:
"For instance, Bill Keller, after years as a New York Times columnist taking largely liberal stands, became The Times' executive editor.
Studies consistently show that reporters and editors stand to the left of the American center. More than 30 former journalists now serve in the Obama administration — nothing comparable happens in G.O.P. administrations.
Clustering of left-of-center viewpoints in the newsroom leads to a cloistering, and thus reporters end up unfamiliar with conservative viewpoints. This shows up in the tone of daily coverage (for instance, “property rights” gets put in scare quotes, while “abortion rights” doesn’t)."
This is the former editor Bill Keller who said, in a talk in 2011:
""We're liberal in the sense that ... liberal arts schools are liberal," Keller noted, during a recent dialogue recorded at the Lyndon Baines Johnson Library and Museum. "We're an urban newspaper. ... We write about evolution as a fact. We don't give equal time to Creationism."
Moderator Evan Smith, editor of the Texas Tribune, jokingly shushed his guest and added: "You may not be in the right state for that."
Keller continued: "We are liberal in the sense that we are open-minded, sort of tolerant, urban. Our wedding page includes – and did even before New York had a gay marriage law – included gay unions. So we're liberal in that sense of the word, I guess. Socially liberal."
Asked directly if the Times slants its coverage to favor "Democrats and liberals," he added: "Aside from the liberal values, sort of social values thing that I talked about, no, I don't think that it does."
Next, I do think journalism schools had a lot to do with shaping attitudes, but this is only my own view. Going to college to study journalism, instead of working your way up on a newspaper as in the old days, means exposure to academia, and academia slants left. So you're going to be taught that set of values, and if you remain conservative in your views despite it all, you probably learn to keep your mouth shut and go along to get along.
So there does develop in newsrooms, as per the Keller interview, this culture of "we're all liberal, we have all the right opinions on various topics and are on the right side of history, we have the same social values and beliefs that all good and right-thinking people have". It doesn't mean they think they are reporting in a manner that is *not* fair and impartial, but they don't think to examine their biases anymore than the rest of us think to do so.
And lastly, there has been a push by the new, younger journalists to engage in activist journalism because they really do believe there is only one right side of history and it would be immoral not to fight the good fight, that error has no rights, and there should be no pretence at 'balance' or 'both sides of the story', because all the good right true facts are on one side and all the bad wicked prejudice and bigotry is on the other. So, right back to advocacy journalism of the old days.
The National Enquirer story about deliberately >not< checking out their sources' stories resonates for me.
For mainstream, i.e. what people consider "serious", media deliberately not checking: remember Rathergate? Documents (that we now know were fake) presented as real on national news about a political story just before an election. Not authenticated, but Rather said they had been, and continued to say so for weeks (as well as the company backing him).
Now, you can argue that this is rare and therefore memorable.
I say it's common, but most of the time outside people don't go to the effort of actually checking, or else when they check and find it fake they are drowned out by the "new news" or by deliberate suppression. For the few times I have been on the scene for something later reported, I have always found something inaccurate about the report.
[edit]The unusual thing was not that Dan Rather lied. Technically I guess he didn't lie, Mapes lied when she said it had been checked and it had not, while I guess Rather trusted her - although it's interesting that he didn't press her on it when the refutations started coming out and didn't get a second opinion on it.
No the unusual thing was that he got caught at it. He thought he could brazen it out, and it didn't happen.[/edit]
[Further edit]From what I can see (I haven't really followed the multi-year saga) the difference with Infowars seems to be that they got sued fairly quickly. Once they got sued, backing down would have meant losing the case. I guess they also thought they could outlast the people suing them.[end further edit]
I’m a day late, $20 short, but I recall the ‘George Bush amazed by grocery store scanner’ being in a nightly news broadcast. (Probably ABC) where it was presented as he was out of touch with the masses…
YEARS later I encountered the truth in my data-collection tech job. It was brand new scanner tech that weighed and scanned damaged barcodes.
Personally I think a very common mode, that is missing from your list is the noble lie and I think a lot of journalists/experts believe they are engaging in the practice. While this is technically congruent with your 7. I think there is a strong association with your phrasing that the person in question is malicious or bad or some such, which doesn't perfectly match up with the noble lie. If you agree with the noble liar it feels weird to call them a liar, but if you disagree that is obviously what they are.
Your definition of lying implies that a person who reasons badly or doesn’t have a good enough model of the world are incapable of lying. This seems like an overstatement.
I encourage people to read the "Nanci Pelosi Hanged at GITMO" article that Scott linked to (https://realrawnews.com/2022/12/nancy-pelosi-hanged-at-gitmo/) thanks to a commenter who wanted to provide an example of actual fake news. I'm sure most would agree that there is no mistaking that particular article for anything but parody fiction (and not particularly good or incisive parody either). I'm not talking about the gist of the content itself (that Pelosi was recently hanged at Trump's order, the way Hillary Clinton was hanged the year before) but the style of narration, which comes across as that of a comedic short story writer trying to amuse themself in a fantasy-fulfillment way. I just don't believe that in any day and age anyone but the bottom-10% stupidest readers among us would mistake a story written this way for a genuine news article, whatever the platform calls itself.
I read your Section 8 (My actual thoughts) as a round-about admission that yes, your definition of lying is unusually high, as many commenters point out, even though you mostly individually refuted them. That is ok, though, it is your choice how strictly to take the definition of lying; if anything, it clearly shows that even you are operating somewhere along that spectrum of truth vs. lie, and your original headline is clearly not a lie given your own definition, but might be viewed as deceptive by others who have a narrower definition of sharing true facts.
I do very much appreciate your article and the discussion though - it makes it very clear that outlawing, or even identifying and condemning false stories is exceedingly difficult. Your point that mainstream media and conspiracy platforms are not fundamentally different, just a different points on a spectrum is the one key and true fact here.
Perhaps we have to take the old adage that "you cannot not communicate" and change it to "you cannot not lie?" Writing a story that is entirely true, based completely on verifiable evidence, and not deceptive at all is very difficult in a complicated, messy and contradictory world.
FWIW, I thought picking apart the ways they give misleading ideas without formally lying was a nice way to teach media literacy and criticism. Kind of like an updated 'How to Lie With Statistics'.
>I asked Jeremy to guess the probability that this was an honest mistake vs. malice. He said (thanks for giving a clear answer!) 60-40 in favor of malice. I think this is pretty high, given that I had to read Jeremy’s comment several times before I realized what the error was supposed to be
>“The intentionality of an agent with behavior sufficiently indistinguishable from malice is irrelevant.”
I don't think it was malice, but excusing The Washington Post (or The NYT, or whoever that's supposed to be at least occasionally reliable; ie, not Cousin Jimmy's Facebook Rant) for an "honest mistake" of this scale is too generous. It's... what's a good name for it, conveniently advantageous stupidity? What did you call it last year- [too good to check](https://astralcodexten.substack.com/p/too-good-to-check-a-play-in-three)? Dishonesty that tells the right story doesn't get checked, and while it's not technically a lie, neither is it usually an honest mistake. Or rather, we can't know the difference between an honest mistake and a dishonest one, and the distinction is irrelevant for truth-seeking anyways.
Rarely do I think the supposedly-trustworthy outlets are *malicious*, per se. But neither do they make *honest* mistakes. You say all your readers are too paranoid, all your readers say you're insufficiently paranoid, and the truth is somewhere in the middle.
However, the NYT said Sicknick WAS murdered, and so WP:RS applies and that false "fact" got repeated everywhere, including Trumps impeachment.
Now I would think that the NYT might want to explain how/why they pushed this lie (er, untruth) and a good way of doing this might be to burn their source for the "murdered" bit. After all, that's a mistake so egregious that literally nobody should use this person as a source again.*
But they won't.
Which is odd, considering that they've been willing to out bad sources before. Why not in this case?
-They are good friends with the source and don't want to embarrass them?
-The source is politically/socially connected in the journalist community?
-The source has been reliable (in the sense of truthful) in the past and has pinky-sworn that they won't make such a mistake again?
-The source doesn't actually exist, but was inferred from a DM-chain?
-The source has been reliable (in the sense of telling the reporter what they want to write about) and they don't want to lose that resource?
*Of course, I though the same thing about Vox after their "those meanie Israelis shut down the bridge between the West Bank and Gaza" article, but apparently most people disagree with me.
One of the reasons for making the lying distinction as narrowly as Scott does, that I have not seen mentioned here or in the other two columns, has to do with which forms of deception are actionable, and in what ways. (I made a comment about this on DSL.)
Lying in the sense Scott seems to want to use, is libel (or slander), and you can sue for it. It's hard, for multiple reasons, but possible. If a journalist makes a claim, and they know it is false when they make it, and it causes damage to you, and you can measure that damage in dollars, and you can convince a jury of all this, then a judge can punish that journalist in a way that journalist will feel, and make that type of lying less likely to happen again.
Lying in the sense a lot of people here seem to want to use, is practically impossible to bring a suit against. Karl Gallagher, in another comment here, mentions "sins of commission" and "sins of omission". Commission is the above thing, and an easy call. Omission, though, can happen for lots of reasons, some innocent, and is technically unavoidable unless every article comes with a bolus of every known true claim in the universe. "Oh, you claimed Saddam was shopping for nukes, but you left out the bit where OBL was Saudi Arabian!" Is the latter relevant? To some people, in some contexts, it is. In others, not. In yet other contexts, it's assumed, and clutters the article if included. Clutter matters. If a journalist makes an important claim in an article, but has to attach megabytes of exposition to avoid any possible charge of omission, readers will glaze over and skip to the next article, and that claim may as well never been made, and so is therefore unmakeable. By extension, every interesting (and possibly true enough) claim is unmakeable. I don't consider that situation preferable.
So omission can't warrant a suit (AFAIK). The defendant can always argue that the omitted claim was irrelevant, or assumed known but not important enough, or out of scope or context, or even itself questionable enough to be un-includable without even more justification. If we ruled these excuses insufficient and we allowed such suits anyway, we would quickly flood the court with them, and nearly any article could be found punishable, and we'd find the suits correlating with "someone had an agenda opposed by the claim" much more than with "journalist was being deliberately deceptive". Ultimately, every journalist would clam the hell up (or find a black market), and there'd be little news to be found anywhere, even the stuff that's true enough to act on.
In a world where one type of deception can bring a formal response by secular law, and another type is effectively unassailable via the same means, it's worth making the distinction. I contend Scott has done so.
Notice I'm saying "via the same means". There are ways to deal with deception by omission, by misleading, etc., and there are even institutional pressures that arbitrarily block those ways, and I think there's changes worth making in those areas. But those ways still require acknowledging the distinction.
As a corollary, acknowledging that distinction will also mean acknowledging that both InfoWars and NYT commit sins of omission, because only then can one group of people with evidence of InfoWars making a sin of commission that NYT isn't, convey that information to the group of people who don't yet have it. The alternative is to yell "one's lying, the other isn't!". That language is so low in resolution, that it makes the former group of people look like they're being deceptive as well. In other words, for example, I see no way to make the case that InfoWars is worse than NYT (or WeeklyWorldNews is worse than WaPo, etc.), without also copping to the charge that NYT cannot be implicitly trusted.
<blockquote><p>[Jestin] Coler is a soft-spoken 40-year-old with a wife and two kids. He says he got into fake news around 2013 to highlight the extremism of the white nationalist alt-right.</p>
<p>"The whole idea from the start was to build a site that could kind of infiltrate the echo chambers of the alt-right, publish blatantly or fictional stories and then be able to publicly denounce those stories and point out the fact that they were fiction," Coler says.</p></blockquote>
Many more fascinating details at the link.
It occurs to me that fake news isn't just run for the clicks (as in the Macedonians) or however you want to characterize what InfoWars does. Some of it is a pretty straightforward effort at embarrassing your opponents, with a potentially large payoff. Let's say the FBI agent murder story (as detailed in this article) infiltrated a congressional candidate's talking points because some low-level researcher tasked with showing that Vince Foster was part of a weird and disturbing pattern found it and didn't vet the source, and his notes went to a speech-writer who then didn't double-check the researcher, and then the congressional candidate is saying something that isn't actually true, and then the mainstream media spend the rest of the campaign talking about it.
If anything, I'm disappointed that you're making this out to be a technical nit-picky point.
Most people seem to care passing judgement on media outlets for their morality, and If that is your goal, then sure, the a strictly literal definition of "lying" might be nit-picky.
But if your goal is to be able to be able to make sense of the world, then being able to tell the difference between outright lies, shoddy speculation, and misleading framing is a crucial skill.
We can reason through own analysis and model our own framing, but we're more or less reliant on outside sources for our raw facts, and the knowledge that we can _mostly_ trust those raw facts is extremely useful.
The Bulletin of Atomic Scientists thing appears to be a genuine miscommunication, not even a motivated misreading of facts or the like. The Bulletin talked to a bunch of the scientists MacAskill cited and five of them really did say that they didn't remember talking to him. This is because they were contacted by a member of his research team who didn't mention that the questions were for his book: they really gave the quotes, but didn't think they were giving them to him, for that.
> Regardless of any of this, the point I’m trying to make with these posts is that the media, while doing all of 1 through 6 pretty often, very rarely does 7.
Why did you want to make this point though? I don’t think a significant number of readers learned something important or changed their minds about anything?
>Could I have saved myself some trouble if I had titled the post “The Media Very Rarely Says False Things”? Or “The Media Very Rarely Makes Up Facts”? I think people would have been equally annoyed that I was using “false things” or “make up facts” in a way that excludes technically-true-but-misleading statements.
That's because people are mainly complaining about your category, and only secondarily about your definition. "Makes up facts" is a category that doesn't carve reality at its joints very well--people find such a category much less useful than they find the category "lies". They are upset that you are applying a word for a useful category to a less useful category, but they are also upset that you are using a less useful category at all. Just changing the word you use doesn't change most of the complaint.
>Oh god, if saying there’s “no evidence” for something counts as a lie, then every media source in the country stands hopelessly condemned.
You seem to be saying "this example makes media lies common, so it doesn't count". A better conclusion is "media lies are common; this example demonstrates it."
(This is especially so since you've emphasized literal truth. If literally true means it's not a lie, and context doesn't count, "literally false" *must* mean "lie" and context can't count for that either. And it's literally false that there's no evidence.)
Reason a couple of hours ago posted something which connects to some of these comment threads (one in particular actually but Substack is being stupid and not letting me find it right now):
"How Do We Solve a Problem Like George Santos? The slippery slope of political fabulism, from the "Jew-ish" freshman representative to the president of the United States."
"numerous comments getting angry at me for saying that I was calling NYT exactly as bad as Infowars, and saying I was being deceptive / lying because of this. " - This is a clear sign of a) so. reading too many of this comments - ACX is a no lies-place, that is one reason we are here
b) having done a post that triggers sub-par comments (this one included). Which may happen. - About the NYT: was their hit-piece about Scott a lie (by his definition)?
A lie requires intent. "Intent" of pure talk - is usu. a hard tell. Why do we talk? Why do I write this comment?!? - I want media to be mostly right, not wrong. I hope.
Bigfoot-tabloids run on interviewing cranks? Disagree. We have (had when me young) one in Germany, too. (No, not BILD.) And we do not have enough people believing in alien-abduction/bigfoot/third eye (as in: a real third eye at the back of one's head). Still that tabloid brought out those stories every month. They were not lying but in the business of writing and selling a regular bigfoot-tabloid. Fiction. Same as Harry Potter. Same as the dozen or so of royalty/celeb-journals that need a dozen "stories" every fortnight; same as Barbara Cartland or Zane Grey. They do not do "research" or interviews, lol, grow up.
I don't understand how this point is confusing at all. Of course there's a meaningful distinction between "lying" and "saying true things in misleading ways". I get that it's pedantic but isn't a rationalist blog the place you're supposed to go for constructive pedantry?
> the point I’m trying to make with these posts is that the media, while doing all of 1 through 6 pretty often, very rarely does 7.
Fair enough, but does this distinction make enough of a difference ? Let's imagine two Bugmaster clones: one agrees with you, and the other believes that the media does #7 pretty often, as well. How will their behaviour differ (with respect to the media) ?
I had a sudden brilliant hypothesis in relation to this: "The media very rarely asserts anything (true or otherwise) other than quoting people." This explained exactly why they never lie, as the only lie would be literally making up quotes. Having concluded this, I opened today's Times to check my hypothesis...
And it was half write. The media makes factual assertions about events (as opposed to what people say about events) all the time. However, there is a pattern to the sorts of stories they do and don't report like this:
(I'll call statements such as "John Smith shot his wife" Direct Reports and "John Smith shot his wife, according to experts" Indirect Reports. All examples are from today's (London) Times.)
It's completely standard for Science/World News articles to use Direct Reports. For example, "President Putin is trying to pin the blame for his faltering Ukraine campaign on rogue Wagner mercenaries as he seeks to consolidate his power within a dwindling band of loyalists." There's no source cited either named or anonymous.
Articles about people being "told" to do things (e.g. "Hand out statins on demand, NHS told") aren't really news stories at all, their editorials printed in the "News" section by means of a friendly think tank/rent-a-quote.
Articles about politics are a mixture of Direct Reports of easily verifiable facts printed directly (eg. what's happening in the legislature), and Indirect Reports of everything else. There are few or no Direct Reports that you couldn't easily work out the source of.
Crime/Court Proceedings articles are mostly Indirect Reports, with Direct Reports interspersed.
Culture War articles are mostly Indirect Reports. Direct Reports only occur where they represent very simple analysis, e.g. "The shortlist for artist of the year is all male and four out of five nominees for album of the year are men."
Finance/Economics articles are all Indirect Reports, other than stock prices and indices.
In any articles, background information will often contain Direct Reports in places which violate the above rules. Looking at my own archive of "Newspaper is wrong" screenshots, these are often incredibly sloppy (eg "The poet participated in the feuding between the Guelphs, a mercantile family, and the Ghibellines, who were predominantly agricultural" or referring to Cecil Rhodes as a "19th Century slaveowner"). I'd give it 70% that these are just thick journalists, 30% that they consider this to be simplification as opposed to falsehood.
This broadly makes sense when you think about it - unless an article is written by an eye witness, the information in it must have come from somewhere. There's no reason not to tell you where it comes form, other than saving space.
I think the imporance of it is that if we accept (7) as the definition of a lie (which I'm happy with), newspapers really do very rarely lie. But the reason for this is that (7) can only apply to Direct Reports or inventing quotes, and Direct Reports just aren't a big feature of domestic journalism, and especially the sort of journalism where anyone would want to lie.
In other words, the paper of record of a country that was ruled by the Guelphs for over a century, and who's largest constituent has the Guelph cross as their flag.
“If you’re 71% confident, and you falsely say you’re 72% confident, then you are lying.”
If that’s not just an offhand comment, and you really mean it, then by that definition every conscious agent who has ever lived or ever will live, aside from perhaps certain AI agents that may come to be, have lied and will lie again. And being in a media position would probably incentivize this kind of lying a lot; in fact many of your examples of media not lying but being deceptive or sloppy fit that category. Overstating one’s confidence is a common occurrence, hell I did it at the top of this post. People don’t usually provide artificially precise Bayesian confidence statements, but if pushed how many times will someone who says they’re “certain” about something back off from their certainty? From my own experience, I know a lot of those times I am in fact ginning up some kind of epistemic courage or feeling pressured to firmly state a belief that in my gut if I attend to it I will realize I have less confidence in that I’ve presented. And after that reflection, if it’s say a job interview, there’s a good chance I will stick to my inflated confidence.
Of course at this point this gets into the qualia of lying and deception, and from an outside perspective if you question someone’s confidence and they back down, it’s an ambiguous judgement call whether you think their initial confidence was on some level a lie or just a lack of reflection on the part of the overconfident one. Any given instance of this may be undetectable but it seems like the sort of thing that happens all the time.
It seems really predictable and obvious that the #7 type of lie would be rare in media reporting, since it's so incredibly easily avoidable. If you can just find literally any idiot to tell you exactly what you want to report, and then you can report it as the truth without technically "lying", then why would you NOT use that super convenient defense? Any 'evidence', any 'witness' is sufficient, and you can just make up a QED at the end. Wham, bam, now you're telling the truth.
I'm not really sure how this is a helpful distinction though; I don't think many people are surprised by the idea that most media misleads without committing to full-on #7 lies. You would have to have some pretty motivated anti-censorship cognition going, though, to conclude "technically they aren't lying in the most stringent sense possible, therefore censorship may not be as easy/effective as most people think". For those considering what sorts of 'news' should be censored or moderated away from certain places based on untruth, I imagine very few would consider #7 lies to be a unique subclass of offense, while everything else gets a pass. Outright lies are not a requirement for misinformation; I can find a source to tell me anything I want you to hear.
There's also use of sources, and while I don't think it's lying as such, it's easy to have the 'reputable' sources that you use regularly be the ones with liberal opinions, or at least ones that fit in with the reporter's/media's biases. 'All right thinking and good people of course think X is a human right, so covering this story I will ask Dr. Y and Professor Z for quotes and they will confirm that X is a human right, then I put in a bit at the end from those kooks and bigots wanting to ban X' style reporting.
For instance, Fr. Thomas Reese gets regularly quoted by media because he's a reliable source, and he's on the liberal side so he'll tell them that the Church needs to modernise for the world or this new encyclical is the same old homophobia or whatever. And that's great, and that fits with their own opinions, and they feel like they are being fair and impartial because after all, they asked a real live Roman Catholic priest for an opinion so he must know! If you said they were slanting their reporting they'd be indignant because how dare you, they went to the proper guy and everything for his views and he's the expert here! It never dawns on them that going to the same reliably liberal guy for the same reliably liberal interpretation every time, and not searching for the other side (unless they're doing a bit on 'unreasonable stick-in-the-muds try to hold back progress') is being slanted.
Why I don't think this very strictest definition works:
Let's say I make up a fact and publish it. Technically, I don't *know* that it's false since I also made sure not to check - perhaps it just accidentally happens to be true. But it's ridiculous to claim that I'm not lying here. If *this* is the standard, then *of course* the media almost never "lies".
Speaking briefly to the John Buridan point about how we develop anti-conspiracy-theory antibodies, and inform our priors... the decisive moment of that for me was reading Michael Shermer's book "Why People Believe Weird Things", which explained at a level my high-school self could digest, all about cults and creationism and holocaust denial and whatnot - and then devoted several chapters to debunking them simply and effectively. The thought patterns that encouraged have been invaluable throughout life, because "knowing what the right questions to ask" in such situations is very much an art. It's a teachable art, but few people blunder upon it all on their own without help.
I heartily recommend the book to everyone's attention. Some things hold up really well over time.
Yeah the basic tension in this is the maximal point that it is by no means clear that Infowars is telling lies is being reacted to in a binary way to defend the basic principle that you can identify falsehoods in some or many cases. I think part of this is to defend the utility of honesty in personal life which feels threatened by the maximal point but is really a different kettle of fish than assessing honesty in reporting about a broad range of subjects that are often filtered through a world model that motivates biases to protect it without the reporter knowing it consciously
Part of the problem here is that "lie" is considered a bad thing, so when someone says something untrue and the listener hates them, said listener can describe the statement using the single-worst word in the "falsehood" Russell conjugation.
The example I had in mind is the large collection of falsehoods that the Nazis stated about the Jews. (Sorry for invoking Godwin's Law here). According to Scott's analysis, the Nazis were not lying about the Jews, since the Nazis believed every one of their antisemitic canards and conspiracy theories. They technically avoid definition #7. Yet we still commonly say that the Nazis lied about the Jews. Why? Because lying is bad and the Nazis were bad, and their falsehoods about the Jews were really egregious and defamatory. It might also be because we have actual, definition #7 examples of the Nazis lying (ie, Hitler promising that he had no further territorial claims on Czechoslovakia beyond the Sudetenland), but careful readers should still note that "The Nazis were liars" and "The Nazis lied about the Jews" are two different statements.
The expression "believe their own lies" shows that the word "lie" is commonly used outside of the strict definition #7 sense. I think that the common definition of "lie" is the union of definitions #6 and #7, and the most negative Russell conjugation of "falsehood" under definitions #2 - #5. When someone says "lie" under definitions #6 or #7, it can involve morally positive examples of lying (ie, a resistance member hiding Jews and lying about it to the Gestapo officer checking in on her house), but when someone says "lie" to describe something that fits into definitions #2 - #5, it is always to denounce.
I think people are generalizing from personal life where an individual has very specific knowledge about events and actions in their life making it hard to say something untrue without being more on the definite lie part of the spectrum to journalistic reporting and general worldview or model dependent ways of making sense of the world
There are two factors required to be able to report accurately. One is the skill to do so and another is the desire to do so. You often need a lot of both and it's easy to fall short one way or the other without consciously intending to. While telling the truth in personal life is rightly strongly held and therefore appealing to transfer on a one to one basis to people who are reporting on a wide variety of issues it just doesn't go one to one
Approximately nobody in the general media is motivated to lie when reporting about fusion experiments yet it's perfectly normal for me to see basic mistakes in fusion articles. You've heard net energy was achived? Big lie. The total energy balance was 1%. You've heard that and therefore fusion is clearly a dead end? Big misrepresentation. Anyone who knows the facts is poorly justified in saying that. The laser technology itself may be capable of 40x greater efficiency and there are other more readily commercialized designs that may actually reach a full net energy gain by 2030. In other words both of the primarily reported facts about the fusion energy gain experiment would be pretty easy to call lies if you thought they were wrong and thought that the reporters had a special bias towards giving inaccurate information about fusion
So now maybe you want to agree with that but also maybe you don't because it takes away the "Just Stop Lying" button that you could just keep pressing hard enough to escape from the Late Pre-truth Era. I have an answer for that. On Metaculus the current prediction for a 100 MW fusion reactor to come online is 2035. That's probably a very accurate piece of information that you won't hear in many other places
> In a perfect world, we would have separate words for all of these. In our own world, to save time and energy we usually apply a few pre-existing words to all of them.
After being distracted for entirely too long with this, I propose the following words for those 7 things:
1. veracinate - from latin verus (truth) and ratio (reason). reasoning truly.
2. misveracinate - now you've fucked up - but you were at least trying to reason truly
3. idioracinate - you're reasoning like an idiot or common man (latin roots)
4. karsioracinate - from the greek karsios (oblique) aka skewed reasoning
5. karsiocept or karsioceive - from latin capere (to take), it's unintentional deception via bias
6. veracept or veraceive - you're taking the truth from people, subverting them with facts
7. deceive - we dont need a new word for this.
But I bet Scott doesn't read the comments 2 days later...
Your 7 categories do not seem to include the most common of them all:
8. Being so certain of the Higher Truth that precise details don’t matter when they obscure that higher truth.
I raise this point because I suspect it drive almost all of this nonsense. Who CARES whether the police kill 15 or 5000 black men every year, “we all know it’s too many”. Who cares whether 1 in 3 or 1 in 300 women is raped, “we all know that all men are potential rapists”.
If you know the TRUTH then you are doing God’s work by limiting the ability of your readers to be distracted by irrelevant details that confuse the point.
The difference between 8 and 7 is that you don’t even believe you are making up something; you are simply reporting the reality and if facts say otherwise, well, soon enough we’ll realize those facts are in error.
I think what you’re describing as 8 is just an unusually intense manifestation of the more general way in which we see human beings react emotionally to certain politically charged topics and then reliably rationalize the moral necessity of the various shades of deception delineated in 5/6/7.
I’d highly recommend HotelConcierge’s blogpost ‘The Tower’ for a far better explanation of this phenomenon and a deeply illuminating treatment of among other things, how our current hyper-partisan, depressingly cynical media climate might be the logical, inevitable outcome of unlocking forces such as the inexorable free flow of information and mixing it with the zero-sum, my-way-or-the-highway instincts of human nature.
I think the difference is that you view people engaged in 5, 6, 7 as thinking "OK, this is a lie, but it's justified". Whereas I view people engaged in 8 as not even believing that it's a lie.
If a Christian talks about the Resurrection, are they "lying"?
OK, now if a Democrat talks about how Trump was a tool of the Russian Government are they "lying"?
In both cases there's zero evidence for the claim that's convincing to an outsider; what does exist is a pattern of being told a claim enough times that the listener accepts the claim as so true that it's now foundational; it doesn't require evidence, rather it IS evidence in terms of which new data should be interpreted. Because we KNOW that Trump is a tool of the Russian Government, therefore such and such a new deal with Putin is obviously structured in some way so as to damage America...
I think different people are arguing aspects of this based on different goals.
Some are arguing with an "ethics" goal in mind (it's not as morally culpable if you didn't make up the statement, or whatever).
Some are, I guess, arguing some sort of "legal" theory of what can you get away with.
But a few (like me) are arguing with an "understanding" goal in mind (what do these people imagine they are doing when they flood the field with these crazy claims month after month, year after year?) It's in this 'understanding' context that 8 is a different sort of phenomenon from 7 or earlier.
We're a community with a self-selection bias that sorts for folks who fancy themselves smart (you know it's true, c'mon), and I hate to break it to you fellas but we have more specific words in English to cover all of these edge cases without blurring definitions. The first two that come to mind are duplicity and subreption.
I think that if we're going to strive towards goals like: cultivating sophistication, encouraging and preserving nuance and enhancing individual sovereignty, we've got to recognize when lazy colloquial language is doing harm to our ability to discuss a particular subject.
I can never keep up with all the comments on your great posts, so apologies if this was already posted as an example, but here's a clip from a CNN report where they are not just lying, but they are deliberately saying the exact opposite of what's actually happening. https://www.youtube.com/watch?v=zfOcz--2AjE
The context is a gathering at a neighborhood vigil after a police shooting of a young man. There was rioting in response to the shooting. The deceased's sister is speaking to the crowd, and the CNN voiceover describes it as, "Smith's sister calling for peace...". The reality? She was encouraging the rioters to go riot in white neighborhoods. (They actually issued a mealy-mouthed apology for it: https://www.buzzfeednews.com/article/davidmack/cnn-sorry-for-not-airing-full-clip-of-milwaukee-sister).
I don't see how this can be interpreted as anything but a deliberate lie, and not just by selectively showing only one part of her words, which on its own would count as seriously misleading, but by characterizing it as they did, it shows a willingness to unabashedly lie to promote a desired narrative.
Scott, I think you are equivocating here. In section IV of your original post you said the point of this was to show that fighting misinformation was hard. How (for the most part), you couldn't get very far simply by removing posts from "liars saying completely false things".
However, if your threshold for labeling something "misinformation" and removing it is not only that what it says (or maybe even clearly implies) is blatantly false, but you also require that the author of the article didn't believe what they were saying at the time, it would be impossible to identify misinformation even if it were prevalent. I mean what do you expect Facebook to do, bring Alex Jones in to take a lie detector test every time InfoWars says something blatantly false?
Any reasonable argument in favor of censoring obvious misinformation is going to label something "misinformation" if it contains prominent claims that are blatantly false or maybe even if it is clearly designed to convince the reader of an obviously false claim. But in any case, your censor probably isn't going to care about the state of mind of the person who wrote the article. So if the ability of censorship to fight misinformation is what you really care about here, you shouldn't care about whether Alex Jones actually believes Obama's birth certificate is fake either.
I think the concept of intent may be part of the puzzle here. Notably, if you assume intent then probably you should assume positive intent, and then the test for lying becomes difficult to fail.
Tl;DR Making false facts is Thing 7 and misreported confidence is Egregious Thing 6, but only if a person is reasoning well. A person reasoning poorly can commit a Thing 4 at worst. But a poorly reasoning person can intentionally make false facts and overstate their confidence, recognizing that they are doing so. Merely noting someone's unreasonable state does not guarantee a lack of intentional deception. Furthermore, a deceiver can often exaggerate their unreasonableness to escape consequences.
You say in Section 8:
|3) Reasoning badly, because you are dumb.
|4) Reasoning badly, because you are biased, and on some more-or-less subconscious level not even trying to reason well.
|5) Reasoning well, having a clear model of the world in your mind, but more-or-less subconsciously and unthinkingly presenting technically true facts in a deceptive way that leaves other people confused, without ever technically lying.
|6) Reasoning well, having a clear model of the world in your mind, but very consciously, and with full knowledge of what you’re doing, presenting technically true facts in a deceptive way intended to make other people confused, without ever technically lying.
|7) Reasoning well, having a clear model of the world in your mind, and literally lying and making up false facts to deceive other people.
|...I prefer to reserve lying for 7 and the most egregious cases of 6
And in Section 2:
|I don’t want to say you’re allowed to sound more confident than you are. If you’re 71% confident, and you falsely say you’re 72% confident, then you are lying. But if you are very dumb, and seeing a random piece of toast makes you 100% confident that Obama’s birth certificate is false, and you vomit some random words to that effect onto a page, then you’re an idiot but not a liar.
I object to this because:
A) Many people simply do not reason well and have a clear model of the world in their mind when discussing controversial topics. Under this framing, a person who is sufficiently mindkilled cannot lie, even if creating a false claim from thin air.
For example, I recently learned that a friend, devastated by a breakup and trying to express her distress, uttered the words, "He raped me". Another friend took her at her literal word, and started to file a report. She immediately realized her mistake and tried to stop him. (There were no charges, but I don't know if she successfully prevented the initial filing.)
Suppose she had decided that her treatment was morally equivalent to rape and should be prosecuted, and had proceeded with the report. It would have been a mere "Thing 4", because her reasoning was compromised by emotion.
I think most intentional falsifications about the world involve a highly unreasonable individual. It can be very useful to distinguish a cold-blooded and hot-blooded deceptions, but unreasonable people can still intentionally deceive. If internal states matter, then the person's consciousness of their deception should be a major factor, even if they are also unreasonable.
B) Under this framing, a person can vastly reduce the apparent badness of a falsehood by merely pretending to have been reasoning poorly. Because we don't have access to peoples' thoughts we can rarely distinguish this. I think this is a major source of contention by the comment section, especially regarding InfoWars.
C) Unless I grievously misunderstand, the original framing was about moderating false reports such that only "lies" are removable. However, it is extremely difficult to determine whether someone is exaggerating their confidence by 1% (Egregious Thing 6). It takes very little bias for a person to overstate their confidence, and (as in B) even if somehow caught, one can escape moderation by blaming poor reasoning. Even in the case of explicit Bayesian reasoning, you could nudge the priors or introduce a subtle math error, and if needed pretend it was because you made a mistake (Thing 3). This form of lie is effectively unaccountable.
I wonder if there should be a term for the absence of Gell-Mann Amnesia. I just read an article about a subject I had personal knowledge of, and it was surprisingly fair and accurate.
Although what's even weirder is one time last year when I read an article on a topic close to me and it was overall fair and accurate but had one offhand statement that was completely false. It was kind of baffling how that could even happen, and it wasn't ideological or bias or anything either. Presumably, the source for the article just misremembered something and they didn't bother to check.
As per Chomsky, this was rarely done by direct lies, in the form of front page “EVERYTHING FINE IN GUATEMALA, SAY SOURCES”
==
By Scott's current definition, if there are indeed sources who are willing to say "Everything fine in Guatemala," that would not constitute "direct lies."
I think the fundamental issue is sentences of the following format:
--
"[completely false assertion]," said [totally dishonest person], who studies [general field of knowledge that the dishonest person told reporter he studies and which relates to the completely false assertion].
--
Under Scott's current definition, as long a newspaper accurately quotes a source, it's never a lie. So you could have a lengthy article consisting 100% of false assertions, but so long as they are attributed to the people who said them, the article has not lied.
The problem is, there are 8 billion people on the planet, and you can find someone to say *anything*. And even Scott in 2015 called things like this "lies."
>This is why I’m so convinced people are erring on the side of too mistrustful - you can fill your articles with sentences about how you’re not claiming X, and people will still find ways to accuse you of lying because you said X.
The problem is that you wrote a post strongly implying it, and that still holds true even if you were careful to insert a couple disclaimers saying you didn't actually think so.
"My main argument against this is that millions of people believe conspiracy theories - if they didn’t, we wouldn’t care so much about them!" This seems like a bit of a tautology to me.
Technically speaking, we might care about them because we *think* that millions of people believe in them...not because that phenomenon actually exists.
And if anything, I also think we're always somewhere between intrigued, horrified & entertained by the idea of people who believe totally ridiculous things. So much so that there there is a kind of psychological bias to *want* to believe that conspiracy theories exist & have loyal believers (which, by the way, I think is part of the actual audience for InfoWars as well...namely all the people constantly pointing at it with some judgemental lens).
So, all that is to say is just that the logic here doesn't totally hold up for me and I remain unclear about how many *actually* believe in any of these things.
From what I've read, the Weekly World News in particular did totally make up a significant portion of its content, but it also went through a lot of local and other newspapers for weird-but-supposedly-true stories and republished them. (The employees thought of themselves basically as entertainers, I think.)
Is part of the thesis "lies in headlines don't count as long as the article doesn't explicitly lie?" Because you lost me in the very first post of this series when the very first example of "not a lie" was "New Vaccine Data Shows Alarming Number Of Stillbirths And Miscarriages Caused By Covid Shot." That is a lie of at least grade 6, probably 7. The data do not (and cannot) show causation. So I stopped reading at that point, because that's what everyone does when an article seems like obvious bullshit within the first few sentences (based on their own individual biases, of course).
Any discussion of lying in the media that ignores headlines is irrelevant.
A more accurate headline for your article would be "The Media Very Rarely *Technically* Lies," but you probably didn't go with that because:
To address the 1-7 taxonomy you presented at the end: I think 6 is definitely lying. I understand that you're making a narrow point about 7 in these articles, but 6 seems to me to fit entirely within the accepted understanding of the word "lying." If I come home after a late night and tell my wife, when she asks where I was, "I was out with the guys, we got a few drinks," but I intentionally omit the context that the drinks were at a strip club and I also paid for a private dance, I think most people would say that I lied to my wife about where I was.
Also, I think we're actually pretty close to having separate terms for each of these numbers. 1 is just being correct, and 2 is just being incorrect in the most straightforward way. 3 is also being incorrect, although you might tack on a pejorative like "stupidly". I think 4 would rightly be called "willful ignorance," although we could certainly get into a debate about the implications of the word "willful." 5 is the only one I don't think there's a good concise term for. (And of course I think both 6 and 7 would be "lying," although 6 contains kinds of lies that have more specific names, like "lies of omission," and "statistical lies.")
I didn't really know about substack until I stumbled on a link to the original post of this series. No idea how many blogs are of this quality or if it's just a few, but I feel like I've warped into a place on the internet where people are not only smart and reasoned, they make me feel downright dumb. I legitimately thought I was smart before.
< humor > If we don't think your post said what its title self-evidently means (to us), does that give us a license to accuse you of lying? < / humor >
"Scott isn't *lying* to us when he says the media doesn't lie, he's just holding himself to the usual media standards" :P
"Sources say media seldom lies"
As with newspapers, the headlines here aren't chosen by Scott, they're instead decided by editors who may have very different vibes intended than the actual content.
...would have been the funny addendum to >Please don’t have opinions based on the titles until you’ve read the posts!
I really hope that headlines for articles posted here are chosen by their authors, rather than anyone else. I particularly hope they aren't chosen by "editors" with goals unlikely to match those of the authors, never mind the readers.
There are more than enough sources of click bait headings out there already. The medium term (months) result is that people remember "links to this site are unreliable" and stop visiting the site. Scott really doesn't need editors driving his potential readers away.
It was intended as a joke, though I do notice Scott tends to have an unusually high proportion of sub-optimally titled posts relative to other Substackers. Others it's impossible to imagine anyone else doing the titling, because the titles are so...on brand. FdB's post titles are always just as rude as the content, so I can tell he writes those. (Said with approval, does-what-it-says-on-the-tin.)
Sometimes I think Scott just likes to/unintentionally subconsciously bury the lede a lot. Like this threepeat series where The Moral Of The Story is the difficulties of setting censorship bright lines. But that's not remotely obvious from any of the titles, and one must actually read the posts in full with a bit of attention to grasp the thesis <-> supporting arguments distinction. Thousands of comments debating the supporting arguments, not many grappling with the obvious thesis. Maybe because it's too banal, what with rationalists generally being opposed to censorship/"mis-disinformation" already? So we fight about what's more interesting instead.
("You can't blame Scott for writing these articles, he's just throwing red meat to his subscriber base...")
Honestly, the fact that so many people are having trouble grokking Scott's actual thesis is 100% his own fault. These posts remind me a lot of the frustration I get with Ayn Rand's use of the word "selfish" (or philosophers' habit more generally of taking ordinary words and giving them weird, idiosyncratic definitions that the reader has to dedicate half their mental energy to keeping straight): sure, I get what he's trying to say, but I don't understand the doubling and tripling down on a rhetorical strategy that clearly isn't working.
If "the virtue of selfishness" really means "there is nothing inherently moral about self-sacrifice; don't feel ashamed for living your life according to your own values," and you're not telling everyone to be anti-social pricks who lie, cheat, and steal their way through life, maybe it's counterproductive to keep using the phrase everyone assumes means the latter, and the time you're spending defending your slogan could be better used making actual, substantive arguments. (Although I get the impression Rand was being deliberately inflammatory for the attention, because she was at least as much a public figure/celebrity as she was a philosopher).
If I were to be similarly uncharitable to Scott, it almost seems like he's using deliberately inflammatory headlines to boost engagement. I'm sure he's not, as that doesn't jive with everything else I know about him from reading his blog over the years, but it's frustrating that he won't just admit that "yeah, it wasn't the best title. My definition of the word 'lie' is unusually narrow, and I probably should not have started a blog post intended to convince my progressive friends that it's not as easy to distinguish the kind of lying Infowars does from the kind of lying the NYT does as they think it is, with a headline implying you should be ~more~ credulous of the media."
I understand what he's saying, but he insists on saying it in a way that's almost guaranteed to confuse people.
I agree that these kinds of flare-ups are unseemly and unnecessary; it's sad when a blogger and their commentariat get into petty split-hair fights, and both sides have too much pride to admit they fucked up or went too far. Somewhat excusable for pundit-type bloggers whose true profession is Take-Slinger, less so for professed Rationalists. Principle of Charity doesn't extend to cover bullheadedness.
It's also still weird to me because...I use a similar truth taxonomy as Scott? A really shocking amount of Obvious Nonsense is, in fact, true[1]; it's challenging to litigate intent and ignorance if we aren't clear on whether we're judging what-the-text-actually-says, versus what-the-typical-takeaway-is. Not just in a legal-censorship way, but it's where so much of the difficulty with Bounded Distrust[2] comes from. It's like the Lizardman Constant: one must assume that everyone, of all persuasions, is trying to corral inconveniently-flexible facts into a narrative...that's the background level of noise in any news communication. One's priors will be weighted differently for favoured speakers, ingroups vs outgroups, certain topics and not others...but you still gotta Do The Math in each case, or the whole exercise falls apart.
Plus I think it shouldn't be as confusing for longtime readers, since Scott's been making more or less these same nitpicky points for years now. The old SSC posts "Against Lie Inflation" and the "Lies, Damned Lies, And Facebook" series, for example. I sometimes think there's been a bit too much Rally-Round-The-Schelling-Flag effect wrt NYT, so even though there's a very large deserved amount of incredulity, sometimes tribal defensiveness leads to excessive...I wouldn't say "paranoia" like Scott did, that's too combative. But like, confusion of the narrative-news ecosystem we actually have, and a Platonic Ideal of News Media which somehow turns a profit via telling The Truth, Only The Truth, And Nothing But The Truth. Businesses sell what people are willing to buy, and it's not unvarnished distortion-free facts.
I don't know. I just feel there is some real and important inferential distance between unreality-as-stated, and creating-the-illusion-of-reality-via-sophistry-and-misdirection. To me, the former is a much more egregious epistemic sin than advertising an inaccurate map. Scott didn't do a great job making that point clear in these posts, but it's an important point nonetheless. How does the apocryphal Soviet saying go..."Pravda is useful because it always lies, the NYT is less useful because it sometimes tells the truth"? I think MSM skeptics would rather it be Pravda all the way down. Sadly, things are not so convenient - so we must be *more* vigilant, not less. (I do fully agree the headline is a dud, misleading at best. Hence my joke. Scott doubling down by saying we didn't read the subtitle is...well...silly.)
[1] In the classic sense: https://www.readthesequences.com/Your-Strength-As-A-Rationalist
[2] I liked Zvi's attempt at formalizing this solving-for-the-equilibrium. Specifically, if the obfuscations and distortions are the weak cliche ones you'd expect - that's simply the best they could afford: https://thezvi.substack.com/p/how-to-bounded-distrust
Great points. The annoying thing is that I also think I use a similar truth taxonomy as Scott. Zvi's method is actually remarkably close to how I already approach the news. I don't think I actually disagree with much of anything Scott's said in any of the three posts on this topic he's made. If I really zero in on what about the posts bugged me, it's that Scott didn't use sufficiently critical words to describe people he fully admits are doing bad things. It's kind of a strange and petty criticism when I lay it out like that, but it's interesting to see in myself (and many other commenters) just how much "I agree with you who the bad people are. Your feelings are valid" is something I'm looking for here. (I think this also explains some of the "you did not just draw an equivalence between the NYT and Infowars" people).
Scott may have accidentally stumbled across this community's scissor statement. The original post managed to call everyone dishonest without calling anyone a liar, which managed to piss off just about everyone, left, right, and center.
I think that I consider it qualitatively different to employ a word in a narrow, idiosyncratic way when you're trying to expand that definition beyond the original scope, rather than fighting to maintain a semblance of the original boundaries. It's very difficult, nowadays, to construct any argument for preserving the utility of certain words without meeting opposition that dismisses the effort as naïve (i.e. you can't fight the colloquial erosion of specificity, so why even bother?)
The major problem I have with this position is that it seems to completely miss the point. Influencing the colloquial evolution of this language is exactly what any thought-leader is trying to do when writing a post like this. Accusing them of needlessly discomfiting people is, in a backwards sort of way, acknowledging their success.
After all, this sort of discussion is a small part of how the vastly complex memetic landscape is shaped.
"maybe it's counterproductive to keep using the phrase everyone assumes means the latter"
I agree with this in principle, but sometimes it threatens to become arbitrarily hard to have a conversation about topics *without* trying to redefine a word.
Take "capitalism", which I've observed a lot of left-leaning folks to use as a synonym for "greed" (sometimes as "systemic greed"). I get why they're doing it, but there's a perfectly good existing word for it ("greed"), and I'd wish they'd back off from trying to give a different perfectly good existing word a permanent synonymous connotation. It makes conversation difficult.
Arguably, this is exactly what you're saying - we should avoid redefining words. But in this case, redefining it worked (for that tribe), and it bricked a perfectly useful word in that sphere of communication.
Which is to say a few things: (1) I think it's hard to guess when attempting to redefine a word will work and when it won't, (2) I think it can sometimes be a good idea to try and redefine words even if it seems likely to fail. (Although in my example, I think people have managed to partially evade the issue by using the term "free markets" instead.)
Re: Scott, I previously commented that (in these circles, at least) I think it's mostly a question of which definition of "lie" is useful for your model of the world. I share Scott's definition of "lie" and it's been useful to me and my interaction with people to have that distinction in my head, but I can easily imagine situations where that isn't true.
Do you really think that when the average person hears "the media is lying to you", they interpret it to mean "the media is presenting you with information that is technically accurate, to the best of their ability, but is leaving out important context in a biased manner"? I would predict with very high confidence that this is false, and that Scott's idea of what "the media lying" means is the normal one. I could *maybe* believe that people who talk about the NYT "lying" are usually applying a broader standard than Scott is here (although even then I'd place roughly 80/20 odds against), but I'm almost 100% confident that this isn't true in the context of "alternative" media like InfoWars.
Scott's idea of lying isn't even *Scott's own idea of lying*. At the end of this very post, he endorses sometimes calling "#6" "lying", and yet that directly contradicts the definition he bases the rest of his argument on.
The problem is that the thesis is so watered down as to be utterly trivial and meaningless (as the comment about Weekly World News showed).
At what stage of going over the comments do you develop your list of categories?
I guess my issue with this series of posts is that the headline initially sounds like an interesting, debatable topic, but then the word "lie" is used in such a narrow sense the conclusion is meaningless. A post titled "The Media Very Rarely Completely Invents Facts Out Of a Whole Cloth With Zero Relationship Whatsoever With the Truth" would be a pretty boring and pointless post, and that seems to be the argument being made here. By this definition of "lie", even people like Trump or Andrew Tate or whoever are not lying, because they can point to some underlying shred of fact or say it's technically an opinion or they are passing on something they heard. That might be true, but who cares? If you define any term that narrowly it won't cover much.
"He's a liar, but not a fraud" might be a useful concept here.
Agreed. Scott's definition of lie is very narrow. He sees to be claiming that if an article offers even a smidgeon of support for an assertion, no matter how weak or irrelevant, then people shouldn’t call it a lie. But this clearly leads to absurd results. By this standard, even an article that says something like 'John is a murderer' and gives as its only evidence for this claim 'John seems like the kind of person who would be a murderer' would not be classified as a lie.
There's a point where an argument is so incredibly bad and misleading that it's equivalent to a lie.
Jesus Christ, why does Scott even bother writing anything? We're three posts and thousands of comments in!
But has Scott considered that his definition is too narrow? Maybe he should put some caveats in the title or subtitle.
But the whole point of this is that there are a lot of ways to convey an incorrect understanding of the world that aren't overt lies. And this is important because it tells you how to read the New York Times. If you know basically what narrative they're looking for, and you know that they will not lie but will also maybe leave out stuff or shove stuff to the end of the story to avoid messing up the narrative, this makes the NYT's reporting more valuable.
For example, consider: "Oh, wow, there was a really serious shooting but the article never mentions the race of the criminal and never shows any pictures even though he's in custody. I guess he's black." This wouldn't be a sensible inference except that most major US newspapers go out of their way not to mention the race of criminals or show their faces when they are black, apparently as an overt policy of trying to avoid supporting stereotypes.
Reading a story about a hate crime against a Jewish or Asian grandma in a big city where the race of the attacker is not mentioned, similarly, well, you can infer the race of the attacker by knowing the narrative the newspaper prefers. If the perp was a MAGA-hatted white guy, there would be a dozen pictures of him in his MAGA hat. If the perp was a crazy black homeless guy who beat up the grandma because the little voices told him to, the reporter would just accidentally somehow not ever mention his race and it would turn out that there was no room for a mugshot.
What you're talking about here is related to Bounded Distrust: https://www.lesswrong.com/posts/rvpEF2mBLeZE9j53n/how-to-bounded-distrust
I was thinking about that post (and probably should have mentioned it) when I wrote my comment.
Right. The New York Times is a tremendous resource, _especially_ if you know how to read it perceptively.
"If the perp was a crazy black homeless guy who beat up the grandma because the little voices told him to"
Speaking of ways to convey an incorrect understanding of the world:
In the US (with presumably similar numbers elsewhere), people with mental illnesses are FAR more likely (like an order of magnitude) to be the victim, rather than the perpetrator, of violent crime.
Iirc, they also make up a smaller percentage of perpetrators of violent crime than they do of the general population, ie mentally ill people are "underrepresented" among violent crime perps.
You could have easily given your example of race reporting in violent crime without bringing the issue of mental illness into it.
I think you should have.
You recall incorrectly. People with schizophrenia are considerably (4 to 6x says the first study when Googling, and corroborated as approximately correct by subsequent studies.) more likely to commit violent crimes than the general population. The fact that they're more likely to be victimized by it than commit it is a sad, but irrelevant fact.
Of particular predictive power for violence is, in addition to psychosis, substance abuse and command hallucinations. So if you encounter a crazy black homeless guy who has voices telling him what to do, he's much more likely to attack you than the average, random person. Though most people, regardless of all those factors, are nonviolent.
Having an explicitly irrational actor as the hypothetical illustrates the point more effectively, if you ask me.
Also, if it matters to you, I write this comment with a person diagnosed with a psychotic illness lying in bed next to me. You don't need to ignore reality to have compassion.
Now do "assault weapons."
I could have also given the example without incorporating race or homelessness, which work similarly--blacks and homeless people are overrepresented among both criminals and their victims. The whole point of my post is that media sources will omit a lot of possibly-relevant information (like the fact that the dude who beat up a Chinese grandma while screaming ethnic slurs was a crazy black homeless guy) in order to either support their preferred narrative, or to weaken some narratives that support stereotypes like that a lot of crime is committed by blacks, homeless people, and (some kinds of) crazy people.
But what I'm looking for in an information source isn't someone to sculpt my mental landscape in a socially-beneficial way or to fight back against harmful stereotypes by omitting relevant information in their stories. Instead, I'm looking for sources that will accurately inform me about the world. So if the Chinese grandma was bashed by a MAGA hat wearing Trump supporter, I want to know that, but if she was bashed by a black homeless dude who was in the middle of some kind of psychotic episode, I also want to know that.
He literally did.
"With a title like that, obviously I will be making a nitpicky technical point." is literally the subtitle of his first post.
He didn't even sneak a "the the" in there, which, as we all know, would definitely have rendered that entire subtitle invisible.
This thread is my punishment for attempting to convey sarcastic irony on the internet. Unless I'm missing your sarcastic irony here.
I do not think completely missing the obvious subtitle in the first post and then ignoring the explicit mention of it in this article is more absurd than people reading a text that constantly repeats in slightly different ways "The media will very rarely lie in the sense of explicitly saying falsehoods, instead, they will convey misinformation by proxy, through quoting biased sources or signal-boosting some evidence while ignoring other" and then linking to a news post of the media doing just that, and claiming it somehow disproves the central point.
If people are lacking enough reading comprehesion to comment such things, I don't think it is unreasonable to imagine they also didn't see the obvious subtitle.
At the end, even Scott tacitly admits his definition is too narrow, given that he himself endorses calling "#6" a lie in some cases.
> By this standard, even an article that says something like 'John is a murderer' and gives as its only evidence for this claim 'John seems like the kind of person who would be a murderer' would not be classified as a lie.
> There's a point where an argument is so incredibly bad and misleading that it's equivalent to a lie.
Perhaps, but the standard you describe doesn't reach that point, or come close.
The problem with your example is that it's not misleading at all; when you say "John seems like the kind of person who would be a murderer, so he must actually be one", no one is going to believe you.
> Agreed. Scott's definition of lie is very narrow.
This is not Scott's definition, he's using a standard definition of "lie": https://www.merriam-webster.com/dictionary/lie
1. To make an untrue statement with intent to deceive (Scott's 7)
2. To create a false or misleading impression (Scott's 6)
Unfortunately #2 also includes "bias", so Scott says he reserves only egregious examples of that as the media lying. Totally reasonable.
From the same source, "Love" https://www.merriam-webster.com/dictionary/love
(1): strong affection for another arising out of kinship or personal ties
maternal love for a child
(2): attraction based on sexual desire : affection and tenderness felt by lovers
After all these years, they are still very much in love.
(3): affection based on admiration, benevolence, or common interests
love for his old schoolmates
If I were to say, and insistently repeat, "Parents should not express love for their children; it's not good for the children and if at all common is a sign of social degeneracy", I will probably cause a great deal of confusion and get more than a few people angry even though I am pedantically correct using *a* standard definition of "love" (#2).
If a thing is clearly true for one "standard definition" of a word, and clearly false for another standard definition of the same world, and clearly contentious for many people, then it might be best to chose a different word or phrase.
I agree that Scott's definition of a lie is too narrow. Experienced liars agree with Flashman: Suppressio veri is a useful servant, but Suggestio Falsi is a perilous master. Experienced liars are still knowingly telling lies of omission. But mass media is by definition a choke point, unlike social media. No news story can include everything. JD Bernal went so far as saying mass media is always censorship.
So we should not expect more from any mass media than we expect from advocates in court. If I'm on trial I don't want my defender to include every possible way I'm guilty. If I'm a D party loyalist I don't want the New York Times to include President Pedo Peter's bribes from Ukraine or anyone else, much less track down the underage naked girl Hunter tied up and photographed on his laptop to check if she's still alive. If I'm an R party loyalist I don't want Fox calling Arizona for D party before they have to, much less point out every last Trump stretcher.
Lying has a simple and precise definition: "Making an assertion which one does not think is true."
By this definition, the media rarely lies. It often deceives, however, and I would argue that that is sort of worse, because deception (lying using truths) is trickier to refute.
Trump espouses outright falsehoods so often and so strategically that I think it is more than reasonable to accuse him of lying. Now Andrew Tate may, in fact, be so stupid that he believes his claims—I genuinely can't tell. In any case, an assertion does not need to rise to the level of a lie to be harmful. In a certain sense, true believers in falsehoods are more dangerous than liars, because they spread untruth with a relatively clear conscience.
Wouldn't that precise definition be, "Making an assertion which one believes is not true"?
I ask this because you brought up the example of Trump. He generally seems to be simply indifferent about whether his factual assertions are true or not. If the assertion helped him pursue his objectives then it was worth saying, and if it didn't then it wasn't. That seems to be his only criteria.
Does _indifference_ to truthfulness count as lying? Seems like it should....but maybe that is actually a different category.
That's a great point. I think you're absolutely right. Wherever there is positive indifference to the truth, you can't really call it lying.
https://en.wikipedia.org/wiki/On_Bullshit
I always thought that the above book, due to its rather crude title, was on the intellectual level of those cringy self help books like "The Subtle Art of Not Giving a F*CK!!!"
But just at a glance, I think that Frankfurt's category of "bullshit" tracks exactly with what you're saying. The book seems pretty relevant to the whole discussion, actually.
The media doesn't lie. It either deceives, or more often simply pushes bullshit.
There is something sort of similar to the cringey self help books you mention, though.
I would say "making an assertion that one does not believe is true" rather than "making an assertion that one believes is not true".
That's sort of isomorphic to John's "making an assertion which one does not think is true", which is precisely what Paul aimed to correct with "...one believes is NOT true."
Law of excluded middle is being assumed here.
I'm a software engineer. There's a usually a deep cultural rift, to put it mildly, between the engineering and the sales sides of every software company. The two groups don't understand each other and have quite visceral distaste for the other.
Trump is a salesman, period. And that's one of the many things, perhaps the main thing, that people who can't stand Trump can't stand about him. The fact that Trump, who was the President of the United Goddamn States, and therefore responsible in at least some way for what happened in the lives of ~300 million Americans, if not in the lives of ~7 billion humans on Planet Earth due to America's outsized influence in the world, just didn't give a shit if the things he said were true or not is beyond infuriating to the "engineering" mindset.
+1
People for whom "true" and "false" are not important categories set my teeth on edge. And that's basically all PR people and most politicians, but Trump dials it up to 11.
Tied to the "salesman" concept is another important truth category that seems to be neglected here: statements that the speaker hopes will become true as a result of saying them. Self-fulfilling prophecies.
As someone in the finance world, the most obvious one to me is this: a bank CEO says "Our bank's financial position is fundamentally secure." The vast majority of the time bank CEOs say this, and the vast majority of the time it is correct.
Now, let's imagine one of those same CEOs were to instead say, "Our bank's financial position is highly dubious. I don't know if we'll survive another day. If you're exposed to any sort of counterparty risk or other financial risk from our bank's collapse, then you're about to get completely screwed over, sorry."
That statement, too, would very likely be correct as a very result of the bank CEO saying it.
CEOs at places like Lehman and Bear Sterns were accused of lying when they said their firms were secure, even though they were actually on the verge of collapse. Which, while I don't exactly have sympathy for those guys, I also don't think is entirely fair. They said the only thing they could say, out of HOPE that it would be true and knowing that if they didn't say it, they were certain to fail.
Back to a more proper salesman: imagine Steve Jobs. "This product is insanely great." People said he had a "reality distortion field". At least part of what made his products insanely great, they'd say, is the strongness of his assertions. Maybe, if he hadn't made such strong assertions about Apple's greatness, people would actually have rated their products lower. Maybe they would have derived less enjoyment from them, fewer endorphins released.
This is even clearer when you go beyond consumer electronics (which have both a fashion and function component) into pure fashion. There it's all just salesmanship, really. That dress, those blue jeans are the bee's knees because we have salesman'd our way into convincing you they're the bee's knees. If we hadn't convinced you, then we'd be liars.
It's an old American tradition. "The Power of Positive Thinking" was a vast bestseller by Norman Vincent Peale, Trump's childhood minister.
It's intriguing how Peale was influenced by Christian Science, which was itself influenced by Hegel. In other words, there is a direct line of ideological descent from Hegel to Trump. That prussian is truly at the root of all modern philosophy and thus, all modern errors.
Oddly, what seemed to upset people the most about Trump were his true statements: Mexico is not sending their best, Norway vs. Haiti, etc.
It's not like "Mexico" was/is actually "sending"¹ anyone, so yeah, clearly true!
And I don't think that's particularly odd: for known-to-be-false statements, they can point at the facts and say "see, this statement is false".
For the disliked true statements, sometimes the only thing his opponents could say was some sort of "hey, don't you know you're not supposed to say that out loud & in public?"
...which seems a lot more upsetting and frustrating as a position.
¹In the sense Trump meant here. Presumably the people Mexico actually *does* send to the US (eg their ambassador and embassy staff) are in a sufficiently high percentile for at least some relevant dimensions that they could reasonably be called "their best" (if not MIB "the best of the best of the best").
The implication here is that nations are no more than their governments, which I'd disagree with. There was still such a thing as Italy when Dante was writing (in fact, it's a word that he uses). Of course, governments can help forge nations -- there is probably more of an Italy today, some 150 years after unification, than in Dante's day. And if Italy were to split into separate states again, or be subsumed into a European superstate, there would probably be less of an Italy 150 years after that.
Mexico as a nation has been sending large numbers of people northward for decades, which the government in Mexico City has regarded with approval (mainly because of the economic boost provided by remittances) but hasn't done much to directly facilitate.
This, to me, comes off as odd.
On the one hand, I have an engineering degree, and have a history of getting pushback from people above me because I insist on things being technically correct. I've had bosses complain when I've provided something that precisely met their specifications (but not what they actually wanted) and teachers punish me because I would insist on correcting their grade-appropriate but not technically true answers.
On the other hand, I grew up in the DC suburbs. News of politicians and political appointees say things which are not technically true is background noise here. Trump was not an outlier (or out-liar) in this respect. Our current president has a long history of telling fantasy stories, and if he's not deliberately lying when he does it now, the alternative is significantly scarier. We have "You can keep your health insurance", "I did not have sexual relations with that woman" and "I am not a crook!" from past presidents, plus countless examples from lesser figures.
The problem is that Trump spoke outside the overton window. I don't the window actually has a veracity condition, it is just about social acceptability. So Biden's lies are largely banal and within historical bounds. It isn't the lying, it is the social transgression.
Indifference to the truth is a category that Scott completely overlooked in his taxonomy at the end of the article, and it's really important. A huge amount of false information comes from this category, because for most people in most situations the truth is besides the point.
One thing that should concern us is that the cost of generating truth-indifferent information is dropping close to zero thanks to ChatGPT and other systems like it, because they are, by their nature, indifferent to the truth. They have no idea what's true or false, but they know how to generate text that sounds convincing. Expect the future to include dramatically more bullshit.
For more details on these ideas, listen to Ezra Klein's podcast from 5 days ago: A Skeptical Take on the A.I Revolution.
Agree – a good way to understand what is going on with a lot of media articles is to forget about true and false, and look instead at what the author feels, and what they want the reader to feel. That's a whole extra layer of analysis that Scott is neglecting in these posts.
I believe most of Trump's "lies" are hyperbole, which often bypasses detection in the same way sarcasm can, and thus is perceived to be a lie. Yes, there are cases when he seems to have lied, such as when he claimed during one of the debates that insulin could, in the future, be as cheap as water. But when he claimed, for example, that COVID was totally under control, that wasn't a lie, I believe, but the normal stuff said by all politicians.
I guess the point is that Trump's claims are not just misleading (which is typical for politicians), but positively false to a degree which is unprecedented in American politics. See here:
https://en.wikipedia.org/wiki/Veracity_of_statements_by_Donald_Trump
The recurrent insistent claim that the election was stolen, for example, is totally groundless. Many such cases!
It has a lot of well-collated information. Some comments:
"In his 2016–2020 financial reports, he claimed that the Trump Hotel in the Old Post Office Building in Washington, D.C. had revenue of over $150 million. In 2021, the House Oversight and Reform Committee revealed that, to the contrary, the property had a net loss of $70 million during that period." I checked the source on this, and yes, that is all it says. Assuming the quote is entirely true, there is nothing that is a lie in it, if it also had $220 million in expenses.
"During a 2018 interview, television personality Billy Bush recounted a conversation he'd had with Trump years earlier in which he refuted Trump's repeated false claims that The Apprentice was the top-rated television program in America. Bush recalled Trump responding, 'Billy, look, you just tell them and they believe it. That's it: you just tell them and they believe. They just do.'" This sounds very typical Trump. Basically, when Trump says something, one should interpret it as if a politician said it, but more so.
"When Schwartz wrote The Art of the Deal, he created the phrase 'truthful hyperbole' as an 'artful euphemism' to describe Trump's 'loose relationship with the truth.'" Again, it takes a special interpretation. Once one gets used to speaking Trump, which ought to take perhaps five minutes, one can see at least what he's trying to say.
But don't expect promises to be kept the way they are stated. If he says he guarantees something, he means he emphasizes its importance. If he says something will make money, even something like "huge amounts of money, tons of cash", it doesn't mean it will be profitable. If that's lying in your book, then so be it, and I won't argue that.
Obama promised "hope and change" without specifying whether the change would be good or bad. Here in Michigan, Jennifer Granholm promised we would be "blown away" by her second term if she were reelected. She was right. I was astonished people reelected her with a line like that. And now Gretchen Whitmer was reelected after promising to "fix the damn roads" and not doing it. Trump may be more blatant, but I don't see him doing anything really out of line with other politicians.
That's always been my problem with the "Trump is a liar" remarks. The proper comparison set is *other politicians* and that set is the absolute worst set to look at and claim Trump is an outlier. I think Trump lies a lot (or has a very loose relationship with truth to the point of indifference), but so does Biden, and both Clintons, and so on. If you put Trump up against Jimmy Carter and call Trump a liar, I might believe it and take it seriously. You put him up against "Wipe it, you mean with a cloth?" Hillary Clinton, I just roll my eyes at both.
Trump is I think an outlier in that he tells *blatantly obvious* lies, demanding that his followers accept them as proof of their loyalty and ingroup status, and does not defend himself against the obvious outraged rebuttals except with repetition of the original lie. Other politicians mostly tell lies with the intent to deceive and with some plausible-deniability level excuse to hide behind.
I take issue with the claim that Trump is only somewhat worse along the scale of normal politician dishonesty.
Obama promised "hope and change," which isn't a statement whose truth value one can easily validate in the first place. This is typical of politicians. Gretchen Whitmer promised to "fix the damn roads." This is normal for politicians, but they do not usually have the power to unilaterally get things done while balancing other priorities, and this sort of electioneering essentially amounts to overselling an offer to make this a priority in office. This is behavior that could reasonably be described as dishonest, and is a good example of why people usually have low opinions of the honesty of politicians.
But before Trump, most politicians have been very wary of making statements that are non-weasel-wordily, unequivocally false, at the moment of their making them, in a way which could be positively proven to be the case. People interpret that very weasel-wordiness, which allows politicians who make misleading statements to backtrack or justify deeply misleading statements as not having been lies if they're caught out, as a sign of dishonesty, because they're so avoidant of clear and unambiguous assertions. This was an advantage for Trump, many of whose supporters saw him as more honest, because he was prepared to make clear, unambiguous assertions which were flatly false.
Saying that he had closed his Chinese bank account when he had not, or that he would publicly release his tax returns after election and then not only not doing so voluntarily but spending his whole term in office fighting against being compelled to do so in court, are dishonesty of a sort that politicians before Trump have rarely exhibited. Just as the media rarely "lies," according to a strict, narrow sense of lie, politicians also rarely "lie." Trump is an exception to that norm, and took advantage of the public's expectations of bounded dishonesty among politicians.
Yea, this. Trump is much more specific with his lies than the professional politicians and also prolific, often weirdly so. And then once he's done it, if challenged he invariably doubles down. For years I thought that Bill Clinton was the most dishonest person to get in or close to the Oval Office during my adult lifetime, but Trump makes Slick Willy's deceitfulness seem almost quaint.
Most politicians lie or exaggerate or spin towards some specific purpose, to cover something up or change how something is perceived or harm their opponents or whatever. And Trump certainly does that stuff. But then he also just literally makes shit up, without being asked, and spews it out. Have you ever attended one of his rallies? Friends of mine have and one of them once made me sit and watch like 40 minutes of live video that she'd recorded -- Trump stood up there at the microphone for that whole time and literally every other sentence included a flat-out lie. About all sorts of random stuff besides big issues -- personal stuff, and even just random topics.
You don't see a problem with using wikipedia for such a claim? WP:RS explicitly gives greater weight to the NYT and WaPo.
I just now checked WP:RS, and the string "Washington" does not appear in it. "New York Times" appears twice, but in the footnotes.
I see a problem with using Wikipedia as well, but I can't say it's because of "explicit" weight given to NYT and WaPo.
I think the giant lie about the stolen election is far more egregious than the distortion most politicians have hitherto peddled. That suffices for me.
https://youtu.be/uoMfIkz7v6s
<i>Lying has a simple and precise definition: "Making an assertion which one does not think is true."</i>
I'd add something about having an intent to deceive, otherwise actors, fiction authors, and people telling jokes would all count as liars, which seems implausible.
In such cases we do not make assertions (that is, we do not put forth a statement as true).
https://www.oxfordbibliographies.com/display/document/obo-9780195396577/obo-9780195396577-0148.xml
I was thinking: what about innocent pranks? Where we pretend that something false is true—that is, we momentarily deceive in a minor way.
I think that because these pranks involve lying, they ought to be avoided.
He also did tell you that he's making a very nitpicky point. So you probably shouldn't have expected a broad debatable post.
But 2, we apparently did get a post that a great many people wanted to debate, considering the number of comments we've had on it and its followups.
And now you know the secret to writing a successful blog!
I think the reason that the extremely narrow point is not actually vacuous is one the article makes a strong case for, it's just not a case that's compressed into the headline.
If you have a goal of censoring false and misleading news, it's extremely salient to that goal if very little news meets strict and uncontentious criteria of "false," and which news one regards as "misleading" depends overwhelmingly on one's political affiliation.
There's news that pretty much everyone, liberal, conservative, or occupying some position outside the left-right divide entirely, can accept is simply factually false, and our information landscape is not impoverished by rejecting it. Unfortunately, once you remove all of it from the equation, most of the actually contentious news is still left, and if you resolve to censor "fake news," you're necessarily going to have a big fight about what it is. If any platform chooses to censor "fake news" according to its perception of what qualifies, that perception is almost certainly going to reflect their own biases, because in most cases, determining what is or isn't "fake" relies on heavily value-laden judgment calls.
By more expansive definitions of "lying," I think that both left and right-aligned news media "lie" fairly often, but as a society, I don't think we have very good means to impartially arbitrate when they're doing that.
YES! This is how I interpreted Scott’s posts from the beginning, and I felt like I was taking crazy pills that no one in the comments seemed to get this. I’m glad I’m not alone.
There are dozens of us!
Dozens!
Right. Also, censoring all news media that engage in sloppy or motivated reasoning, omit relevant facts to tell the story they want to tell, imply far more than they can prove, etc., will mean censoring basically all the news media, most politicians' statements, a large fraction of official government statements, etc.
Very few comments here would survive, either. It's not constitutionally possible to censor the fuck out of "news media" but leave ordinary blog commenters untouched. The First Amendment does not distinguish between the New York Times and Joe Sixpack keyboard warrier banging out a righteous takedown of whatever the guy above him has said. If it's illegal for the first to engage in sloppy or motivated reasoning in print, then it's illegal for the second, and for everybody, large or small. Conversely, if it's protected speech, it's protected for everybody.
Nit: it's constitutionally possible to establish a censorship office that doesn't have the resources to touch more than a tiny percentage of blog commenters and whose internal incentives are best satisfied by using whatever resources it does have to go after the biggest, juiciest targets. I wouldn't recommend it; even if one would consider that end result desirable, there are too many ways for the implementation to go badly wrong. But it is possible.
You're arguing prosecutorial discretion is constitutional? I don't disagree, but you still have to criminalize what you don't intend to prosecute, you can't differentiate in the statute itself. Exempli gratia, the fact that the 55 MPH speed limit was never seriously enforced in Montana doesn't mean it was therefore not illegal. And if you had the bad luck to be one of the few drivers who got a ticket, it would do no good at all to argue to the court that so-and-so blew past a trooper doing 85 and the cop merely waved hello.
Looks to me as just another instance of the decoupling/contextualizing divide. To some people the narrow point that the NYT and InfoWars can be described as committing the same sort of sin is irrelevant, because they and their allies consider themselves capable of reliably telling the crucial difference anyway.
All this discussion about "people want to censor X" would be greatly improved if people would specificy *who* they imagine is doing the censorship in their scenario.
Like, the government needs to be extremely cautious about punishing all but the most obvious and disprovable lies, because when the government acts it's threatening someone with fines and jail. But if I'm a Reddit moderator, I can be a bit more free about saying "I will take down vaccine misinformation, using my own judgement about how misleading your posts are" because the harm of being unfairly banned from a subreddit is far less.
In recent context, I would say that this is Twitter (or other social media) specific at the least. That Twitter weighed in on the side of certain viewpoints (generally towards the things the federal government was pressuring them to support) and against alternatives - including alternatives that were plausible at the time and have since been proven true.
I thought something similar at first, but then I realized that in the context of whether or not we should censor “lies“, it’s actually a very good definition to use.
It's the *only* definition that is useful to the larger argument. Scott was not originally trying to argue in the first two posts that people have the wrong definition of lies, but rather, that we are going to speak specifically about things whose veracity can be objectively determined. It was probably a mistake to call those "lies" given what he was trying to argue.
However, at the end of *this* post, Scott seems to be arguing for exactly what everyone has been disagreeing with him about for the other two posts, so maybe they were onto something all along.
You're right, he did. That definitely weakens his argument a bit, but I guess explains his original framing more.
No it's not, because if it's the definition you use, the censorship stops being useful, precisely because it's not screening out much - and certainly not the stuff the would-be censors *want* to apply it towards.
Yes, I think that is the point. Those who wish to censor want to have (and imagine they will have) a much larger effect. I believe Scott's point is that anything approaching "objective" censorship will have a very small effect, and thus, to get the larger effect most censors seem to want, one must of necessity enter the realm of interpretation, with the inherent cost being the censors' priors significantly influencing what counts as "lying" and that those in power have increased incentives to further expand the definition of "lying" because it will serve their ends.
If you want to sign on to a censorship regime it makes very much difference what counts as a lie and what does not. If you include much more than Lie Type 7, it becomes very hard to justify any of your censorship.
"A post titled "The Media Very Rarely Completely Invents Facts Out Of a Whole Cloth With Zero Relationship Whatsoever With the Truth" would be a pretty boring and pointless post..."
I think this line is the *observation* more than the argument of the posts. The argument appears to be that it would be hard to arrive at an objective model for censoring "misinformation / disinformation" that does any actual censoring, which seems considerably more interesting.
I'd say if the post has made you turn away in disgust from debating about on whose forehead we get to gleefully slap the incendiary and pejorative label LIAR! -- and we need, instead, to take the much more tedious slog to rhetorical victory of digging carefully into the complete picture of the thinking represented by a statement and slotting it carefuly into one of Scott's 7 categories -- why, this is a quite useful result. Conducive to a more civilized public discourse.
Victorian gentlemen use to resort to pistols at dawn when called a liar to one's face. I doubt very much Victorians lied any less (or more) than we do, so I do not think it had any salutory influence on public honesty -- but having that odd firewall in public discourse did mean they did not sling unforgiveable personal insults at each other in the heat of the debate nearly as often or as casually as we do. Might be worth considering.
It's worth making the argument, because there's a sizable number of people who would disagree, and state that the media actually does invent facts from thin air.
It would be nice if we never had to state the completely obvious, but we don't live in that world.
The Dominion vs Fox case is getting some quotes in depositions that get pretty close to 7. The issue is booking a guest who you believe is advancing a false theory, and then giving her a pretty uncritical slot to make her case.
https://www.vox.com/policy-and-politics/2022/12/22/23523385/sean-hannity-fox-news-defamation-dominion-lawsuit
>Two years later, [Hannity] was asked about Powell’s theory in a seven-hour deposition that was reportedly shared during a court hearing earlier this week as part of the Dominion lawsuit: “I did not believe it for one second,” he said under oath. Powell also walked back her theorizing in 2021, with her lawyers stating “no reasonable person would conclude that the statements [Powell made] were truly statements of fact.”
I think the motive here is that Fox is worried about losing market share to programs like OAN, and needs to put some stories they believe to be false on air in order to keep credibility with their audience, who believes them to be true.
I would count that as lying. At a more polished, fact-checking place like the NYT, I think the behavior would less often be a lie of commission, and more often killing a story that the paper thinks holds up, but they think their readers won't believe. (This is the claim made by Nellie Bowles about her experience at the Times).
In your list, you clearly forgot 4.5: Not reasoning about the truth/falsity of the question at all, but instead answering with some mix of:
- what you believe you are supposed to answer
- what you believe people like you are supposed to believe
- what some higher status than you person said about the question, or someone whose reasoning ability and data sources you respect
- what you expect will make the person asking the question happiest
- what you think the person answering the question expects (i.e. how can you avoid seeming "weird" to them)
Arguably some of this counts as reasoning, but except perhaps for the strategy of "I don't understand this, so I'll trust an expert", none of it is reasoning about the topic at hand.
FWIW, I suspect that media use various of these strategies routinely, particularly if you rephrase them in terms of "selling more papers" and similar.
I believe that in some circles that is what is called "bullshit".
Came here to say exactly this. Bullshitting is an important part of the falsehood ecosystem, and it doesn't really fit on the list - maybe it's close to (4), or maybe it's actually close to (7).
Even more typically than DinoNerds' bullet points, bullshit is often motivated by
- what you think you can say to get what you want.
Frankfurt's On Bullshit ( http://www2.csudh.edu/ccauthen/576f12/frankfurt__harry_-_on_bullshit.pdf ) is a classic, and indeed Frankfurt distinguishes between bullshit and lying. But it's egregiously incomplete to have a conversation about lying and exclude bullshit!
Much of this is how we learn truthful reasoning, though, if we're lucky enough to be in a community where social rewards are tied to learning truthful answers.
When you don't already know how to reason about the topic at hand, and you're still learning, you shoot yourself in the foot if you're not willing to "stoop" to some "fake it till you make it". (I learned this after shooting my own feet an embarrassing amount.) "I'm not sure how to reason about this problem on the problem set, but I bet my teacher is good enough to have given me a relevant example, maybe even dropped a hint about which example to use," is reasoning about the social context, not the problem at hand. It's "faking it", to some extent. But it gives you resources to do the problem at hand in order to develop eventual understanding.
Creating a social context that channels the desire to give a pleasing, "normal" answer into finding an actually-correct answer seems crucial to learning from preschool through university. Deep understanding often takes a bunch of shallow conventions for granted, like the order of the alphabet (the ABC song trains children in the order people "like them" are supposed to believe). And, when you don't yet fully understand a problem, going through the motions of how you've socially inferred "it's supposed to be" solved can help you understand. Answering as you're "supposed to" develops the capacity to reason about the topic.
I learned a lot of things in school that weren't in fact true, particularly in ELHI. I learned even more things that weren't true from religious adults and religious leadership.
Children tend to absorb what those around them claim, without too much critical thinking, and that's a good strategy, both for learning useful information and for avoiding doing things that will get them beaten up or worse. They may or may not eventually go beyond parroting "God is good", "Canadians are better than Americans", "Columbus discovered America", "Eating fat is harmful; sugar is great", etc. etc.
I'm not sure that they actually learn how to go beyond it anywhere in the school system, as commonly experienced - or even whether, as children, they have the cognitive capacity to cope with their teachers being wrong, without experiencing it as betrayal and disillusionment, rather than simply the normal human condition.
"I'm not sure that they actually learn how to go beyond it anywhere in the school system"
Some learn to go beyond it, and those who do usually first have gone through it. That is, attributing the success of these "some" solely to individual superior aptitude, rather than the conventional, sometimes-wrong, education which gave them the scaffolding to exercise their aptitude, takes the scaffolding for granted, and it can't be. Most people who contribute to the sum of human understanding won't be savants from some isolated village with little formal education: there are some of those, but not enough; education, for all its flaws and betrayals, really does help.
It would be strange to say someone who plays piano well doesn't really understand playing piano, or someone fluent in Portuguese, or at constructing correct proofs in geometry and spotting errors in incorrect ones, doesn't understand Portuguese or geometry, respectively. Someone successful at fixing cars understands *something* about them, even if not enough to accurately explain the science of how they work. For much understanding, doing something well enough is adequate evidence that you've gained understanding about the subject itself, not just BSing.
Then there are other subjects, where it really is much harder to tell...
To say that (almost) every genius first went to school, so school must be helping, doesn't really follow when schooling is compulsory. Might as well say that paying taxes (the act itself, not anything the taxes are spent on) is essential for motivating scientific advancement, or something.
Modern schooling does an awful lot of teaching "guess the teacher's password", and regrettably little of teaching "find what is true"
I'm not claiming the education we were forced to attend must be of the best kind – because, look, most geniuses have gone through it, too!
I'm saying even the very talented and intelligent aren't born knowing how to reason about the subject they develop outstanding aptitude in, and typically benefit from social prompts to learn. I'm saying that being in a community where true reasoning and aptitude are socially rewarded matters to developing that aptitude, and that development often includes (even for geniuses) a component of "going through the motions" before really understanding what you're doing and why you're doing it.
I'm also saying that the community can be far from perfect at rewarding true aptitude and be better (even for geniuses) than lacking community – but saying even a flawed thing is better than nothing isn't praising the flaws themselves. Even so, intelligent individuals good at spotting the flaws might underestimate how much their schooling did prepare them to learn true things.
Kids who learn reading, writing, and arithmetic, and that libraries are wonderful places where you can learn stuff for yourself, from their teachers (including parents and private teachers) have been schooled in something truly useful for finding out truth, even if that schooling came with a fair amount of guff. That doesn't make the guff good, or prove that their schooling couldn't easily have been improved on. But the less literate and numerate you are, the harder it is for you to even tell when your teachers are wrong: An education that leaves you literate, numerate – and interested in truthfulness – enough to resent what your teachers got wrong is an education that has partially succeeded – including at teaching you to value truthfulness.
Something like this issue came up back in the bad old days, when good upstanding anglo protestants were sure there were Jesuits around every corner, and each one had an imprimatur from the Pope itself to lie their heretical little heads off.
John Henry Newman actually reminds me of you in that he is sometimes possessed of such moral rectitude that it can be painful to observe him from the outside... pearls before swine is a phrase which comes to mind. Anyway, he discusses a similar issue (lying vs. deception—that is, what exactly constitutes a lie, & the exact limits of the morality of lying) here:
https://www.newmanreader.org/works/apologia/detail8.html
To me this all seems quite simple. You're absolutely right. *Lying* (presenting false propositions) is not *deception* (presenting true propositions in a such a way that, taken at face value, they seem to imply false ones). Whether intentionally or not (whether culpably or not), the media is very deceptive (1-6)—yet very rarely do they go so far as to outright lie (rarely do they get all the way to 7).
The problem is that, as you note, people sloppily equivocate between lying, mere falsehood, deception, and a mere disregard for the truth.
The only way to get out of this, in my opinion, will be to return to moral realism—and ultimately, to natural law. People only began to lie because they did not observe the 9th commandment, which obviously proscribes both lying and deception (which latter thing all ought to agree is wrong, excepting the edge cases Newman discusses above).
To this you might reply that, even without the ten commandments, all men agree that deception is wrong. But in fact, on consequentalist ethical systems, it simply isn't.
I've long been puzzled by the hatred many people have for consequentialist ethics. Your comment is helpful--I agree that a naive, zero-step lookahead consequentialism would allow more deception than would be good for society--according to consequentialist ethics.
Try thinking of consequentialist ethics as ethical epistemology, and virtue or duty ethics as pragmatic implementations of ethics that you can give to computation-limited people in order to implement the results of consequentialist conclusions in the real world.
By ethical epistemology, I mean that consequentialism is the right method of deciding what's good and bad for someone with omniscience and infinite computational power. A consequentialist with lots of computational power would look ahead and see that telling everyone that it's okay to lie if that seems like them to lead to the best outcome, would lead to most people lying a lot, a breakdown of trust, and failure to cooperate or compromise.
Consequentialism gives an approximation to ethical judgement that gets more accurate as you give it more information and computational power. The approximation can be very bad if you use a greedy consequentialism with no lookahead.
A consequentialist who understands human bias would also realize that most people are overconfident in their opinions and computations, and that most people are too dumb or ignorant to act very ethically if they think for themselves. People with different degrees of intelligence, or different kinds of expertise or jobs, should have different ethical heuristics. The whole system may, if conditions are right, evolve into an ecosystem of different ethical systems which together implement an approximation of consequentialism.
(I don't believe anybody has ever successfully *designed* a distributed virtue ethics to implement a consequentialist ethics, nor that any attempt to do so is likely to succeed.)
Attempts to give everybody exactly the same virtue ethics can also work, but it should be realized that this is a pragmatic approximation, not adherence to some eternal code of behavior dictated by a great sky-being. It will break down with people who have to make very high-stakes decisions that need to be correct, like soldiers, doctors, and the rulers of nations. Such people must have the leeway to do things that society at large, with its simpler ethical heuristics, calls evil because they are discouraged from thinking consequentially.
So in building your society, you promote virtues like truthfulness. But these aren't absolute, eternal values; they're heuristics, things easy to grasp which work well most of the time. Only when society is faced with a long-term, complex, high-stakes moral problem do you break out the big guns and apply consequentialist reasoning, slowly, carefully, and preferably collectively, because only the "ultimate consequences" determine whether some action is good or bad.
("The ultimate consequences" may not exist. Total utility, even time-discounted, isn't guaranteed to converge as you look farther and farther ahead. I don't think this is a problem in most situations. You do the best you can with the resources you have, and call it ethical.)
We teach children virtue ethics because it's the foundation. Lying is bad is simple and easy to comprehend, but what about when you are older and ready for more complexity? If you are hiding Jews and the Nazis come to your door, is lying bad? The consequentialist question becomes "when is lying bad?" Phil makes that point well.
That's pride, man. Be humble.
>People only began to lie because they did not observe the 9th commandment
Suppose I think that the Bible is false, what reason do I have then for following the 9th commandment and/or the natural law? The fatal problem with religion in the modern age, that many philisophers have observed, is that it's moral authority is inextricably tied to plausibility of its ontology and factual claims, and those look ever more untenable as our understanding of how the world actually works grows. The God is dead, we killed him, and he's not coming back, so maybe it's time to seriously consider alternatives?
The idea lying is wrong doesn't depend on belief in revelation. I think that most people would agree with it intuitively, but you can argue for it on purely natural grounds.
On a personal note, I resorted to God because I felt that I had exhausted all alternatives. In retrospect it seems like a living example of reductio ad absurdum.
Sure, aversion to lying is arguably both an evolutionary adaptation and could be justified on consequentialist/game-theoretical grounds. In practice people don't derive normative rules from first principles (and are incapable of doing so reliably), but instead use time-tested templates provided by culture, which is where deontology and virtue ethics ultimately come from.
I've heard from a few people that they find religion a more tolerable alternative to despair of nihilism, but I just can't understand it. I can't make myself believe something in spite of overwhelming disconfirming evidence, and I probably wouldn't want to even if I could.
The idea is that their intuitions depend on truths, even if these truths are implicit to most people (and even if most people are far from living up to what they say they believe).
A brief argument to justify that lying is really morally wrong: If we hold to the golden rule, then we ought to reject lying. Because we don't wish to be deceived (or as you said: even if it is possible to wish for that, it shouldn't be), we shouldn't deceive others.
On consequentialism one ought to become religious, provided that it can be shown that doing so would make one happier! But I don't hold to that logic—it's not that I became religious because it was more comforting (I agree that it's evil to try to believe in something untrue), but I turned to God because the alternative premise had lead to contradiction and absurdity.
I should specify that on the religious picture, conversion happens because of grace, not just free will.
I think that we basically agree regarding lying, just frame it in different terms/coming from different aesthetics.
In terms of absurdity, it seems to me that both accepting and denying the existence of a (deist) God leads to absurdities, like there being an uncaused cause or there having passed an infinite amount of time prior to the current moment.
However, even if you accept a deist God, I've never seen a good argument for further subscription to any particular dogma, e.g. a Catholic, Muslim or Buddhist one. They all seem strictly epistemologically inferior in positing a vast edifice of outlandish unsubstantiated claims.
You can't distinguish between someone who tells false claims and doesn't care if it's true or not (4) and someone who tells false claims intentionally (7) you just incentivize people like Trump who always fall back on a type of "insanity" defense. Trump tells false claims all the time but he (and apparently you) defends himself by saying that he didn't know it was false, even though anyone reasonable presented with the same facts as he was would know that it was false.
At that point, it is impossible to identify anyone as lying according to your definition. Even though Trump is a clear liar, who has lead fraudulent enterprises on multiple occasions, one could never criticize him for it. And that is a loss, because his reckless disregard for the truth greatly harmed our societal ability to have any conversations. In addition to the immediate effect of many people believing the false claims that he made (even though you won't allow me to use the word "lie" to describe it).
In theory, it can be very difficult to distinguish 4 and 7 as you say. In practice, people in 7 send each other messages saying "haha the suckers fell for it" in a slack channel called "liar liar pants on fire".
Except that Alex Jones did that and Scott still questioned whether that was actually what he felt.
Additionally, unless everyone (including Scott) agrees that Trump is in category 7, which I don't think everyone agrees on, then he is taking advantage of the loophole that Scott's "useful idiot" stance creates.
All that stuff about how paranoid your readers are speaks to me, I’m also often shocked by the paranoia of my comments section, often relating to the exact same issues and in the exact same ways. Do comments sections just draw paranoid people or are we naive dopes?
I am one of the people who was/is very hard on the media.
But my guess would simply be that people like you and Scott who are trying to create positive information transfer have a greater appreciation for how difficult it is than people like myself on the outside who can just get riled up about all the failures and make angry japes to vent our disappointment.
I do think the core desire for strong condemning language around this issue is the vast gulf between how much imprimatur as arbiters of truth the journalists/outlets/acolytes of the MSM allocate to themselves, and how much they really deserve.
The focus isn't on the NYT/NPR as a bunch of flawed humans doing the best they can. And it isn't on that because the NYT/NPR presents and postures itself as THE PLACE the go to get information about what is really going on in the world. And it manifestly both fails at this regularly, and often isn't even trying to do this.
And the frustration is so large people just want to bring out the largest rhetorical guns to blast them they possibly can justify. Which of course is ironic. Humans are dumb, especially me.
I have a hearty dislike of the media also, but it has little to do with how much they lie, bullshit, or deceive, because I don't think they do so to any significantly greater degree than good ol' boy Walter Cronkite or fast-talking Ed Murrow back in the day.
What annoys me is the saccharine unction that drenches everything, like they're all college sophomore Sir Galahads rescuing damsels from Sir Mordred all day long, instead of just hacks pulling down a salary by shipping out a dog's breakfast of what seems to be roughly true and approximately interesting at fast-food speeds.
If they re-adopted the world-weary cynical hard-drinking edge of the Raymond Chandler reporter ("Ah...I'm just a hack trying to make a livin'. Didja know about this bum politician's shenanigans? Whaddya think...?") I'd probably be fine with them, even if they were just as often wrong, and pushing some establishment narrative or other just as hard.
It's like going to the hardware store for a screw to fix the microwave, and the sales help says "A...screw? Nonsense, you poor ignorant peasant, what you want is a fastener, and they come in a wide variety of types and sizes, which passeth your understanding I am sure, but I majored in Truth and Wisdom, so if you stick with me for the next hour I'll begin with basic metallurgy and Roman techniques of construction, and I daresay we'll uncover the real root of your personal problems -- tell me, have you ever kicked the dog because the microwave won't start? Tsk! -- and how you can get therapy to learn to just accept the microwave as it chooses to be in the moment."
The urge to clop them upside the head is strong. Just give me the fucking screw. Just tell me whether Backhmut was overrrun, the price of gas went up or down, and whether Amazon is hiring again, then shut up and play some Cheap Trick or Mozart. If I want to be scourged I'll go to church.
This is both an astute observation, and very entertainingly written :)
Thanks, that's a kind thing to say, and I appreciate your taking the time to do so.
Your account of Raymond Chandler amused me, because that's pretty much the impression I got from Trump. I could easily imagine him saying "I'm just a guy trying to MAGA. Here's this problem happening at the border. Whaddya think?". I (not a Trump supporter, but just opposed largely on account of his lack of experience) naturally felt I couldn't believe anything he said unless I'd checked it some other way. However, I also felt like that was *obvious*. Like a guy selling knockoffs at a street corner. I could see it coming.
Counterintuitively, this worked in Trump's favor. Why? Because he was basically competing against the clean-cut guy in the department store... who was *also* selling knockoffs.
Who writes the most memorable and negative reviews on Yelp? People of rigid patterns of thought who get hung up on one small negative and just go nuclear over it.
Someone who reads a long nuanced post and thinks "eh...there's a couple points here where I think this wavers from Gospel truth...some footnotes/caveats/elaborations/softening adjectives that I would definitely add..." is not going to be sufficiently driven to wrestle the Substack comment system into submission in order to make a long nuanced comment of weak partial disagreement -- and even if he does, it won't stand out as much to you, the author.
I think it's sort of the parallel to the Twiiterverse being far more shrill, dogmatic, and Manichean than actual people in their actual lives.
Comment sections absolutely attract the paranoid, as well as a few other categories of unusual people. (I’m not sure which categories I fall into, but I recall that traditional comment sections represent a fraction of a percent of readers, and even Substack comment sections represent only about 10-20% of readers, who are already self-selected in all sorts of weird ways.)
Yeah reminds me of when I was a volunteer developer on an extremely popular open source game, and all the developers and friends of the developers did mostly MP, and all the "regular" players on the forums etc. did at least some MP.
So imagine our surprise when the data showed that as far as total hours played SP was ~97% and MP ~3%.
I saw an example from World of Tanks, where apparently if you post on their official forums you have to log into your account and it shows some basic stats about you. Some guy was complaining a lot about the game or some update. He had played about 100 matches and had over 9,000 posts on the forums.
for what it's worth, Kenny, I've seen you leave a lot of comments, and they are often good and never (as I recall) paranoid.
My guess (and I do not claim a lot of confidence here): The combination of the internet and the financial collapse of a lot of media has left us with tons of very visible examples of lies and deception and general dysfunction by all kinds of high-prestige, formerly-trusted institutions. The result is that everyone starts out kinda defaulting to skepticism or wondering if they're being lied to all the time.
I'd compare this with how the replication crisis has affected my reaction to reading about new results in the social sciences--like I should start out with high skepticism and a cynical "this was probably p-hacked or something" assumption, because I know how much BS was published in the social sciences in the past, and it's hard to believe that there isn't still plenty being published now.
Similarly, you can now access rebuttals to assertions in the major media, which was very difficult to do pre-internet. Some random guy with the time and inclination can point out the falsehoods or misleading information in a major news network's reporting, which I think has resulted in a lot of us noticing how frequently the news does this. It may have been similarly common in the 1970s (to pick a random pre-internet date), but very rarely known.
I think Dan Rather's scandal was the point of change, where it became obvious that the public could actually affect things by fact checking the major media.
Speaking of which, I haven't seen Rather's name in this discussion. While I'm pretty sure he didn't technically lie in the initial report, his continuing to maintain his belief in the documents gets to the point where it's impossible to distinguish from deliberate ignorance.
OTOH, but 60 minutes did that misleading edit of their interview and was caught on it in (I think) the 80s, and seemed to keep their prestigious trusted place in American society. Maybe the existence of alternatives matters here?
60 Minutes did have some prior scandals, but to my knowledge they were opposed by companies with the ability to hire lawyers to challenge them in civil court, and it took time for the courts to work. Rather was exposed, not by the target of the segment lawyering up and taking the time to go through the court system, but by a blogger, and the false material was exposed as false in a matter of days.
And 60 Minutes (and even Rather himself) still haven't completely lost their trust.
Hah! If you and Scott are naive dopes, honestly, I am, too. ...I also rarely comment, in general, so I suppose I'm not helping the statistic. I think Carl Pham's comment here is spot-on - people with stronger opinions are just more likely to post, and paranoia breeds strong opinions.
By the time you get to "Reasoning badly, because you are biased, and on some more-or-less subconscious level not even trying to reason well.", you're full blown trying to deceive people in ways which you would not if you felt accountable for what you were saying. This is the *central thing* that accusations of lying are for, and so the term "Lying!" absolutely fits when someone makes real sure to not reason carefully because they anticipate that reasoning carefully would cause them to know that the things they are saying aren't true. After all, there's conservation of expected evidence, and if they know to avoid reasoning carefully they damn well know enough to conclude that what they're saying isn't likely to be true.
Scott's category there is not people doing it about a specific issue, but that that sloppiness is part of their character. These people are not "anticipating that reasoning carefully would cause them to know that the things they are saying aren't true", they just don't reason carefully as a matter of course, perhaps because they consider feelings to be more important.
I'd say most people fall under that category, and it doesn't seem useful to say most people are liars.
No, they definitely are aware that careful reasoning would lead them to conclude that they're speaking falsehoods.
The test is to point them at the flaws in their thinking and see what they do. If they say "Oh shit, thanks for pointing it out!" (i.e. show genuine surprise), then they weren't aware. If they don't even falter in their BS and keep on as if you hadn't said anything to challenge them, then it's possible that they genuinely don't care enough to even anticipate and you get no data (though they likely do anyway). Generally though, these people will make active attempts to avoid looking at the thing you're pointing at, and you can only do that if you already know that looking there will do something you don't want.
Popularity of lying cannot make it not a lie. You just have to be careful with how you apply your limited attempts to shame people for things, since lying is too ubiquitous to just knee jerk it away by using the word "liar" as a shaming tactic. Christianity knows this distinction well. "Everyone is a sinner", "Love the sinner, hate the sin", etc.
I think most journalists are consistently ignorant about pretty much every technical subject. This is part of the point of "Gell-Mann Amnesia" - every time I read an article in a field where I have technical knowledge, it's either wrong or so garbled that it's not even wrong.
Garrett complained about the media's problems with reporting on firearms, but the problems there are not particularly distinctive. I've just come to mostly accept that articles in the general press will make egregious errors on everything other than easily-available, easily-understood public information. This isn't primarily because of bias, and it's certainly not lying. It's just people being dumb.
I would say the reporting related to guns is particularly bad because the particular tribe journalists tend to come from is substantially less likely to contain firearms enthusiasts. So there are a lot of newsrooms where nobody has a good bullshit detector for gun related information, because they don’t live in a social circle where a working knowledge of guns is in the water supply and where your buddies will laugh at you if you do something like call a folding stock “the part that goes up”.
Journalists are rarely “experts” on technical subjects, but there are definitely areas where they are more or less ignorant. You don’t need to be an expert to have sufficient working or casual knowledge to be able to sniff out likely bullshit.
This is one area where I think increasing diversity of backgrounds could be legitimately valuable.
++
Also, its not hard to find a competent gun expert, the Times, Post, etc just actively avoid it. This is not like employing a full time Biomedical Engineering expert for the 3 annual stories you are going to run about new developments in artificial organs. You are running hundreds of stories annually on the subject of guns, and you don't have anyone on staff who knows the difference between semi-automatic and automatic firing.
Right. There's a point where intentional ignorance becomes no different from knowing falsehood, at least morally. The standard for 7 should be "knows or should reasonably know that the statement is false". Reckless disregard for the truth is morally indistinguishable from a knowing lie IMO. It certainly is for libel standards (the former is what's know as "actual malice", in one of those "legal terms that don't mean what you think they mean" instances.)
Sometimes I do think it’s “actively avoiding it”. But other times I think it’s not quite that. It’s more like “so completely ignorant of the subject and unexposed to anyone who isn’t that you don’t even realize that a distinction might exist, or if it did, that anyone would care”.
Now I do agree that at some point if you’re going to write about a subject a lot you have a responsibility to cast a wide net and at least become familiar with a range of experts and opinions. So I don’t want to imply that I’m excusing ignorance or laziness.
But if you and literally everyone you interact with on a frequent basis all share the same blind spot, how do you know the right question to ask, who you should ask for help finding the right questions, or even whether there are any questions worth asking?
Because people have repeatedly told them that they're wrong. And experts have schooled them. And they've just closed their ears and said "nah nah nah, I can't hear you!"
Have “people told them” or have “people they don’t know yelled at/about them on Facebook or in media they don’t consume”?
I think we should be careful to avoid ascribing to malice what can be explained by bubble effects. Again, not an excuse - but maybe part of the reason. And I’m making a point that’s more expansive than just the gun issue explicitly.
Editorials in their own papers. Reports in academic journals. Court documents. All the things they're *supposed* to research as part of their jobs. Heck, even experts *they themselves talked to*. Many of whom have explicitly said that they told them differently, but they ignored the experts or misquoted them to say just the opposite and then quoted 'anonymous' sources which said utter obvious garbage.
On the gun issue in specific (and in many others in general), the kinds of crap they spout is of the order of "1 + 1 = banana". Not even *plausibly* wrong. And requiring willful blindness to reality to say without malicious lies.
Edit: And at some point, I don't believe there's a difference between actual malice (the literal meaning) and a suitably reckless disregard for truth or falsity. At least not in practice, as far as anyone else is concerned.
But this still explains too little. Liz Cheney is now kind of a NYT darling for being anti-Trump and probably has slipped pretty far into being purely anti-Republican. But I'm pretty sure she's been hunting and knows the difference between an AR-15 and a M-16. There are lots of people like her. Now, most aren't Vanderbilt heirs who can afford 3 years of unpaid internships followed by 10 years earning $40k/year in New York or DC, but that is their own weird problem.
Not just firearms. I am from Eastern Europe and have closely watched coverage about war in Ukraine. Journalists are extremely bad if talking about military matters, military weapons, weapon systems, etc. Making mistakes, not understading the subject and creating bizarre interpretations.
Very good point.
Historically, I think there was also a culture in progressive spaces where firearms knowledge was considered morally bad. Lots of people seemed fairly proud of their complete lack of knowledge, which confused me greatly, and I didn't witness that same effect with other technical subjects.
In Sunday school I was taught about lying being divided into "sins of commission" (category 7) and "sins of omission" (category 5 & 6). The point being that leaving out key points to deceive someone is still lying, just another kind.
"Commission" and "omission" may be a useful start on your names for every category.
Another post called this distinction lying and deception, respectively. I think that's a good way to look at it.
Re. this comment by Tytonidaen:
"I think a more likely explanation is that many people are choosing to attribute deaths to the vaccine that are not actually from the vaccine. For example, let's say Person A gets vaccinated and dies shortly after of some completely unrelated cause. And let's say Person B, the loved one being polled, has priors about vaccines or the medical establishment or whatever that cause them to be convinced it was actually the vaccine that killed Person A.
"In hypothetical reality, Person A lived a rather unhealthy lifestyle, had lots of risk factors for a heart attack, and would have died from a heart attack, regardless of whether they'd gotten the vaccine. Then, when Person A does, indeed, die of a heart attack, and by sheer coincidence had recently gotten vaccinated, Person B blames the COVID vaccine when polled, but it wasn't really the vaccine that killed their loved one."
This seems fair, since this is how Covid deaths were counted. Once enough tests were available, anyone who died, was tested for covid; if they tested positive, they were reported as a "covid-related death". News outlets reported covid-related deaths as Covid deaths (which was, BTW, a very common media lie).
For one example, an employee of my brother-in-law was supposed to get a liver transplant. He was tested for Covid before being admitted, tested positive, and so was taken off the list of eligible liver recipients, and died soon after of liver failure. This was counted as a Covid-related death. Which it was, in a sense.
What's confusing is the fact that, in the face of unreality, people claiming to propagate The Truth actually propagate untruths, hence fascist 'anti-fascists', racist 'anti-racists', the proposal for a national Office of Disinformation -- presumably to identify which disinformation the government approves of.
Regressives declare that norms are obsolete and oppressive, then strain credulity to 'normalize' the bizarre. It all makes perfect sense if one accepts chaos. But if you're building your regressive paradise around chaos, you can't turn about and fault others for not recognizing your new, improved 'norms'. You just told us norms are obsolete. Oh! Wait, I get it. Our norms are obsolete. Yours, on the other hand, are golden. It's a pretty juvenile style of rebellion.
Brilliant, best piece I've read in awhile that exposes the sinews of the fundamental disagreements that hover over and stalk almost every expression on social media. And REALLY depressing. Thanks anyway.
This is a really straightforward case of "Taboo Your Words". It literally does not matter whether the media is "lying" or not. Scott's original claim was that the media rarely does number 7, and that this is important /specifically because/ that means it is very, very difficult to gain consensus that a given media outlet has done something worthy of censorship. Scott's point is that "reasonable people can disagree" about almost anything, which is proven in how many reasonable people are disagreeing about every one of these issues in the comments.
Given this wide swath of reasonable disagreement, what kinds of statements can be responsibly prohibited or punished? A vanishingly small number. Almost none.
That is the point. That is Scott's claim. That is what you have to engage with. To say "No, I think this behavior should be called lying" is to change the subject. To say "Actually the New York Times is better than/just as bad as/worse than" InfoWars is to change the subject. Your task is to find an article so egregiously false that almost no one in the world would take issue with you trying to get it banned. Until you can do that, Scott's point stands as almost trivially true and impossible to deny.
Yes. Well-said.
Yes well said and I thought this was also the whole point of Scott raising this topic, so I was a little surprised that he didn't underline that in this post and seemed focused himself on parsing words. I felt like if I were Scott I'd be shouting "you're not hearing me!" but maybe he feels that way a lot and so this discussion didn't feel more egregiously that way to him.
Censoring is very difficult to start with for the stated reason, and even more difficult when there is uncertainty involved about an alleged future outcome that is unprovable. Many of the most emotionally entangled subjects up for censorship fall under this category.
I do think this argument goes the other way and we should try to build a consensus for censorship that goes beyond category 7. The hard part is figuring out how far to go.
Nice.
>Given this wide swath of reasonable disagreement, what kinds of statements can be responsibly prohibited or punished? A vanishingly small number. Almost none.
Which is absurd, because this media furiously demands that people not literally lying need to be censored.
But most calls for censorship are not about truthiness, but specifically and carefully couched in terms of "harm". In other words, they would not care too much about deceptions that they think reduce overall harm, like claiming that vaccines prevent transmission which incentivized people to get vaccinated, just like they *would* care about scientific truths that they think will cause harm, like acknowledging differences between genders/races/whatever-group.
I can find dozens of articles that are egregiously false. Scott will refuse to put them in category 7 because he will blow smoke about how they may have been honestly mistaken. Moreover, even if they fell in category 7, many would disagree that they should be censored. So this categorization doesn't actually help in the censorship conversation.
How about separating the factual claim of whether people actually lie from the conversation about what appropriate censorship is, which may or may not have anything to do with lying at all!
I agree this is the argument Scott is trying to have, but I think a lot of the comments here are not as orthogonal to Scotts claim as you seem to think. Specifically that because a media outlet rarely does 7, it follows that it should be hard to censor 'misinformation'.
However this is totally contingent on accepting the idea that only 7 constitutes real misinformation. Arguing about what is or is not a lie, is directly arguing with a core assumption of Scotts arguments. If somebody holds that anything beyond 2 is misinformation and should be subject to censorship, that is a consistent principled position that they can maintain. If Scott can not defend the idea that only 7 is misinformation that can be fairly censored then his 'true' argument is necessarily unconvincing.
It is not enough for one person to believe that anything above 2 is misinformation to collapse Scott's argument: it would require that everyone else agrees on that that is the correct tier, and to agree on to what tier any given claim belongs.
Scott is focusing on tier 7 because people often say, when defending censoring misinformation, that outright fabrications of that kind are a) uncontroversially bad, b) easy to identify, and c) common enough to be a problem. Scott was trying to disprove C, but ended up disproving B as well.
The goal of the censorship scheme is to appear as fair and unbiased as possible, and Scott is arguing that claims from tiers 3-6 are so fraught with emotional and political valence (or sometimes just genuine complexity or uncertainty) that any misinformation filter that applies to them will necessarily appear unfair and biased to large groups of people.
The claim you have to defend is not that "anything beyond 2 is misinformation", consistent and principled as that position may be, but rather that /everyone agrees/ that anything beyond 2 is misinformation, and /everyone agrees/ which articles fall into the category and which do not. Otherwise the misinformation filter is going to start removing stories you believe are true.
I was responding specifically to your point that arguing about what is or is not a lie is orthogonal to Scotts position, 'changing the subject', when in fact I think it is directly relevant and a central part of the argument. Scott must defend his focus on 7 and if he fails to do so his argument fails to be convincing. There are tons of largely unsupported and unargued assumption about censorship and what the public believes, how they do/will behave. You think the comment section shows people being unable to understand Scotts position, I think it shows that Scott's model of the problem and actors is wrong, such that drawing conclusions from his model is likely to produce error.
Honestly, I don't think retreating to 7 produces the results that Scott seems to think it would, I think a very large cohort of the relevant population specifically believes that some 'facts' are more false than true, and for example, reporting 'factual' race crime statistics is actually reporting something that is false, because it fails to grapple with the broader implications and context of the issue. 7 is IMO a false category when it comes to how groups of people view reality and truth and far from something that society can/is coordinated around.
These kinds of fundamental disagreements with Scott's model of reality are what the comment section is reflecting.
I don't believe the comment section shows that people are unable to understand his position, I believe it proves his position. If there is any substantive disagreement about what constitutes a lie---which this comments section undeniably demonstrates---a censorship standard based on "lying" is worthless.
I think there is a really fundamental disconnect here, I saw Scott as making the point, to quote you,
"Scott's original claim was that the media rarely does number 7, and that this is important /specifically because/ that means it is very, very difficult to gain consensus that a given media outlet has done something worthy of censorship."
which I think is wrong as a comment about the world that we live in. There is significant consensus on the kinds of things that people want censored that do not meet the specific definition of 7. In fact I think there is a pretty strong consensus here in the comments that 7 specifically is not a good standard for censorship/labeling something misinformation. We currently live in a world where such standards are being practices everyday, and presumably the people who practice them find them to be valuable and are able to act because they have the support of their local consensus.
I think Scott is making a point that works assuming you accept his priors, and a lot of people don't, and these last two posts are basically Scott trying to defend his priors. I think that conversation is directly relevant to Scott's point, and you seemed to disagree.
The only consensus I have seen in the comments is the number of people who think Scott is wrong. They don't agree on what Scott is wrong about, or with each other on what is right. I think that's a very big point in favor of Scott's argument.
> We currently live in a world where such standards are being practices everyday, and presumably the people who practice them find them to be valuable and are able to act because they have the support of their local consensus.
If this were true, we wouldn't even be having this conversation. We would already be living in a misinformation-free world.
In the second article, Scott responded to several examples of things commenters claimed were obvious misinformation worthy of being censored. Do you endorse the claim that all of these were, in fact, valid examples of misinformation? Do you believe everyone but Scott endorses that claim? If not, your proposed consensus does not exist.
The American legal system has built out a robust framework for classifying and evaluating different levels of “being misleading” (ranging from intentional, unambiguous, damaging misrepresentation to unintentional, slight, inconsequential carelessness with respect to accuracy). Securities fraud case law is one example of a large body of reference material. Just noting that in case helpful to anybody doing a deep dive into this sort of discussion, no need to reinvent the wheel.
Other legal systems have done so too, I’m sure, but I’m unfamiliar.
Absolutely! In particular Infowars was just raked over the coals in that exact legal system for that exact reason!
How about "Reasoning well, and just having an interest in pretty (ostensibly, to most of us!) interesting things". The laptop, sure, but we've run over that ground enough. Compare it with Amy Carter, whose habit of reading during state dinners was grounds for a dozen stories. Or the Bush twins, whose famous underage margaritas will follow them to the grave.
The incuriosity of the media, even when it seems like they could "sell papers", has amazed me most. (Crime no longer automatically interests them - when they must know that their readers, or at least their female readers, will never tire of the agony column.) Reporters used to like to dig up ... weird stuff. Maybe that was a job best done away from a keyboard. Maybe it was not very edifying. Are we edifying now?
Because it was mentioned, I thought of the Zimmermann/Trayvon story. Now, I have no opinion one way or another about whether the "white frat boy" Zimmermann should or should not have been convicted. Actually, I don't even know what happened with that. But I one time listened to Loury and McWhorter on the former's podcast; and they presented some really strange info about the prosecution. Like, strong evidence that the star witness was not even who the prosecution claimed she was. The two of them had already smoked this, based on their shared and immediate suspicion about the incongruity of a "cute" boy like Trayvon, and the girl put forth as his girlfriend, whose testimony involved what he said to her on the phone that night, as best I recall.
This seemed pretty crazy!
Well, the world moved on from this case, so maybe it's not so strange that none of these discrepancies pointing to a rather shady prosecution, made any news. Still, I rather expected that when two well-respected academics made these claims and inferences on their podcast, that some news organization or other would take an interest.
As an attorney I can’t lie to a court. But I absolutely CAN selectively choose which facts to highlight, in service of the position I’m advocating. I would have been ethically allowed, had someone hired me to do so, to advance InfoWars birth certificate theory in court (standing problems aside). There are a few things considered lies of omission, but they’re much more black/white than this, e.g. if a case I’m citing was later overturned and I know that.
So the bar association agrees with Scott.
It all goes back to people’s desire that newspapers, at least the ones on the righteous enlightened side of issues where their readers surely reside, be thought of as motivated by a purer motive than lawyers. But lawyers and journalists are the same on this: we’re all just saying the best version of what our customers want us to say. Scott’s definition of lie is fine for both.
The customer of a lawyer and of a newspaper ought to be very different people.
Here's another corner case: What about when someone does something 2-5, then is made aware of their error and chooses not to correct their statement and issue a retraction, because doing so would make them look bad or go against their desired narrative. Is it reasonable to say that their statement is now, retroactively, lying?
The original statement is not a lie, just a bad take. If you ask them directly and they repeat it, the follow-up is a lie.. But not going out of your way to bring the topic up again doesn't make it a lie,
So then Infowars lied, because they were corrected on every issue and continued to disseminate false information. But Scott insists that they might have continued to believe the false claims even after being corrected. There's no possible way to move into category 7 according to Scott!
Define 'corrected'. Just people saying 'you're wrong'? That's not enough.
There, now you've been made aware of your error. Further statements of your original point will thus be counted as lies.
If you would like to do a better job than the media does when they "don't" "lie", you can pick up some tips at Notes on Honesty (https://www.lesswrong.com/posts/9iMMNtz3nNJ8idduF/notes-on-honesty) and Notes on Straightforwardness, Frankness, Sincerity, Earnestness, Candor, and Parrhêsia (https://www.lesswrong.com/posts/haikNyAWze9SdBpb6/notes-on-sincerity-and-such).
Being misleading is not the same thing as lying. But it is bad. Lawyers, for example, are not supposed to mislead the court. That includes more than just not lying to the court.
People being misleading little sneaks is something that's perfectly valid to call out. They're just not technically liars.
I understand your point that there is legitimately no way to draw a bright golden line between awful Infowars style reporting and honest, excellent reporting that uses true facts to erroneously report a false conclusion.
There's also no bright golden line between negligence and failed due diligence, or reckless disregard for life and acceptable risk, or good actions and bad actions.
Using similar criteria we'd struggle to ever punish wrongdoing in any field. I agree that speech demands considerable more care than other things and would warn folks off of government control of speech for this reason. But to suggest that we avoid any line we can't draw in gold marker is to suggest no lines at all.
This is the fundamental flaw of Scott's argument, and I wish he'd grappled with it, rather than using needlessly inflammatory language and then getting caught up in arguments about it.
With regard to the word "Uncritically":
1. You may not be able to fact check a scientist yourself, but you can talk to other experts in the field. As Beowulf notes, an actual epidemiologist could likely have told the LA times that ocean aerosols weren't an area of concern. But they didn't ask, either out of laziness or because they were concerned their sensational headline would be endangered.
2. Usually when I criticize someone in media for acting uncritically, it's because they're citing a source with a known poor track record. Repeating something a white house press secretary says without comment is flatly irresponsible. Same with anything involving "anonymous officials within the intelligence community".
Much worse that not asking out of laziness is not asking out of ignorance. Like "this is a scientist, therefore we have no need to speak to another scientist".
A lot of journalists will speak to an expert on any subject and then that's sufficient: they have spoken to an expert, the expert has given them an expert opinion, there is no need to speak to a second expert.
This has been one of the fundamental problems with all science coverage: they don't know that they are doing it wrong, because "asking an expert" is doing it right by journalistic standards.
Merely copying the opinions of impressive seeming people is frequent in the rationalsphere, as well
>they did not show a consistent tendencies to make things up or say outrageous things (except for one who listed their religion as “Satanist”).
Hey, at least one of the branches of Satanism seems like a pretty decent group, and it's not like they're worshipping the literal theistic Satan.
One really good example of 6 is a technique used by certain (tabloid) newspapers in England. They will report that "a source close to Downing Street" said something or other. Normally, that sentence would refer to someone who works in Number Ten, ie a politician or a senior political staffer.
But sometimes, they just guess what is going on (a smart journalist with good background sources usually has a pretty good idea even if no-one will actually tell them), so they send a junior staffer to go and stand in the street outside the building, call into the office on their mobile phone, and repeat whatever quote they decided they wanted to print. The person was close to Downing Street - he was standing right outside! - so it isn't a literally false fact.
Most of the time, the good political journalist has correctly assessed the situation and the result is that everyone is convinced that they have incredibly good sources and no-one can work out who those sources are. Occasionally, they guess wrong and they would look like idiots ... except everyone just thinks they were misled by a source who wasn't in the loop on that occasion.
I think this highlights a problem with the categorisation: if your distortion of the facts is in aid of a substantively correct conclusion, then that is a much less problematic situation than distorting the facts in aid of a substantively incorrect conclusion. That's not particularly correlated with the question of whether you are lying though.
I think there's lots of "mood affiliation" here: if someone is saying something they believe and which the reader doesn't believe, then they will think that they are distorting the facts to prove something that is wrong, and that's much more like lying than distorting the facts to prove something that is correct. But the definition of "wrong" and "correct" is, all too often, "what the reader agrees with".
Your categories of things-which-might-be-lying is missing "reasoning poorly because you don't care about the truth".
It sounds like you want number 4 to encompass that, but it really is a different thing than being biased. And it's different from number 7 unless you include "A implies B" implications as "facts" which can be made up.
And I think we should call this thing lying, even if we wouldn't call number 4 lying, and that this is what InfoWars does at least some of the time when it makes physical-world claims which are false.
---
But regardless of whether you call this lying, I think that it is worth at least noting that InfoWars is doing a _different thing_ than the rest of the media here. The NYT mostly manages to hedge its physical-world claims enough to avoid them being outright false. InfoWars does not. So defending InfoWars against a charge of lying requires us to give them the benefit of the doubt in their motivations in saying false things, whereas we can defend the NYT on the basis that they didn't say false things in the first place. This really is a difference of _kind_, not merely of degree.
---
Kind of a sidebar, but - I'm fine with giving people the benefit of the doubt, but that means we don't say "they are lying", not that we say "they're not lying". To say "when [establishment papers] do err, it’s by committing a more venial version of the same sin Infowars commits", that requires us to actively claim that InfoWars is not lying, which is different from just giving them the benefit of the doubt. The most we can say is that, lacking access to their motivations, they _might_ be committing the same sins.
Your sidebar seems extremely important to me. That's definitely a relevant difference.
On the Jeremy Goldberg Inflation comment: In high school I took AP Economics. The teach was a history teacher assigned to the class despite no economics background, but he'd been teaching it for many years, so I'd expect him to at least be able to parrot the textbook. In any case, he told us he could not understand what the difference was between deflation and disinflation. Note that I'm not saying he glossed over it or explained it badly, I'm saying he literally told the class he didn't see how there was a difference between these two things. I only talked to a few of the other students about this, and most of them didn't think his confusion was problematic, even though they were taking calculus at the time and so were definitely taught about the difference between functions and their first and second derivatives. So personally I find it *very* easy to believe that the entire newsroom at the Washington Post didn't notice any problem with what was being published.
See also any news article that talks about "salary cuts" when they mean "people are getting smaller raises this year than they did last year".
“<Party I don’t like> SLASHES budget for <thing I do like>!” where “slashes” means “reduces proposed increase”.
If the proposed increase is at the rate of inflation, it is a slash in real terms, just not in nominal terms.
Sure, but I rarely see that distinction carefully and consistently made.
Does “disinflation” actually mean a decrease in inflation rates while they remain positive? Or does it actually mean a (brief) period of decrease in prices that may not be long enough to drive year-over-year rates negative?
Disinflation is a negative second derivative of the price level, typically paired with a positive first derivative; deflation is a negative first derivative.
Both derivatives being negative would probably be described as "accelerating deflation" or similar.
Honestly, I think keeping track in your head of something like "first and second derivative of price level" is hard for most people. Hell, otherwise reasonably smart people get confused about comparing stocks and flows (approximately x and dx) all the damn time.
Were there any takes on your core conclusion (that stopping "lying" is an invalid argument for censorship enthusiasts) worth highlighting? I found myself frustrated that the comments I read were all focused on the nitpicky definition that was clearly highlighted as nitpicky, and not on what you were actually arguing *for*. If your nitpicky argument is invalid, does this mean that censorship standards can be invoked as long as journalists are avoiding points 6&7? All 7 points? What regulations would be put in place to address these?
> I found myself frustrated that the comments I read were all focused on the nitpicky definition that was clearly highlighted as nitpicky, and not on what you were actually arguing *for*.
On a couple of occasions I've tried to start making a point by establishing some common background material.
It's never worked. If you start with background material, you can't reach your actual point. You have to start there.
Okay, so now that we've discussed the problem for three posts, is it finally time to propose solutions? Because I'd *really* like to be able to trust at least some sources of media to not outright mislead me to false conclusions, and I'd like the ones doing that to be disincentivized from continuing.
The only way I think this can work in general is if we have media sources who have:
a. An incentive to get things right and convey them to you honestly.
b. A way to find out when they're doing that well/poorly.
c. A reason to care when (b) says they're not doing a good job.
The classic version of this is Vegas oddsmakers. They have a genuine incentive to get the odds right, they'll learn quickly if they're doing badly at it, and they won't keep their job if they keep making bad probability estimates.
My sense is that journalists are mostly not in the learning-reality-and-conveying-it-honestly business. That is, the best journalists care about those things because of an internal ethic or integrity (thus, the NYT tending to keep the contrary-to-narrative facts in the story but de-emphasize them), but I think the actual thing they're being paid for is writing stuff that brings in revenue--either from ads or from subscriptions. From the perspective of the NYT's management, a sequence of factually-accurate and honestly-conveyed stories that piss off a bunch of subscribers and advertisers is much worse than a sequence of factually-shaky and rather deceptive stories that make the subscribers and advertisers happy.
That is, I think a lot of our collective frustration with journalism is that we're mad it's not doing some other job we wish it would do. I would like NPR, the Wall Street Journal, the Washington Post, the New York Times, the Economist, etc., to make a genuine effort to understand what they are reporting and then convey it in a clear and accurate way, including links to raw data, unedited interviews, etc., so I can follow up. But I'm in a very small minority, even for those generally high-end news sources. I think a careful and accurate series of reports on what urban crime looks like might actually be financially bad for NPR, because it would upset the donors. They're probably better off running ideologically comfortable stories about historical racism and how it echoes through to today. (And indeed, I'm sure there are places where historical racism hurts blacks now, but I'm like 99% sure that the impact of historical racism on current-day blacks is a tiny fraction of the impact of crime on blacks.)
I think some people keep their audience by being careful with facts, because that's their audience. Nate Silver would probably lose his audience by becoming another writer of fuzzy-headed ideologically-reassuring stories. But I think they are the exceptions.
Think about it in terms of Silver's home turf, sports reporting. Some people want an accurate assessment of how likely the home team is to end up winning the playoffs, but most people probably really want to hear a nice happy story about the strengths, virtues, and heartwarming story of the home team, how they deserve to win the playoffs and will surely win because of their many great qualities. Telling the fans who want that kind of coverage that the odds are 4:1 against the home team is bad for business...unless you've built up an audience that actually wants the hard-headed stuff instead of the heartwarming stuff.
A lot of people go into journalism not to debunk bad ideas but to use their verbal skills to promote what seem to them at first glance, working under time pressure, as good ideas. Malcolm Gladwell is a striking example of this.
I wrote a number of critiques of Gladwell when his reputation was sky-high in 2005-2010, and no doubt played a role in the decline of his reputation.
But, my conclusion was that Malcolm's faults as an explicator of the real world were due not to anything contemptible about him, but instead due to likable aspects of his personality: he really admired college professors and their pet theories, so when he heard a clever new idea from one of them, rather than wanting to see if he could debunk it (as I tend to do), his natural inclination was to get enthusiastic thinking about how he could use his writing ability to persuasively frame this new idea so that more people would understand it and appreciate it.
Gladwell didn't have strong critical thinking skills or much of an urge to debunk, but then most people don't either.
I think that, in addition to media sources have your a, b and c, we will also need *readers* who have:
a. The ability to quickly visualize what claim(s) an article is making that the author believes are most important.
b. The ability to view what material in that article supports any given claim it's making.
c. The ability to check that material against other sources.
This is basically a first order logic theorem prover. And it's not necessarily a pie-in-the-sky contraption. (I spent years working on the precursor to such a thing.)
Part of the idea here is that different readers know different things, and such a technology could starkly illuminate the difference between what this new article is claiming, and every other article that reader has read. So, a "what's new" section that will report things like "the vaccine offered prior to December 8 is believed by the CDC to *not* be effective against variant 92.33.1411D" or whatever. It would tease out every referent, and even paraphrase the summary multiple ways. And the reader could even ask it questions like "does this mean I have to get another shot?" and it will answer either "yes, given these conditions" or even ask the reader reasonably simple questions, playing the role of an expert acting in that reader's interests.
Unlike GPT, such an "auto-reader" would be employing discrete symbolic reasoning logic, not a neural net or some Markov chain method. That discrete approach would have many other applications as well.
> you can fill your articles with sentences about how you’re not claiming X, and people will still find ways to accuse you of lying because you said X.
I think this is because if an article spends 5,000 words seeming to say one thing, and 50 words saying the opposite, people figure the 5,000 words outweigh the 50.
It's like the disclaimers on car commercials - the whole commercial is designed to say emphatically "drive fast! drive offroad! drift! buying this car will make people want to have sex with you! you will be cool if you own this car!" and then tiny text says that's not true. The ad is leaving people with a pretty clear impression, and it's not what's in the fine print.
Okay but in this case it’s more like “Scott spent 50 words saying 1 controversial thing (“the media rarely lies”) then spends 5000 words explaining the very technically precise way in which that’s true and why it matters” and everybody just wants to argue about the 50 words.
Unfortunately, those 50 words were right in the title, and therefore colored everyone’s thoughts about it.
That’s assuming they even read it in detail at all
I think at some point, you have to accept that if the reader couldn't be arsed to read beyond the headline/title and first paragraph of an article, their opinion on the article probably isn't worth much. I mean, if Scott were being dishonest in the titles, that would be bad, but as it is, he's just making a precise statement that won't fit in the title.
If you "read" an academic paper by only reading the title and the first paragraph of the introduction, I'm not actually very interested in your opinion of the paper and its results.
I don't think the 5000-word controversial thing is the title, although that's controversial too, it's the clearly implied equivalence between infowars (total trash, never believe them), and the nyt (generally credible so long as you understand their neoliberal bias).
This seems like a low/high decoupler divide.
Scott says, neither infowars or the NYT do 7 which is very bad. Even though infowars is quite bad, worse than NYT.
A low decoupler hears: Infowars is not maximally bad! If NYT and infowars share exactly one good trait means they are equally as good!
I don’t think it is charitable or reasonable to call something a “clearly implied equivalence” when Scott explicitly and repeatedly denied that he was implying that equivalence.
I'm trying to explain why people reacted how they did. A lot of people seemed to think the post spent a lot of time implying equivalence.
The way language and human beings work is if you spend a lot of time/space seeming to say one thing, and then briefly explicitly deny that thing, people will come away with the impression that you think and meant the thing you spent a long time seeming to imply.
Most of the post implies an equivalence between InfoWars and the NYT in the sense that it denies any qualitative difference between them. Briefly explicitly denying that doesn't undo in people's minds all the time they just spent seeming to hear the opposite.
You don't have to like it, but that is how people work, as evidenced by the comments.
You’re still being uncharitable because you’re saying Scott is actually contradicting himself with his denial. He’s not - the idea that he’s implying equivalence, that the point of his article is to say “InfoWars and NYT are the same” is a facile and superficial reading of what he wrote. The explicit denial should only be necessary as a prompt to the most uncharitable readers, but they ignore even that.
“NYT and InfoWars are different, but mostly in degree not kind” is the obvious intent, and that’s not “equivalence”
It's interesting that you're doing the exact thing reading my comment that I'm saying people did reading Scott's original post - you're responding to the emotional tenor of the bulk of my post, and ignoring the little caveats sprinkled throughout. E.g. you glide right over "seeming to say", or "equivalence ... in the sense that it denies any qualitative difference" and just respond to what it feels like my overall point is.
That's a very normal thing to do. It's how human beings read unless they're being very very careful. Which is my whole point. The original post was, presumably unintentionally, written in a way that led a great many people to misunderstand it.
It spent a lot of time *seeming* to say one thing, then briefly denied saying that. People are going to latch onto lots of "seeming" over brief explicit statements. Proof: all of the many many commenters who clearly did exactly that.
I'm literally just describing why, in my opinion, that happened. I'm not being charitable or uncharitable.
Super dumb question, but what does the word "prior" mean when you use it in a sentence like this:
My prior on “a randomly selected egregiously wrong person is lying” is much lower than the sort of people who make these accusations.
I did some searching, but wasn't clear on which meaning applied.
Prior belief e.g., "I have a prior belief that the entire staff of Infowars are idiots and/or insane so take every single word on their website with heaping tablespoons of salt"
Note, this is an update of my old prior that they were all liars
It’s a Bayesian prior. Basically it’s how likely you are to believe X before some new piece of evidence about X updates your beliefs. You can look up “Bayes’ Theorem” for the mathematical context the term comes from.
But to answer the question more directly, it means the probability that you give some thing in your model of the world.
The meta-answer is "it's part of the in-group jargon here"; the other factual answers are also correct.
The biases you had going in. I can read a perfect argument for why the world is flat, but my prior for "the world is round" is strong enough that I'm still not going to believe it even when I can't find flaws in it; I'm going to conclude it's more likely I'm too dumb to catch the flaws, than that there aren't any flaws.
You see a lot of arguments around here about "falling back on priors", mostly when an argument is not obviously wrong but still kind of weak.
Prior beliefs. This is used in roughly the Bayesian reasoning sense, which has a math definition, but approximately comes down to the idea that you started with an understanding of the world, and then saw some new evidence--your resulting beliefs at the end of that will be a mix of the starting ones and the new evidence, and will depend on the strength of your prior beliefs and the new evidence.
Suppose I have a lifelong friend I know to be a gentle and peaceful person. A local crazy lady claims to have seen him murder someone. My ending assessment of the probability that he murdered someone is probably still really low--the quality of the evidence is low, and my prior beliefs are strong.
By contrast, suppose the police then find the body of his ex-wife shot with bullets identified as coming from a gun registered to him, and a surveillance camera at a local convenience store appears to show him hauling a dead body in the back of his pickup truck the night she disappeared. At the end of that, my assessment of the probability that he committed the murder is going to be much higher.
😧
I'm not saying this piece is exactly as bad as infowars...
But that's a sentence just begging for a 'but' now isn't it?
Most journalism is hearsay (reporters telling readers what somebody else told them), but people still want it be admissible as evidence in the court of public opinion. I think that's behind people's dissatisfaction with the argument "this article didn't say X was true, it just said that so-and-so said X was true, therefore it's not lying".
Side note: I help fund a YouTube channel whose mission is producing documentaries about historical firearms. The main guy behind it is really, really serious about getting things right, but obviously he wasn't alive or there when e.g. Sam Colt was developing his first revolver designs, so he has to rely on books. To his horror, he's come to find that most books that claim to be authoritative on the subject, and that people treat as authoritative on the subject, don't cite any sources, have inconsistencies, etc. People have been repeating the same "facts" for decades as one book cites another book that cites another book going back decades upon decades and when he finally traces it all back, finds that the original source just said things without any kind of primary source documentation and presented them as fact. Or that the primary sources themselves are contradicted by physical evidence. It's got to be exhausting. The idea that anybody can really know what happened hundreds of years ago honestly seems ludicrous. (Edit: link to a video summarizing some of these woes here: https://www.youtube.com/watch?v=whYgOjAkC2g)
I long ago gave up on the idea of ever being able to actually know what happened about any remotely controversial event, e.g. Rittenhouse or whether COVID was a lab leak, because by the time it's become controversial enough for me to care what the truth is about it, there's already a zillion articles out there written by people with their own biases and agendas. Frankly, the world might be a better place if more people simply accepted that if they weren't there, they can't know what really happened. All they can really know is what people say happened. And even if you know what the people who were actually *there* say happened...I mean, would you trust anybody who was inside the Capitol on January 6th, 2021 to give a completely unbiased and factual account of what happened there? Do you even trust their memory of what happened, given that memory is imperfect? And yet whether the events of that day was an Trump-organized attempt at a coup d'etat against the very foundations of our democracy, or a just a bunch of idiotic protesters larping as revolutionary war heroes, seems really, really important to a lot of people. And so they want to be able to point to stories in the the news and say "see, this NYT article says X, so I'm right!"
"William De Cognisby came out of Brittany with his wife Tiffany and his son Manfras and his dog Hardigras."
https://www.youtube.com/watch?v=qEV9qoup2mQ
Oh yeah, that was a good one.
The lab leak thing is complicated by the fact that everyone from some hapless lab tech who stuck himself with a needle by mistake and came down with a weird flu a week later all the way up to Chairman Xi has a huge incentive *not* to allow any evidence of a lab leak to come out, and at this point they have probably had the time to make sure that most such evidence is gone. That doesn't mean that it was a lab leak, just that it is very unlikely we will ever see evidence that it was a lab leak if it really was.
Yes. I meant to include some words along the lines of how it's quite common for the folks closest to the truth to have the least to gain for the truth to be known, but I'd already spent enough time working on my comment at that point.
> It's got to be exhausting. The idea that anybody can really know what happened hundreds of years ago honestly seems ludicrous.
To repeat a story I heard somewhere (or in other words, I make no claims as to factuality):
Sir Walter Raleigh was once imprisoned in the Tower of London and decided to use the time by writing a history of the world.
While he was there, some work was being done on the tower, and on one particular day, a fight broke out among the workmen.
Shortly afterwards, Sir Walter made inquiries of the workmen.
He found that he was unable to determine either (1) what had happened during the fight; or (2) the subject of dispute.
From this, he concluded that writing a history of the world was mostly just a waste of time.
Your view of media outlet truthiness:
UTTER BOLLOCKS ...................X (INFO WARS)........X(NYT)............................TOTALLY TRUE
Readers who complained you were too easy on INFO WARS/hard on the NYT:
UTTER BOLLOCKS ..X (INFO WARS)................................................X(NYT).....TOTALLY TRUE
Readers who complained about you being too easy on the NYT:
UTTER BOLLOCKS...........X(NYT).......................................................................TOTALLY TRUE
[INFO WARS not shown, off the left hand side of the page in the abyss]
You aren't going to be able to reconcile those three models even if you're upfront about expected disagreements and load your article with caveats and very precise subtitles. I wonder how much of the technical points about the finer details of the denotative interpretation of the word "lying" is rationalization over a gestalt model of where the players fall on the truthiness spectrum.
>I almost always see people erring in the direction of accusing people too quickly, rather than too slowly.
On this note the devil is called “the accuser of our brethren” in Revelation 12:10. There’s definitely a Christian notion that eagerness to accuse is a major human evil.
Something that I was surprised to not see addressed is the distinction between media and written articles. Written articles are the easiest to analyze, but "the media" includes other, well, media. InfoWars is most famous not for the website but for the radio show. Fox News is most famous not for the website but for the TV channel. It may well be that the same analysis holds equally across all forms of media, or roughly equally across most forms of media, or it may vary widely. But I don't recall seeing this addressed anywhere in the various pieces by Scott or the comment sections, and I found the omission worth noting.
I’m in camp “This is a really important point, but framing it as a semantic question about lying is distracting people from the real argument being made.”
You aren’t actually making a nitpicky technical point, you’re saying that *for the purposes of censorship* any objective measure of truth would either be over-broad (and thus inconsistent in practice), or else so narrow to be useless. I think people are just getting distracted because it’s been framed as something much more pedantic.
This is something I'm not sure about. There are places where Scott seems to be saying this and there are places where he seems to be trying to make a broader point. I think that more clarity on the scope of the claim would help.
Now we have this taxonomy of dishonesty we can look at it from the other side. A taxonomy of levels of belief correlated with 1-7. I think someone who wishes for people to believe x may not have to resort to 7 or even 6 very often or at all. 4 and 5 can funnel someone into believing just as effectively in most cases. It’s a different question though. From the perspective of designing standards of truthiness, it might have to be requiring (like some states’ ballots) an argument, a counterargument and two rebuttals. I don’t know.
He is strongly suggestive that the point of this is about censorship. But I think that is not actually the central point of any of these articles - just a corollary he hopes we think follows from the claims here. (I actually think we should just be more comfortable with censorship of some things that aren’t literal category 7 lies.)
I guess you could look at it that way.
from my perspective, if you focus on censorship, then his definition is reasonable and useful, even if a little distracting.
If you look at it towards any other point, then it's just nitpicky and pedantic. This might be an example of the very common phenomenon where someone starts with a very specific point, but then drifts elsewhere while responding piecemeal to the specific comments of others.
But Scott framed the point about the media never lying as merely a lemma needed to reject a certain facile argument for bans on misinformation.
If he was a dead author then maybe we should reinterpret him but given he's an active participant in the dialog I think we should assume that's the point he wants to make and why he finds it interesting until he indicates otherwise. Besides, I think good Gricean norms mean we should assume that the speaker thinks the interest of his claim will be apparent to his audience. Without the initial framing it's unclear why we should care about the fact the media never lies under this definition. And some writers are obscure like that but Scott pretty much never is.
Why should we be more comfortable with that kind of censorship? Seems to me that the long history of censorship by powerful entities suggests we should have a strong rebuttable presumption against censorship of things that seem intuitively 'bad'.
I don't see how we could possibly have the kind of evidence needed to rebutt that presumption. Since the harms from censorship are usually of a sort that isn't recognized at the time we'd need a long time period in which we were comfortable with what you suggest that we still judged was desireable long after the fact.
Even at just the object level it's not even clear that the kind of censorship you suggest is even an effective way to decrease the bad beliefs in question. It's not even clear that the European laws about Nazi propoganda have increased of decreased the appeal of the far right (being forbidden adds cachet).
Not to mention the fact that the mere knowledge that certain views are being suppressed risks lending credence to all the worst kind of beliefs since they can (more credibly) just insist the reason you aren't hearing about it from reputable sources is the censorship.
I'm not particularly relieved to know after decades of reading that the NYT excels at "ethically" misrepresenting the world on certain subjects. For most things that don't touch the culture wars the NYT can write some exceptional articles where fact checkers do their job in a professional unbiased manner.
Otherwise ... they paint a not representative picture of the world primarily through selection bias and framing bias. Selection bias is rather obvious, excessive highlighting of the right's culture war flaws and the near dismissal of the left's culture war flaws. One is forced to read multiple news sources, you just can't know what is missing.
Framing bias is another art form that any skilled reader can easily spot lately. Which subjects are given the news analysis treatment and which subjects are allowed to have opinions from biased experts quoted liberally throughout the piece virtually unopposed. Expert laundering has become pervasive across the news industry partially leading to the decline of trust in experts. Some subjects are heavily investigated and others are ignored.
Subjects that don't fit a preferred narrative that cannot be ignored are usually given a "just the facts" treatment with a strict line on what are facts and what are not, as well as zero curiosity to ponder obvious questions out loud. No opining experts for context here. These storylines typically last no longer than a day or two.
Random example of common framing bias today, subhead on front page:
"The revelation is sure to intensify Republican attacks on President Biden, who has called former President Trump irresponsible for hoarding sensitive files.
It is not clear where or when they were recovered. But aides have been scouring various places since November, when a handful of classified files was found."
The story is apparently Republicans Pounce and the Biden team earnestly working hard to find all the missing files. Reverse the identity groups and these headlines will change. It has become very tiresome.
Sometimes group identity information is reported, sometimes it is not. When it is included in the headline or first paragraph it tends to be a preferred identity narrative, otherwise it is relegated to the end or simply not reported. For most news sources you can simply look at which anecdotal stories they choose to cover and that will highlight the preferred narratives with near certainty.
Knowing exactly where the industry's current ethical lines are and adhering to those slavishly doesn't stop one from being able to cleverly misrepresent the facts.
I guess people like accusing others of lying because there's a universal(?) human norm against lying so it's a quick and clear way to discredit someone. Saying someone “is grossly negligent in their reasoning and is happy to not investigate beliefs any further that are emotionally satisfying” isn't quite as pithy and there's not as strong a norm against this behavior. But it would be good if we had words for this.
Should we have a contest to propose words for these concepts?
If the target audience gets just as convinced by 2/3/4/5/6, they don’t have to attempt 7 very often (or at all).
If most journalists in a given newsroom would top out at 4, I think all you need is one editor or even a colleague who does 6 to start seeing effects in the writing. I think it’s slightly contagious, higher IQ is often (not always) more resistant to it, and the more it happens, the more the higher IQ people leave the profession. But I appreciate this taxonomy of dishonesty, it’s a helpful scale.
Damning with faint praise is the whole point, guys.
More praising with faint damns, actually.
One important point is that unless you are smart and good at acting, delusion is much easier and cognitively more efficient than lying.
I'm sympathetic to your views on this overall, but I do think I'd categorically consider 6 to be "lying". At the end of the day, when we talk about why lying is bad, what matters is really the intent to deceive. "Lying by omission" is a standard phrase, after all.
On the other hand, intent is nearly impossible to discern with certainty, so I largely avoid trying to make any judgement calls on it. Accuracy rates and targeted bias are measurable enough that I can assume no one is actively trying to lie without it affecting my opinion on how I should respond to any given statement.
I disagree with the line-drawings on the Infowars claim in section 2. Infowars saying "the document is a shoddily contrived hoax" (if that's a direct quote from Infowars) is doing something different than what most media outlets do.
Most media outlets generally try to avoid explicitly saying false things. They are willing to do various other things that are misleading, like presenting information in distorted ways which suggest a false conclusion, or abridging a quotation and altering its context, or citing a false claim that someone else made in a way that suggests that it's true. And they sometimes make mistakes and accidentally say false things, e.g. when they're confused (often in an ideologically convenient direction) or when they get sloppy and fail to find artful phrasings that avoid making explicit claims. But they'd try pretty hard to avoid calling an accurate document a "hoax".
If this discussion is trying to offer models of how the media does and doesn't distort the truth, then IMO there is a joint here where things should be carved, putting this Infowars stuff on one side and most of what you'd see in the Washington Post or whatever on the other. And that seems more relevant than line-drawing around the word "lie", and separate from trying to weigh in on how much we're bothered by different sorts of distortions.
The defense of including "the document is a shoddily contrived hoax" in the same category as everything else retreated to the idea that whether something is a lie depends on the mental state of the person saying it, and we can't know for sure what the author believes.
But that's steering the discussion away from trying to actually model what the media is doing. Plenty of people believe that Epstein didn't kill himself, and it would not be at all surprising to see a reddit post that asserts "Epstein didn't kill himself." But it would be surprising to see a Washington Post article that asserts "Epstein didn't kill himself."
Even though there are probably reporters and editors at the Washington Post who believe that Epstein didn't kill himself. Because the Post, and most major media outlets, are doing something different than what Redditors do. By asserting a claim about the world, like "Epstein didn't kill himself", the media outlet is engaged in *reporting*, and that carries with it a process of 'trying not to report inaccurately' by checking if the claim is true / what it's backed by. And 'the author believes it's true' or 'there exists an argument for it' are insufficient.
Possibly dropping the word "lie" from this discussion altogether would be an improvement, because the question of whether the author might have believed what they were saying is more relevant for arguing over whether it technically counts as a lie than it is for talking about how the media usually works and out-of-the-ordinary an example like this is.
I’m not convinced that the 9%/7% inflation thing is actually false. Reported inflation rates are usually not point-in-time estimates of the current annualized rate of change of prices, but are usually instead reports of the year-over-year change from 12 months ago. If the yoy change in one month is 9% and the yoy change in the next month is 7%, it seems quite likely that prices actually *did* come down in the intervening month. But it would take a bit more digging to be sure.
Off the top of my head, there were already talks of dangerously accelerating inflation last December (at least in Europe), so the argument that the Dec 2021 baseline is about the same as the Nov 2021 is at least questionable. I know for Germany that prices did indeed come down, but only by -0.4% MoM (give or take) in the face of a -2.x% YoY decrease, suggesting a fairly high MoM increase from November to December 2021.
I think the miscommunication here is that when talking about whether a statement "A => B" is true, for some people that means "given known and fixed truth values for A and B in the real world, is the statement true", and you make the point that it's true because B is true (and A, too, but that doesn't even matter).
However, other people might treat it as true only if it is universally true for all A and B because B inherently logically follows from A. And under this interpretation, the statement is fal.. not universally true, because a price decrease cannot be inferred from shrinking YoY inflation rates alone.
I agree however that it's fair to expect media to be in the business of reporting facts in the real world rather than mathematical reasoning on an abstract level.
It is literally true that prices declined (not just increased at a slower rate) in December, as measured by e.g. the CPI index - Federal reserve series here: https://fred.stlouisfed.org/series/CPIAUCSL . So the commenter who claims it is a lie is instead mistaken. This is of course another vector by which people might think the media lies more than it does.
EDIT: Looking back the commenter's datestamp was 12/29/2022. Not sure what article they were looking at, and not sure what month's data was being reported. That said "judge is incorrect" (like me, in this comment judging another comment, maybe!) still stands as perhaps additional way the media might be perceived to lie.
Exactly — I came here to say the same thing.
I think some of the confusion might be between "not lying" and "we don't know if they're lying or not, because we don't know what they really believe." We begin in ignorance of other people's beliefs and often don't learn enough to be sure either way. But if you say "not lying" then it sounds like you're making a claim about their state of mind, when often we don't know that.
Maybe it would be clearer to say that proving strangers to be lying is hard.
The story about Jones and the radioactivity to me isn't all that clear-cut, maybe actually supporting Scott's point.
If Jones just wanted to push some false claims about radioactivity levels, why go through all the effort? Why send a team to California, why use real Geiger counters (somebody must make fake ones for movies, right?), why not use some legal, naturally radioactive substance to deliberately create a fake measuring spike (in relative terms, even if not harmful), why take the risk of live reporting? Given that they were absolutely not set up to produce fake reporting, Jones screaming at their lack of finding something suggests that he seriously expected another result and irrationally didn't want to update, resisting it quite ferociously. Otherwise, why not accept that there's nothing to then instruct them to just make something false up, rather than screaming and telling them to merely *not* say that they didn't find anything (which, for all he knows, might well be because of $other_conspiracy_theory).
Did the reporters eventually report something which they knew to be a lie? The quoted excerpt actually sounds like they *didn't* because they "couldn't stop [saying they weren't finding anything]", but I'm not familiar with the original story. Either way, it sounds like a reporter not being willing to claim evidence they didn't find, and his boss being furious at him for not finding evidence he is certain should be discoverable.
Jones has certainly deliberately pushed lies in other contexts like Sandy Hook, but if the above story really is the best known example, I remain unconvinced of the prevalence of that being exorbitantly high.
I recall Fukushima fairly well because I am a night owl who was up late as the news broke from Japan. Mainstream news media headlines tended to run something like this from Day 1 to Day 4 (note, these are just semi-comic reconstructions of my memories, not quotes):
1. Fukushima Nuclear Power Plant Swamped, but Experts Say No Radiation Danger
2. Building Blows Up at Fukushima, But Danger Minimal
3. Second Building at Fukushima Explodes, But Risk Is Limited
4. Blast Obliterates Third Fukushima Building
As it turned out, California wasn't drenched in radiation from exploding buildings, but events were not reassuring at the time. You didn't have to be an Alex Jones fan to be worried by this progression of events.
From the anecdote, it sounds like Alex Jones spent real money in the hope of getting a real scoop, which speaks better of him than I would have guessed.
It sure seems like a news organization that cared could have gotten together people with the right expertise to make pretty accurate predictions about the consequences. My assumption here is that they didn't because they didn't care that much--panic was a better response for them in terms of ratings, and they had no incentive to make accurate and boring statements like "yeah, this is probably pretty unhealthy for the people trying to stop the crisis, and I wouldn't go fishing right outside the plant and eat what I caught just now, but it's almost certainly not going to affect anyone in California."
Making accurate predictions is not their job, and is hard and takes work and knowledge they mostly don't have, so they don't do it.
The mainstream media kept implying that there was little to worry about at Fukushima, and then, three days in a row, buildings blew up. I was surprised that the MSM was downplaying the risk from a nuclear power plant disaster, but that was the general tone of coverage.
Personally, I found this pattern worrisome at the time.
Fortunately, after awhile, the nuclear power plant finally stopped blowing up.
I posted the excerpt (and Scott linked to the full NYT story). You make some great points.
For background, I think IW consists of a spectrum from true believers to psychopaths. The latter prefers to keep their lies deniable. Pointing to a specific statement as a lie basically requires a glimpse behind the scenes (which aren't exorbitantly common).
Here's my model of the situation:
TL;DR Jones sent out true believers to California. He wanted them to lie, but they didn't understand (or at least, would not lie without direct orders).
In terms of procuring legal radioactive substances or fake Geiger counters- I think that would be legitimately difficult. There were significant time constraints, and these aren't clever or well-connected people even if someone did think of it. (And I suspect that an InfoWars employee attempting to procure radioactive substances "ASAP" would be denied and flagged regardless of quantity.)
In terms of why the employees were sent out live in the first place: I think the strategy has been successful before- it's easy to convince an audience that the person in front of your camera is a baddie, even if it's live. But this time two true believer-ish employees were sent out to check a machine's reading of the air. I think what happened is "Mistakes were made"- Jones and upper management simply didn't consider that perhaps the camera crew wouldn't have the means or willingness to fake a dangerous situation.
As to Jones' behavior- I don't think he breaks keyfabe very often. He wanted them to figure out how to fake the result, or if nothing else, pretend to be run off the road by feds. (The employee in the quote later mentions Jones doing the latter.) But usually this is communicated via "Alex is mad at you, make him happy." With two (true-believer) employees in another state who didn't realize what their task was, he wasn't willing to risk directly ordering them to lie. Doing so would mean realizing he's a fraud when he isn't there to manage the situation.
On the “fake news” thing - I do recall seeing a few “articles” on sites like weather737.com, claiming “meteorologists forecast largest blizzard in history to hit North Carolina in 12 days”. Totally outrageous, beyond any reasonable forecast, but coming soon enough after a polar vortex event and about a date far enough in the future that it will have lost social media currency by the time it is checked. I don’t exactly know why these briefly proliferated (probably some time in the 2014-2016 era?) but then disappeared, but it was a fascinating phenomenon, and not only about political news.
This series has really reinforced my acceptance of 'Politics is the Mind-Killer'.
This really has been primary takeaway as well.
I worry this definition of a lie is so narrow and so unprovable that it's a bit of motte and bailey when taken back to the original justification (that censorship attempts are ill-advised since the media rarely lies) and weakens that argument. The call is to censor misinformation or disinformation or 'fake news' and that is not equivalent to censoring lying if you're going to use a such narrow definition and is likely much more straightforward (if imho still ill-advised).
In particular, your definition of lying leans heavily on intent which is obviously pretty hard to conclusively determine. Short of leaked chat logs saying 'ha, ha I'm totally making this up' from the 'We Are Lying Liars' chat room it's going to be difficult to prove they are lying even if they are (though if you follow Matt Levine this seems to totally be a regular thing in finance so perhaps the absence of regular leaks of this nature should say something on earnestness / self-delusion of the journalist cast). I think the a better title for the first piece is 'The Media Can Vary Rarely Be Proved To Be Lying' rather than 'The Media Vary Rarely Lies' which is in some sense is unknowable for many cases using this very narrow definition of lying and probably works just as well for your argument.
But back to the censorship point, the steel man case probably doesn't take intent into consideration at all. But rather seeks to ban factually incorrect statements (whether by malice, incompetence, bias or otherwise) or mandate the presence of certain context on particular issues. And that seems potentially possible and orthogonal to whether the media lies to you in some very strict sense of lies. We have (flawed) social mechanisms to determine is something is proven to various degrees of certainty and use them in critical places, courts for one example.
So, there's been a huge amount of discussion, what the word "lie" means, which is very hard to resolve because it's a fairly vague term and even more to the point, it's an emotionally loaded word with accusation and anger as part of its meaning. Not only that the discussion was focused around some material that's already very emotionally charged, like Alex Jones vile attacks on families whose children were murdered. I think it's kind of impressive that the discussion has been relatively sensible given this context.
With regard to your conclusion, Scott, I think it's still a bit superficial. You seem to assume that finding some other source for what you're saying makes it a different issue than outright making things up. Tell that to an AI that turns racist reflecting its input. And we humans often make up our minds about things for non-rational reasons then construct rational sounding arguments to support what we're saying, especially when talking politics, especially the sort that's basically an emotional minefield.
You seem to dismiss the idea that Jones decided it would popularize him to come up with an attack on the Sandy Hook victims, then looked around to find some flimsy stuff that could be used as his "source", so he couldn't be accused of making up what he was saying. How is that different than making it up, except to give him a way to claim he didn't make it up. It's very easy to "support" pretty much anything you want to say this way It didn't take Musk long to find a specious source of really nasty stuff to retweet about Paul Pelosi, and was that different in any meaningful way than his made-up "pedo boy" crap?
I don't see how making distinctions between between misleading statements supported by the flimsiest of sources and making up facts is especially meaningful, in a context where you can find a tweet or a posting to support most any position you want to take.
But also, a story composed of made up facts can contain truths that can't be told as effectively any other way. (I guess that's why you make up your allegories.) A factual account is always going to be subject to bias and error. It's how humans communicate. People communicate to help each other understand or to mislead each other, among other reasons. To me, distinctions between made-up facts and misleading interpretations ultimately don't really make anything much clearer, especially when we're talking about communications with content that's basically emotional rather than factual.
>[your categories]
I've agreed with all your posts on these topics. I find myself not motivated to post when I agree.
7 is lying
5 and 6 are being deceptive
3 and 4 are being incompetent
2 is forgiveable
I find 3 and above bad enough, that once we get to 7 I just don't care very much. My starting point is that the news media I want to consume should be trying to convey the truth, and anything past 3 is a failure to even *pursue* that mission.
I think some of the frustration in comments is that it seems like your standard for the media is "they aren't doing a really bad thing", but many people's standard is "are they doing a a good thing".
Media : Lying :: Doctors : murdering :: Teacher : losing kids
Media : Conveying Truth :: Doctors : Getting you healthier :: Teacher : Teaching things
And if Scott had written an article called "Doctors very rarely murder people", commenters would be missing the point if they replied with lots of examples of doctors failing to cure people.
More importantly, it's not the case that Scott is being overly lenient or complimentary towards the media. His point is that even when the media do these clearly bad things, they're not technically lying, therefore it would be extremely hard to come up with a set of laws or censorship algorithms that prevent the bad things. He's not disagreeing with you that the bad things are bad.
The question is around intent. Doctors very often kill people (medical mistakes kills 10s maybe even 100s of thousands each year). But do they murder them? That requires intent and definitions vary, is operating drunk murder? While ill and sleep deprived? After a bad breakup? Civil and criminal courts and judicial codes have a range of intents and responsibilities and muddle through. Scott's using a First Degree Premeditated Murder as his standard of lying, and sure fine. But I don't really care if my Dr. murders me or just kills me tripping on LSD or something and the more defensible misinformation censoring is more akin to mandating that Dr.s not be high and pilots have 8 hours of rest than outlawing murder.
You as the dead person may not care whether you were murdered or accidentally killed, but the rest of us care very much. A person who kills by accident will probably feel remorse and will take care not to repeat the act, since he has no reason to do so and probably good reasons not to. But a person who murders on the other hand is much more likely to repeat the act -- since he obviously found a reason to do so that was not sufficiently restrained by conscience and strong social disapproval.
Such a person is a far greater danger to the rest of us than an accidental killer, even though for you (one victim) the difference is academic. That's why the law puts a very, very heavy stress on intent. Intent tells us how likely this person is to be a continuing or greater problem.
Banned for this comment. I think I might have posted a link to an article about why this was false, but even if not, accusing me here just seems weirdly aggressive.
Could you (or someone) post such a link again? I can't remember when this ever came up but would be genuinely curious.
Why should we care if the media doesn't lie under your definition of lie? How does that relate to the original argument claiming to discredit the facile argument for blocking misinformation?
I thought you wanted to argue that you can't say that such bans only bar clearly valueless content because they would inevitably have to ban content that isn't lying.
But with this narrow definition of lying why should we assume that you can't only ban clearly valueless content (say the weekly world news) just because it's not lying under your definition?
I ultimately agree there are problems with such bans but I feel like the debate has gotten a bit too focused on whether the media lies to the point of losing sight of the original reason this claim mattered.
A lot of us care about understanding how the media does and doesn’t “lie”. It doesn’t only matter as an argument against censorship.
First, to be clear, I meant why should we care in the context of the argument Scott was making.
But on the general point I suspect you care about how the media behaves not whether the meaning of the word lie is such that it accurately describes their behavior.
If you don't give any special weight to the word lie then there isn't really anything illuminating about media behavior in Scott's claim. Scott could have conveyed the same content about media behavior merely by saying: the media almost never make assertions in their own voice and even the most misleading claims are usually written as a literally true statement that some other entity made some claim. Very few people would have doubted that claim (whether or not that's literally never outside of disgraced reporters or virtually never doesn't seem very important)
The only reason it seems like some substantial claim was made is because of our prior associations with the word lie. Once you simply take it as a stipulative definition then Scott's claim doesn't really have much substance.
Thanks for pointing this out.
Indeed we were sidetracked by the arguing about definitions, while it's not even that relevant to the original question.
> I’m kind of sensitive to this because for almost every article I write, people in the comments accuse me of lying, or “pretending” I don’t know why the statements I made are wrong, or some other offense which I plead innocent to.
You asked for that treatment. Once you've made statements like this:
(A) When there's an official orthodoxy, why can't scientists just pretend to believe it in public while privately knowing it's false? ( https://slatestarcodex.com/2017/10/23/kolmogorov-complicity-and-the-parable-of-lightning/ )
(B) Huh, it sure is weird that all the research on racial differences in intelligence says one thing while the official orthodoxy says something completely different. Oh well, what are you going to do? (Early ACX; not sure how to find it)
(C) I know about Steve Sailer, and I think he's right about this, but I'm deliberately avoiding citing him. ( https://astralcodexten.substack.com/p/what-caused-the-2020-homicide-spike )
...people are going to conclude that when you say something, you probably mean something else.
Not to put too fine a point on it, but if Steve Sailer is persistently right about some important issues that most media sources are persistently unable or unwilling to report well on, why does that lead you to not cite him?
A lot of people hate me and believe that their hatred for me must be because I am hateful rather than that they are hate-filled.
It's a difficult conundrum for Scott. He wants to tell the truth and thus do good for humanity, but he knows that likewise I've been telling the truth for decades in order to try to do good for humanity. But I mostly inspire mindless rage and increasing ignorance and stupidity. So Scott tries to subtly obscure and bury the turth under a torrent of words so that only the bright will understand while the less intelligent will assume that, while they can't quite follow him, obviously he can't be like that horrible hateful Sailer.
God bless him, and I hope it works.
> I asked “Why? A woman said she had been raped, and Rolling Stone believed her. The woman was making it up, but Rolling Stone wasn't” and Deepa commented “Isn't it the job of a reporter to investigate? And be good at it?” I don’t want to pick on Deepa, but this is what happens when you have an overly expansive definition of “lie”!
On a psychological level, it's plausible this is an "honest mistake" on Rolling Stone's part.
On a practical level, 1. it led to them writing a widely-disseminated article that made very many people believe a straight-up falsehood, and 2. it is pretty obvious that this failure of due diligence was in service to advancing their ideological agenda. (Compare how Rolling Stone would treat e.g. a story about a trans woman assaulting someone in a women's bathroom. In this counterfactual, despite being very similar on the object level, it's obvious they would do their very best to either debunk the story or bury it; there is no world where they credulously run a story like the Jackie one.)
You say:
> When someone says “Joe is a liar”, I don’t want to have to ask every time “Do you mean you have some actual evidence for this, or that they said something you disagree with and you instantly leapt to ‘this is reckless disregard for the truth because nobody could ever be so dumb as to honestly disagree with me’?” I think if we let people use the word “lie” this way, then the overwhelming majority of accusations of lying would be false. Why would we want to define a word in a way that dooms it to constantly be used incorrectly to mislead people?
But this is fundamentally missing the point. The main inference people are usually trying to convey by saying "X is a liar" is something like "X is an utterly unreliable source of information and should not be given any credence". This is a statement solely about X's claims and their relationship to the truth; it need not rely on any inference about X's state of mind.
It may be psychologically interesting to dive into the question of whether X is knowingly making false statements, or knowingly deceiving, or merely speaking with reckless disregard for the truth. But from the point of view of the information consumer, these distinctions don't matter. All the reader needs to know is that X is grievously unreliable.
In practice, almost all media outlets pass the bar of being grievously unreliable as information sources. If you're the priest or the therapist of an NYT reporter, you should make the careful distinctions about the exact psychological impetus behind their deceptions. But if you're merely a reader, the important thing to take away is the fact of their unreliability.
An audio tape of Rolling Stone reporter Sabrina Rubin Erdely's conversation with UVa coed Jackie Coakley was released during the libel trial. The funny thing is that the student is clearly the alpha female of the pair, dominating the conversation with a steady flow of BS, while the 40ish journalist chips in with a few nerdy true believer factoids about how studies show that rape on campus etc etc that Jackie then runs with.
A really interesting historical work could be written about the collaboration amongst activists, journalists, and the Obama Administration in the years up to 2014 that laid the groundwork for the notorious Rolling Stone fiasco. It had been pretty clear to me from 2013 onward that the New York Times and the White House were conspiring to make Rape on Campus into a huge crisis. But nobody seems very interested in documenting afterwards how these type of Democratic Coalition campaigns are organized.
It seems like there are two things that went wildly wrong with this story (though note that I didn't follow it beyond reading the original story and some online discussion as it unraveled):
a. Erdely should have been a more skeptical observer of the story. There were some obviously implausible bits (broken glass), and it seems like even minimal checking was enough to discover big holes in it. So why didn't Erdely chase down those details and realize there was a problem?
b. Rolling Stone's fact checkers and editors should have been a second line of defense against RS publishing a story that would end up blowing up in an embarrassing way that damaged their reputation and got them sued.
In Erdely's case, maybe she was just totally convinced by Coakley's story and mannerism and believed her. You can imagine how a lot of cultural messages about believing women and rape culture on campus and such might have contributed to this effect.
Alternatively, perhaps she thought that the details of the story were unimportant because it was conveying an important truth--that campus rapes are a serious problem, that frat boys are bad, that UVA is full of a bunch of dirty outgroupers who deserve bad things, etc. Certainly, a huge amount of the support the story got in media and online before it fell apart hit those points pretty hard. How dare you doubt that this friendless old lady I just grabbed is a witch--are *you* a witch, too?
But it sure seems likely that she basically knew this story was full of holes, but thought that nobody would ever check deeply enough to disprove it, and that she'd covered her ass sufficiently by having a source she could say she'd accurately quoted. That is, she may have been lying in one of the ways Scott says media do sometimes lie--by quoting a source saying something that's pretty implausible and full of holes, but then not engaging in any further questioning or thought that might mess up the story. I think Megan McArdle used the term "too good to check" when describing the story initially.
The first motivation doesn't apply to the editors/fact checkers, but the last two might. Perhaps they thought "this is Socially Important" and "one does not doubt the word of a woman who claims to have been rape, that's just Not Done." But also, perhaps they thought "this is going to be a very big story, we'll win awards--so maybe it's a little shaky, but who's ever going to dig enough to find out?"
Did enough information come out in the lawsuits and such later to untangle which of these is more likely? If Erdely and her editor/fact checkers were reasoning along the lines of "this story is probably made-up but who's ever gonna know?", that implies that probably a lot more big stories along these lines were similarly shaky, only nobody every investigated further and so the full-of-holes story just became accepted as truth everywhere. That certainly has happened in a number of high-profile stories over the years, but I have no good intuition for how common it is.
https://en.wikipedia.org/wiki/Janet_Cooke
The Wikipedia entry on Erdely is also interesting, highlighting two other news stories about rape victims, one of which also appeared to have major problems.
An important thing to remember is that the Rolling Stone "Rape on Campus" story was publicly admired by many leading journalists, such as Jeffery Goldberg of The Atlantic, who went out of their way to tweet congratulations to Sabrina Rubin Erdely. It was the Spirit of the Age. Hadn't we been running articles for 18 months in support of the Obama Administration's "Dear Colleagues" letter to college administrators about the frat boy rape threat?
Five days after it was published, Richard Bradley, an editor in the financial press, became the first professional journalist to doubt it publicly, and I was the second four days after that. That seemed to open up the floodgates for skeptics. But that was a long time before more than two journalists doubted this celebrated story.
Most of the journalists who retweeted the article appear to have reasoned: Of course blond fraternity boys at the U. of Virginia are smashing glass in their rapine and hitting the poor source on the head with a thrown beer bootle. That's what blond boys do: they hold kristallnachts.
Ironically, the Rolling Stone hoax led to an actual kristallnacht on campus with Woke students smashing the windows of a frat house.
Here's my 2014 analysis of Erdely's absurd article.
https://www.takimag.com/article/a_rape_hoax_for_book_lovers_steve_sailer/
The main thing I've changed my mind about is that in 2014 I gallantly assumed Jackie Coakley was some poor girl in need of protection from exploitation by Rolling Stone. But my impression now is more that she's a George Santos-type who likes concocting tall tales about herself and that she exploited poor, naive Rolling Stone with her ridiculous story lifted in part from "Dawson's Creek."
Heavens, redefining "liar" to mean "source of unreliable information" is quite a big jump. The Internet would become a very quiet place if we were all to think that way.
Why would you say that? Quiet requires that the speakers stop speaking, not that the listeners stop listening (or believing).
In any case, the admonition used to be to "not believe anything you read on the internet"[1], and I'm not at all convinced that it does not continue to be sound advice. Insofar as my question remains "whether I should update my model of the world based on this information?", the sort of distinctions being discussed here are a pointless distraction.
It does not matter whether someone is technically a "liar" under some stupidly pedantic definition of the term, and never has. What matters is "can I expect the received information and proffered conclusions to be more reflective of the truth than not?", and a taxonomy of the various ways of answering "no" to the foregoing is more of a hinderance than a help.
It's not even particularly helpful to considering the issue of censorship, because the key issue there is "can I trust you, who will be doing the censoring, to only censor the bad stuff and not the good stuff?" Which can instantly be answered in the negative for all values of "you, who will be doing the censoring", "bad stuff", and "good stuff".
If you *really* want, the matter can be reasoned out in greater detail, but I find that "Fuck no!" is a perfectly adequate argument in this case.
---
[1] We're talking about the later '90s, when the various cranks and conspiracy theorists had already hit the net big time, but mainstream media still mostly functioned in the physical realm
I'm pretty convinced that if the listeners stop listening, the speakers will stop speaking. Very few people will put the energy into speaking if the evidence suggests nobody is listening.
Also, I suggest you are completely wrong for the general population (i.e. those other than yourself) about the importance of whether someone can be called a "liar" to general approbation or not. As evidence I would adduce the heated conversations which you can see have formed right here on the subject -- clearly most people *do* care about this, a lot.
The reason they care is because "liar" is highly pejorative. It's the difference between saying a woman is "sexually confident" and saying she's a "slut." One will get you punched in the nose, the other won't. So when people debate "When is it OK to call so-and-so a liar?" they are debating (in addition to the named topic) "When is it OK to say so-and-so is an Untermensch, an undesirable, someone we are justified in socially shunning, mocking, subjecting to both informal and formal social penalties?" It's a big deal. The distinction between saying "He was wrong/misinformed/bullshitting" and "He's a liar" is very strong.
Consequently, people who permit the word "liar" to be applied to people who are merely mistaken, or reasoning poorly, or even bullshitting, will find themselves increasingly socially isolated, since that's not what everybody else means by it, and it would be as offensive and likely to end badly (for the speaker) as replacing "sexually confident" with "slutty" in college classroom conversation.
When I took an intro semantics linguistics class in collage, I was told, bluntly, that there is not and cannot be a single satisfying all encompassing definition of either "a lie" or "lying," and that the best we can do is jargon that covers restricted cases (e.g. perjury, defamation). Otherwise, we're stuck with the unsatisfying conclusion that the concept of lies and lying is a broad category with a clear center, but fuzzy boundaries similar to the concept of a heap or flatness. This dogma also came with a whole host of interesting edge cases, for example:
A man lives in a home on a suburban cul-de-sac, his neighbor asks to borrow his snow blower. The man, knowing that his neighbor had destroyed his possessions in the past and not wanting to loan out his snow blower replies "I no longer own a snow blower." This has everything you could want in a case of lying: 1) a statement that is false, 2) intent to deceive, and 3) creative genesis of a narrative (the "making" part of "making stuff up"). You can change the story to remove each element: what if he thinks he has a snowblower, but his wife sold it earlier in the week without telling him? What if he just hallucinated that he didn't have a snow blower? What if he transferred ownership of his snowblower to his daughter? Any of these would be enough to definitively make it "not perjury," at least one of them would not be enough to avoid "defamation," but the charge of "lying" isn't so easily or cleanly evaded.
I'm not familiar enough with the state of the field to know whether this is indeed the dogma of the field, let alone to know if it's justified. But I do know enough to know that a lot of good work has been done on this project and you don't need to re-invent all of it yourselves if you don't want to.
>Any of these would be enough to definitively make it "not perjury,"
I doubt this.
There's a reason why witnesses need to testify to "the whole truth and nothing but the truth".
Check the math here.
"8% of Americans said they had a relative who died from the COVID vaccine."
What percent of Americans have a relative die each year of any cause?
According to this... ~4%!
https://www.therecoveryvillage.com/mental-health/grief/grief-statistics/
So if we stretch out the time frame to be the past two years we get 8% implying that every single person who died died of the vaccine! We could have stoppered death forever!
Okay but this also makes it implausible that they just incorrectly attributed post-vaccine deaths to the vaccine - there weren't enough post-vaccine deaths!
Exactly. I wanted to rule out that hypothesis. With each discarded hypothesis, the plot thickens.
Yes, 8% doesn't seem crazy, and we can always ask how many people were in the pole.
How many relatives do you have?
If your cousin tells you on New Year's Day that her cousin told her on Christmas that her cousin's cousin told her on Thanksgiving that her cousin's cousin's cousin died from the COVID vaccine, how many relatives are we talking about?
Putting some numbers on Steve Sailor's complaint:
As a rule of thumb, 1% of Americans die each year. (That' why our life expectancy is ~100).
If each of them has 4 relatives, then we get 4% of the population having a dead relative. That's probably about right if you count children, siblings, and perhaps(?) grandchildren, but nothing else.
As soon as you count uncles and cousins and perhaps even more, the number will be a lot higher. Especially because people who are old now often have/had a lots of siblings.
People very often believe false things after they read news stories. This is by design. Naive readers think that news stories are credible without having a good understanding of what it means to be credible, and willfully ignorant readers don't care if they believe false things. The act of "staying informed" has positive emotional valence for these people, but they would believe fewer false things if they stopped reading the news.
Many people are angry about the status quo, and they're trying to convince everyone that the news is not credible, whatever that means. They want "staying informed" to have negative emotional valence. They interpret this series of blog posts as a defense of the status quo, which should not be defended. They don't believe that careful caveats about nitpicky technical linguistic questions can change the directionally incorrect impact the blog posts will have on emotional valence.
This is a bravery debate. Scott is technically correct about everything, but nobody cares about that. They care about emotional valence and downstream consequences. They care about how common it is to believe false things.
(This is just one perspective, and it's oversimplified, but I think people need to be more aware of this dynamic.)
You just explained my emotional response to these posts that I've been having trouble expressing. "Lying" is worse than "not lying," and so a blog post centered around the idea that "the media rarely lies" feels like an undeserved defense of a frequently deceptive media, regardless of what Scott's intentions actually were. And his defensive follow-up posts sound a lot like "what? I said that I didn't come to praise Caesar and that Brutus is an honorable man, I don't understand what you're complaining about."
I agree with the broader point Scott is trying to make, and I can't really find technical fault with anything he wrote, but it's framed in a way that seems to be making excuses for dishonest people. Even if he's not, really. He's still providing a level of respect and decorum that I don't think the media really deserves (they may not be 100% forthright all the time, but at least they're not liars). I know I'm probably being unfair to Scott, but his continued insistence that "the media very rarely lies" isn't a terrible headline almost guaranteed to be misconstrued is frustrating. If you're going to use a misleading clickbait headline for your article, then you don't get to complain when people ignore your buried lede and spend their time reacting to the clickbait headline. You reap what you sow.
I think one of the problems with this discussion-- which I'm sure came up in the comments, but I haven't looked through those-- is that the title used the word 'rarely', which means that by definition Scott and the people in the comments are going to be talking past each other when people in the comments bring up examples of what they think are 'lies'.
> In a perfect world, we would have separate words for all of these.
Let it never be said I didn't make the world a bit more perfect. ;) Let's give 'em names:
1. Reasoning
2. Erring
3. Flubbing
4. Rationalizing
5. Spinning
6. Deceiving
7. Lying
(Multiple commenters also pointed out the variation commonly called bullshitting.)
Hmm, no. In my mind there's maybe a relation to simulacra levels. https://thezvi.wordpress.com/2020/06/15/simulacra-and-covid-19/ But I'm not sure if 7 level lying is on that scale. It's a multi-axis world.
Yeah, I think most of your readers have agreed with you all along. Your critics are what make you stronger. Anyway, I want to carve out a category:
2.5) Reasons well but is dumb. Dumb in this use means uninformed. I don't know the true facts. That would be my category. Oh and after dumb, whatever bias I have takes over.
I've stopped consuming the news. People at work will tell me if something important happens. And I just focus on local stuff, work, my kids, the house, cars, neighbors, friends...Oh and the neighborhood here online at ACX. It's great, thanks. (Oh I follow weather and local sports on the news, and getting politics into sports (and weather) is something I hate.)
I have found the reactions to this series of posts bewildering. You're stating an extremely simple premise, and for some reason you have comments full of people attacking...something else. I would expect readers of this blog to share a level of clarity of thinking and basic literacy, and yet you have numerous comments throwing out counterexamples that are unrelated to your central point.
I'm not trying to say people commenting are dumb and wrong, because that seems unlikely. But reading these comments I genuinely don't understand what is going on in the people's heads. (I'm thinking specifically of comments that are like "oh yeah but look the media once reported that person X said Y and this wasn't true so the media is lying". Or even "You're saying the NYT is as bad as info-wards.) Again, non-judgemental, I'm just clearly missing something about how people's brains work that is causing this behavior.
I think most of the people commenting in that vein go into this discussion with very firm beliefs that they have caught the media lying, and not infrequently so. I therefore would just call the phenomenon you observe "update resistance", which is how people's brains work most of the time, as I learned from reading ACX and SSC.
Of course one must be different to not blanketly chalk any disagreement up to update resistance - some of the cited complains do convince me to some extent that Scott's definition is so pedantic that it's borderline useless, many others however don't seem to be convincing at all because they sometimes even inadvertently prove Scott's point (see my comment on the Alex Jones thing). So I definitely see your point.
I think this is definitely part of it. What confuses me most though is the fact that the counterexamples don't pass the bar Scott established. Maybe they already classified those things are lies by their own definition, and didn't recompute the label with Scott's when they posted them? I think that could explain some of it.
Unrelated, I think Scott's is sort of my default day-to-day definition of lying--it would never occur to me to call any of these other things lies. So whether or not its useful, I would say it's not uncommon in my experience.
He's not directly saying the NYT is as bad as Infowars, but he's putting them in the same category for reasons that are not really clear. And people are reasonably upset about the implication and suspecting he has an axe to grind.
(Like, this article is literally about how someone with an agenda can make technically true but misleading statements that imply your desired conclusion, are you really shocked that people will be looking at what the article implies rather than just what it literally says?)
"He's not directly saying the NYT is as bad as Infowars, but he's putting them in the same category for reasons that are not really clear. And people are reasonably upset about the implication and suspecting he has an axe to grind."
I don't think that's reasonable! This seems like a decoupling thing, right? It's fine if they're in the same category or share traits or whatever, the same way that it's fine if you talk about the fact that Hitler loved his dog. It doesn't mean you like or support Hitler, it's just a true thing about the world. I'd expect people who don't think-about-thinking much to get upset if you say nice things about Hitler, but I wouldn't have expected that from this blog's readers.
If someone was specifically trying to draw attention to how both Adolf Hitler and Joe Random own dogs, I think it would be reasonable to suspect they have a problem with Mr. Random. Sure, there's nothing wrong with owning a dog, but you still have to wonder why they think it's important for us to draw a comparison between Mr. Random and Hitler.
(The fancy name for this is Gricean implication - bringing something up in a conversation inherently implies it's relevant.)
In this case, the reason Scott said he's putting Infowars and the NYT was to argue that there was no principled way to distinguish "the sort of lying the NYT does" and "the sort of lying Alex Jones does." But many people (including me) disagree. Even without a 100% objective measure of the truth there are many different axes on which you can differentiate those types of lies, and saying that they're both the same type of lying is ignoring those distinctions.
I think this is a good point, and I agree with the first two paragraphs (though not your third). But what is weird to me that people argued against it so _badly_, like bringing up examples that obviously played into Scott's hands.
What I'm settling on as the process is this:
* People read the thing
* The social implications of the thing but not the object level thing make them uncomfortable
* They feel they need to provide counterarguments, but aren't exactly in touch with what is prompting this, so they come up with arguments that are just sort of a mess
This is not to say that's 100% of what happened, but it does explain some of the more bizarre counterexamples.
My prior on Scott having an axe to grind with the NYT is not exactly zero, but I don't necessarily think it's fair to suspect this based on the post in question. What he's doing is perfectly in line with his philosophy from even the SSC era, before the NYT thing.
Most people have a gut reaction along the lines of 'Clearly, Infowars is waaaay worse than the NYT', but gut reaction isn't good enough from a rational standpoint, so let's analyze why we think that is - *without* relying on the conclusion that Infowars is made by despicable humans and only read by deplorables.
(My memory is really bad so I don't remember the title of the SCC post, but there's this one about 'other people [not] getting "the thing", where "the thing" is some understanding that the person of the opposing faction is willing to fight just as hard and is just as convinced about their positions as you are about yours' - possibly butchered it. But I think that's where he's coming from when he consciously de-emphasizes judging the messaging/narrative of the respective publication.)
And then we see that, yes, maybe NYT has better fact-checking, yes, maybe NYT has a higher bar of who they are willing to regard as a 'credible source' (didn't work too well in the Russian collusion story, but that was not just NYT). But most of this is quantitative - we'll have a hard time making a qualitative distinction in the sense of "Infowars does X (which is egregious), and NYT has never done and would never do X".
Some people may say that the quantitative difference is so large that it becomes qualitative. Others might say that the distinction comes from the NYT having good values and Infowars having obviously bad values. These distinctions may work for you, but from my perspective it's fully in-character for Scott to dismiss them as too arbitrary, I don't see how this is picking on the NYT in particular.
I am also absolutely amazed by the response. I can't believe so many people would like things they believe to be lies by some definition to be censored - what are we, back in the Soviet Union? I also can't believe most people think they can easily tell whether things are true or false by whatever their definitions are.
Not only I wouldn't want a job censoring things under any circumstances, I also wouldn't want a job checking facts for being outright false unless I was paid really, really well, because it's a very difficult job. Most stories of uncovering academic fraud involve crazy amounts of difficult detective work - this stuff is not easy! Same goes for many other kinds of fraud, the kinds you can go to jail for.
So, do you think that there shouldn't be any laws against libel or slander? Because that's a fairly common case of the government punishing things that it believes to be false and harmful to someone, which actually did happen to Alex Jones.
Oh no, these laws are fine. When a case goes to court, it goes through the process of establishing the facts and the intent. The process is public, and the accused party gets to speak to defend itself. This might go wrong, but this is pretty much as fair as it goes.
That's a far cry from "lets just prevent everyone who thinks something we deem to be a lie from publicly speaking about it".
EDIT: The legal process is a non-negligible burden on the accuser. This makes sure that it's only used on rare occasions, rather than every time someone says something the accuser doesn't like.
I've had this relevant Dr. Seuss song playing in my head since morning:
https://www.youtube.com/watch?v=4Cn-r8YfXBk
"With a flick of my thumb, with the greatest of ease, if I wish I can stop all the buzzing of bees." That's exactly what Twitter, Google, etc. were doing - and I would think this would bother almost anyone who gives it a minute of thought, but apparently that's not so.
I have definitely detected an increase in people who think censoring is both necessary to prevent various forms of harm and not particularly difficult to implement. This group usually likes to put square quotes around the term free speech.
What I haven't seen a single incidence of is a person who believes in such a censorship regime that doesn't additionally believe the current holders of cultural power are on their side. I kind of get whiplash between arguments that censorship laws are a net good and that they believe their opponents are malicious authoritarians in hiding.
They say in rationalist culture there are 40 words for being wrong
+1 for the sensible chuckle
Okay, but how do you get from the extremely narrow and pedantic point that deception rarely involves literally inventing facts whole-cloth to the point you were originally trying to make about censorship?
That was the original reason you argued that this distinction matters, at the start of this sequence - you said that "people" (unspecified) want to censor Infowars but there's no way to do that in a principled way without also censoring the NYT for doing the same sort of thing.
But the legal system *already* knows how to make these distinctions. Judges know and apply all sorts of nitpicky categories like "reckless disregard for the truth" or "matters of public concern" to identify exactly how much slack we want to cut someone for telling a lie that isn't an outright fabrication. And this system has demonstrably managed to punish Alex Jones for his most egregious type-6 lies while not punishing the more minor lies of the NYT.
People don't use these sorts of distinctions in everyday speech, sure, but when it comes to actual potential censorship by the government these distinctions are already well-known. So what exactly are you arguing should be different?
The US government isn't ever going to be doing this type of censorship directly due to the 1st Amendment. The legal system uses a non-criminal civil process to potentially punish a single instance, it doesn't impose lifetime bans of speaking. The censorship regimes at social media companies do their work in hiding with little transparency. There is a significant burden on those who wish to prove a lie legally so the process is not used frivolously. The FBI sends private emails to "trust and safety" personnel. The legal system is accountable through the voters.
The vast majority of things censorship supporters want to ban will not pass legal muster used in the current court system. Perhaps this is the point you are trying to make but the counterpoint is that this is not the type of process these regimes would support because it doesn't fulfill their goals. Thus. back to the original point of you aren't going to be able to articulate a neutral law that achieves your goals of silencing the wrong-thinkers in practice without figuratively sweeping up both InfoWars and the NYT.
> I asked “Why? A woman said she had been raped, and Rolling Stone believed her. The woman was making it up, but Rolling Stone wasn't” and Deepa commented “Isn't it the job of a reporter to investigate? And be good at it?” I don’t want to pick on Deepa, but this is what happens when you have an overly expansive definition of “lie”!
Doesn't this open you up to manipulation by bad actors? What if I tell you that I have a 100% guaranteed method to make you money. Many people are suckered in. I collect fees from them and become quite wealthy. And it turns out its all a scam. They call me a liar and I say "Well, my friend Bob told me about it. So I wasn't LYING technically." If lying is only saying something which the speaker knows to be untrue then the concept effectively does not exist.
Fwiw, in defamation cases and the like you have to prove you did sufficient investigation and all that. Which Rolling Stone did not. Which is why they settled. They were going to lose badly.
I think your example might be unfair, because in that example the scammer has independent reasons to know he's running a scam, other than Bob's story. He's running the scam! Collecting money and getting rich.
A fairer example would be if Bob is running the scam, and you're not part of it, but you hear from Bob that it's a fabulous get rich quick scheme, totally legit. You pass the story onto a friend of yours, enthusiastically, and your friend invests with Bob and loses everything. The question then becomes: can your friend sue you for fraud in the same way he can sue Bob? Shouldn't you have done due diligence[1] before passing on the recommendation? Aren't you just as much a liar as Bob himself? I suspect most people would have at least a little problem with this.
---------------------
[1] I mean, you absolutely should have, morally and legally, if you're acting as your friend's financial counselor, and he pays you a fee for that purpose. But a newspaper is not in a fiduciary relationship with its readership. Nobody has created a contract, explicitly or even implicitly, in which you pay money for a product of a certain level of guaranteed accuracy. The news is sold by newspapers and media on an "as-is" caveat emptor basis. It's up to you to read the prospectus carefully before investing.
No, no Carl. That's what happens when you have an overly expansive definition of lie! You see, yes it was untrue. And yes I was benefitting from it. And yes I could have discovered it if I'd done a little investigation. But I wasn't making it up. Bob was. I was just credulously repeating what Bob said in good faith.
Your example is a good example of the same phenomenon. But I was purposefully pushing it to an absurd point with mine.
As to the whole newspapers are caveat emptor: They're not actually. The story did meet the legal definition of defamation. Yes, truth is a defense to defamation and untruth is not automatically defamation and the law is not the exact remit of morality. But it doesn't change that Scott's definition here is so strict he's required to believe many convicted fraudsters never lied.
Yes and no. Defamation is not unique to newspapers, a blog commenter can be guilty of defamation as easily as the New York Times. Furthermore, I don't think there is any implicit contract between a newspaper and a reader *other than the defamed person* that the newspaper will not defame anybody.
Indeed, if anything the existence of tabloid journalism (or InfoWars for that matter) shows that the public actually likes and wants newspapers that come as close to defaming people[1] as is legally possible.
I'm comfortable with concluding some fraudsters are not liars. There are many ways to mislead people other than telling outright lies. That's why we have a separate word "fraud" instead of just merging "fraudster" and "liar" because they are in one-to-one congruence.
But if we go the other way, if we expand "lying" to include a host of much more minor human foibles that are intended to influence -- perhaps even influence a bit beyond what objective fact strictly merits -- we are going to make the word as useless as the word "racism" has become[2].
We need strong, toxic words, words that indicate stuff that is so far beyond the pale nobody will defend it, and we all recoil from it. Their existence, and occasional use, serves to set clear boundaries on civilized behavior that stiffen the spines of those otherwise too weak-minded to resist momentary temptation. ("Do this and people will call you a liar. It's not worth it.")
But to preserve those clear markers we also need to be stingy in their application. They really need to be applied where even very generous or hihgly biased souls would (perhaps reluctantly) mostly agree. For example, we should orobably only call "lies" the things Donald Trump or Joe Biden say that most people who voted for Trump/Biden would agree are "lies," so we can achieve a 80-90% social consensus on who is a liar and who is...something else, maybe something undesirable, but not over the line with the Nazis and molesters.
Preserving the social consensus around what a toxic word means is far more important, in the long run, than winning any momentary social battle. Or rather, failing to preserve the social consensus means winning today's battle for your faction but losing the long-term war to preserve general social cohesion. A poor tradeoff for which our great-grandchildren will hate us.
------------
[1] Although by "people" here I obviously mean the half-human half-orc out-group bastards who are sapping and impurifying our precious bodily fluids.
[2] Not to mention it will kill salesmanship, politicking, and most Internet commentary dead. We'd all be reduced to the bland anodynes of the Camazotzian press, approving mildly to strongly of the latest revised Five Year Plan.
> Defamation is not unique to newspapers, a blog commenter can be guilty of defamation as easily as the New York Times. Furthermore, I don't think there is any implicit contract between a newspaper and a reader *other than the defamed person* that the newspaper will not defame anybody.
I don't see the relevance. Journalists are not a special class. They have no more rights than a random citizen. So the standards put on the general public are the standards we expect from journalists as well. There's no implicit contract. There's instead an explicit rule about what anyone, including a newspaper, is allowed to do. And unless you're going to argue we should expect criminality then we should expect them to stay on the right side of that line.
> Indeed, if anything the existence of tabloid journalism (or InfoWars for that matter) shows that the public actually likes and wants newspapers that come as close to defaming people[1] as is legally possible.
And the existence of the Rolling Stone case and Sandy Hook case show that the public actually likes people who commit actual defamation. I again don't see the relevance. People like defaming people is a true statement. But so is that people like stealing.
> We need strong, toxic words, words that indicate stuff that is so far beyond the pale nobody will defend it, and we all recoil from it. Their existence, and occasional use, serves to set clear boundaries on civilized behavior that stiffen the spines of those otherwise too weak-minded to resist momentary temptation. ("Do this and people will call you a liar. It's not worth it.")
Yes, I agree. In fact I think we should go beyond having the words be toxic. They should be legally enforced through things like defamation or fraud statutes. And Rolling Stone crossed that line. Scott is arguing, however, that this isn't a case of them being a liar. He's implicitly setting up a definition of liar that is more strict than defamation statutes. That's what I disagree with.
> But to preserve those clear markers we also need to be stingy in their application. They really need to be applied where even very generous or hihgly biased souls would (perhaps reluctantly) mostly agree. For example, we should orobably only call "lies" the things Donald Trump or Joe Biden say that most people who voted for Trump/Biden would agree are "lies," so we can achieve a 80-90% social consensus on who is a liar and who is...something else, maybe something undesirable, but not over the line with the Nazis and molesters.
I agree we need to be parsimonious with their application. But this is neither the legal standard or the practical standard as actually applied.
> Preserving the social consensus around what a toxic word means is far more important, in the long run, than winning any momentary social battle. Or rather, failing to preserve the social consensus means winning today's battle for your faction but losing the long-term war to preserve general social cohesion. A poor tradeoff for which our great-grandchildren will hate us.
And yet very common, unfortunately.
Are you sure you actually disagree with me?
https://youtu.be/ohDB5gbtaEQ
No, I'm not sure. I'm sure I disagree with Scott's point that the thing didn't rise to the point of being a liar. I'm not sure if you agree with him or me though.
another post (or series of posts really) for the record books. thanks.
I agree with the central thesis, and should probably be a bit more careful about the use of the word "lie". That said, is the "(make|leaves) other people confused" in 5/6 not overly specific? Did you mean, rather, to say something like "intended to achieve a specific goal other than relaying useful information about the world" (implant an idea, attract readers, sabotage reasonable discussion, etc.)?
You keep ignoring the difference between "a lie" and "a liar".
I believe that the main reason your posts seem so off to many people is simply that: judging the sincerity and the integrity of any given report in isolation is indeed a tricky and delicate business, as you so articulately demonstrate. I agree with you completely on this. But judging the sincerity and the integrity of a given news-source is much more robust, and orders of magnitude simpler. And it's the later that is relevant to your original point, about fighting fake news.
There's a standard argument in cognitive therapy (at least that's where I know it from) against using terms like "liar": they are either overly strict or arbitrary labels.
I'm sure you have lied before, even in the narrow sense of claiming something that you knew to be false (no offense). Does this make you a liar? No, because you don't do it all the time, possibly even rarely. But what is the frequency at which somebody has to lie that it's justifiable to call him a liar? Once a month, week, day? If more than 50% of your statements are lies? So you're good if you truthfully comment on the weather between two lies? Statements of importance? Important to whom?
Maybe you want to use the "I know it when I see it" escape hatch, but this seems pointless in an argument with someone who very apparently sees the same things and comes to a different conclusion.
Your argument seems to me so general to render it useless (is anyone really tall?).
But anyway, we can and should avoid these problems altogether: we're talking about news-sources, not private people. Roughly estimating the trustworthiness of news agencies and labeling those below some threshold as "fake-news sources" does not suffer from all the psychological, sociological and ethical issues that may arise when labeling specific people as "liars".
It does suffer from being easily subvertable by publishing more true but irrelevant (relative to the goal that is likely being pursued by pushing fake news, they can still be globally relevant) articles if you define the threshold in relative terms. And absolute terms don't really make sense either. One would have to measure it as a fraction of volume of information conveyed, where information that is exclusive/reported by fewer other outlets is weighted higher (this would also mean a completely fabricated story would be given a very high weight).
It gets very technical very quickly, and maybe "I know it when I see it" sometimes just is the best we can do. But I would also say that this is decidedly different from saying that someone is tall, because height in an upright position w/o non-body-accessories like shoes can be easily and objectively measured and compared. Nobody is ever absolutely tall (compared to, idk, an elephant), but if you know someone's height, you can tell their percentile within a certain population of humans and make a well-founded claim that they're tall relative to that population. This isn't possible with "trustworthiness" of media outlets.
I don't claim that "I know it when I see it". I straight out claim that it's a tractable problem to formalize and solve. I think that the objections you give and the difficulties you point out are valid, but can all be satisfactorily addressed.
How? We may start by quoting Scott himself: "P(A|B) = [P(A)*P(B|A)]/P(B), all the rest is commentary". But I'm aware that it's not obvious and there's a lot to be said. However I'd like to avoid delving into this (technical, as you rightly say) discussion right now since it's worth doing so only after there's an agreement over the following points: (1) the epistemological issues that worry Scott looks very different on the level of "news-items" and the level of "news-agencies", and (2) it's the later that practically matters, and it's an easier problem to solve.
To clarify: it's conceivable that the problem still remains difficult, but I claim it's clearly much easier, and that most of Scott's arguments are at the very least dramatically weakened when applied to it, if they applicable at all.
For anyone who has published and has accused someone of lying, the consequences, if wrong, can be disastrous. Some people care about the consequences more than others. “They should have known” is not a proper defense most of the time. In general, both writing and research with the intent to publish are painstakingly tedious, so even the best of journalists can err in way that elicits others to call them a liar. The best of us should accustom ourselves, out of habit, to reserve this accusation in public as much as possible. Better to say it is misleading or deceptive whenever practicable? Or if we're in the accusatory mood, we can say someone is trying to mislead/deceive? Doing this can keep the focus on debate rather than defending why something is not a lie.
Furthermore, everyone believes things that are false, and for most of us, a magical revelation of some of those misbeliefs, if we magically accepted them, would be a shock. Many of us want to know if we’re wrong even if it will ruin our day/week/life, but many others don’t want to know, at least on a subconscious level. I wonder if any correlation can be made between people who want to avoid the truth, intentionally or instinctively, and those who want to “wirehead” themselves to a perpetual happiness device. I wouldn’t accuse anyone of fitting this mold, but it would be an interesting study.
> Looking at these people's other responses, they did not show a consistent tendencies to make things up or say outrageous things (except for one who listed their religion as “Satanist”).
This is a legitimate, if niche, thing. There's a Church of Satan which has real religious beliefs, and there's an unrelated Satanic Temple which is atheist, but for all I know some of its members call themselves Satanists.
Paul makes a good point about proving too much. If you define "lie" in such a way that "The National Enquirer Rarely Lies" is a true statement, then who the hell cares. Ultimately, people are interested in how often NYT or Infowars does "That Thing The National Enquirer Does" when writing a story, whether or not you want to call it lying.
This problem would be reduced if the news media had a convention of annotating each headline to indicate whether the editors think the proposition is probably true (PT) or probably false (PF)?
'BIGFOOT FOUND' (PF)
And for question titles, probably (P) and probably not (PN) instead:
'WILL BIGFOOT BE FOUND?' (PN)
'SOURCE CLAIMS BIGFOOT FOUND' (PT) defeats the purpose. So the convention would only be meaningful if it was strictly applied to the inner proposition.
However, I expect that this annotation would reduce readership (by making articles less interesting to read), and that it will therefore never happen.
Scott, I feel like the big issue here is that you're conflating two different things - both of which can be described as "technically true, but misleading/deceptive," but which are nonetheless very different in both degree and kind.
The first category includes your example of FOX News saying "more Whites are killed by police than Blacks" without mentioning that Blacks are killed at a proportionally higher rate. It also includes your example of the New York Times saying "Blacks are killed by police at a higher rate than Whites" without mentioning that police are called to Black neighborhoods much more often. In both cases, the fact is true, and arguably even genuinely useful to know, but there's some critical piece of context missing which leads people to draw flawed conclusions. This isn't quite lying, exactly, but it's not telling the whole truth either. And if the missing context was purposefully left out in a deliberate effort to make people draw false conclusions, then I'd consider it a lie of omission. On your 1-7 scale, these would be ranked 5 or 6.
The second category is when someone just tells a straight-up bald faced lie, but argues that it's "true" in some abstract sense through some opaque trick of semantics. On your 1-7 scale, these would still be ranked 7, or at least 6.9999999! But I think you're making the mistake of grouping them together with the 5's and 6's. Some examples of what I mean:
-Redefining words to mean something entirely different from the common usage without letting anyone know that you're a nonstandard definition
-Taking advantage of homonyms to substitute a word for a different but identical-sounding one (e.g. if your wife asks you to deposit a check for her, and you say "I promise I'll stop by the bank today," but then go fishing at the riverbank instead)
-Speaking in metaphorical terms, despite all context making your statement sound literal (one famous example is Obi-Wan saying "Darth Vader killed your father")
-The practice of mental reservation (i.e. saying "I don't have any money" out loud while mentally appending "in the Bank of Zimbabwe" to the statement, thus giving the version of the statement in your head a positive truth value)
-Hearing someone make a statement that you know to be false, then sharing it with other people without letting them know that it's not true (for instance, if Bob says "I saw Frank's girlfriend Martha making out with Sam at lunch today," and you know for a fact that Martha was alone studying in the library during lunch period, but you still go around telling people "Bob said he saw Martha making out with Sam" and letting them assume it's probably true)
To even call these "lies of omission" would be overly generous. They're just outright, blatant, unambiguous lies, simply worded in such a way that the speaker can retroactively *claim* (however implausibly) that they were being honest. And InfoWars, unlike the New York Times, frequently resorts to these sorts of tricks. ("I didn't say "[false statement]," I just said "X said [false statement]," that's totally different" is a pretty common defense of theirs.)
>Redefining words to mean something entirely different from the common usage without letting anyone know that you're a nonstandard definition
I've been seeing this a lot lately. I believe it was the New York Times that was stating this week that we never had any lockdowns in the US, simply "intermittent state by state mitigation measures" or some other such euphemism. Oh goody, I'll have to go inform the proprietors of the restaurants that were forced out of business in my town in 2020 that we never had any lockdowns, I'm sure they'll be thrilled.
I think Scott's "drug testing" example is a version of this.
Nobody beyond the lizardman constant, and Scott, would claim that being asked if you use drugs is a "drug test".
When was the like button removed? Or am just not seeing it?
Anyways, thanks Scott for discussing that at length with anyone. That's exactly how I feel about the phrase "lie" but would have given up arguing by now.
Just going to add that I was quoted in an article in The Times of London (on the occasion of exam results day) as saying something like, "Anyone who says marking standards are falling is belittling our achievements". Didn't say it! But it did support the narrative of the article.
From Zack's comment:
"I'm skeptical that 10 out of 500 people were unfortunate enough to have 2 household members die: one from COVID and one from the vaccine. (Especially because these are not large households; 4 of these 10 report that they have 1 other household member [...])"
Those 4 are obviously protagonists/narrators in hackneyed "they were dead all along"-twist horror stories.
I think looking at some indisputable cases of #7 highlights how rare they are:
o) Jayson Blair inventing stories and sources: https://en.wikipedia.org/wiki/Jayson_Blair#Plagiarism_and_fabrication_scandal
o) Another guy doing the same thing: https://www.poynter.org/ethics-trust/2022/usa-today-fabrication-breaking-news-reporter/
and I think it's generally individual reporters doing this freelance, so to speak. I can't think of any fabrication scandals where editors issued an order to make things up. Maybe there were some back in the "yellow press" days?
Could it be a helpful contrast to say that Moscow Kreml (or indeed any nation at war, Kiev as well but less in-your-face) lies in the point 7 sense continuously 24/7? It's part of the war effort to spread misinformation.
Most media are NOT participants of an actual physical war and do NOT deliberately spread lies, though they still serve interests that compel them to do all things in points 2 through 6. Contrasted with a clear example of point 7, and thus a bit detached from the U.S. bi-partisan echo-chamber circle-wankery, perhaps the claim would sound a lot less controversial.
Harry Frankfurt has a famous essay on bullshit:
http://www2.csudh.edu/ccauthen/576f12/frankfurt__harry_-_on_bullshit.pdf
He distinguishes it from lying, defining it as "speech intended to persuade without regard for truth". A key summary from the wikipedia article:
>Both people who are lying and people who are telling the truth are focused on the truth. The liar wants to steer people away from discovering the truth and the person telling the truth wants to present the truth. The bullshitter differs from both liars and people presenting the truth with their disregard of the truth.
Now, you might say that bullshitting is just a variant of "egregious disregard for the truth", so it's already been analysed above. But I think that we rationalists have a bit of a blind spot where bullshit is concerned; we're so concerned with the truth that we assume that others are equally so concerned (interestingly, many conspiracy theorists are similarly very concerned about the truth). It didn't appear in Scott's list in section 8 (point 4 was the closest, but "more-or-less subconsciously not trying to reason well" doesn't really cover "entirely unconcerned with reasoning well, and not seeing why reasoning well would even be an issue").
So I think we want to round bullshit to something more understandable. Alex Jones, for example, is mainly a bullshitter. The truth or not of his published statements is not a concern of his. Is he lying? As Scott pointed out, he's not. Is he telling the truth as he sees it? No, he's not doing that either. So I think Scott is reasoning "he's not lying" while others are reasoning "he's not telling the truth" and then get into definition arguments. But if we add the third category of bullshit, then it seems easier to classify Alex Jones.
So, armed with bullshit as a category, what can we say about the other media sources? Weekly World News? Bullshit. New York Times? Going off the descriptions in this article and the original one, not bullshitting (or rarely doing so) - we might need to add another category for what they're doing, but it's not bullshitting. Infowars in general? A high proportion of bullshit, though not all of it.
Anyway, just wanted to remind people that some people really don't care about the truth, one way or the other. They're not lying, they literally don't care about the truth, and probably have difficulty imagining that other people do.
Thanks for posting this. It's a little hyperbolic in it's conclusion but I think generally true.
I'd also like to thank you for posting it.
One might think that you are expending an inordinate amount of time and energy on an undeserving subject, but not me. The actual subject of this investigation is:
THE NATURE OF TRUTH.
On issue 3 and media believing their own lies, among many obvious differences between Infowars and NYT I see one that is particularly relevant.
Imagine that journalist Alice works for a newspaper and covers space exploration. She has opinions on the matter, reads many sources, does her best to decide which stories are true and which are not, and honestly tries to inform her readers as well as possible. There are people out there who disagree with Alice on some topics, sometimes pretty emotionally, but these people do not control most of her sources and do not have the means to manipulate her. When Alice finds a story about Saturn rings of some given relevance and novelty to her readers, Alice would try to establish the degree of truth in this story and will not publish it unless she was 95% confident in it.
Another journalist, Bob, works for another news outlet, covering domestic politics, NSA surveillance, and conspiracy theories. He has the same motivations as Alice and also does his best to decide which stories are true and which are not and tries to honestly inform his readers. There are more people out there who disagree with Bob. However, for some weird reason, Bob honestly believes some of the outlandish conspiracy theories, such that there is a secret email list of journalists that covertly coordinates domestic policy coverage in main news outlets to favour one party, or that Twitter is running a shadow banning program against people who express beliefs similar to Bob's, or that most of media fact checker outlets are coordinated and funded by the same people who fund one of the political parties.
When Bob sees a story about a certain conspiracy theory of the same novelty and relevance to his readers as was Saturn rings story to Alice' readers, he would do his best to establish degree of truth of this story and would not publish it unless this degree was rather high. Would it be rational for Bob to set this threshold at the same 95% as Alice? I do not think so. Bob honestly believes the conspiracy theories about Twitter and factcheckers, so he thinks that his and his readers news flows are heavily biased against this kind of stories. A lot of good and true stories simply do not propagate well enough to reach Bob and his readers because of suppression. So the stories that do get to Bob are more valuable, and Bob would set his threshold at 80% and not at 95%. This does not mean that Bob is less honest of more gullible than Alice. He just believes that he lives in a world with powerful forces that manipulate the news flows and sees his duty to try to compensate.
Nicely argued. I especially like your last series of comments. One problem you hint at but don't really answer is what are we supposed to say/do about the many instances when grossly deceptive information is disseminated without actually crossing the threshold of genuine lying?
There is a huge ingroup/outgroup component to this. When your side is just wrong, you can safely ignore it for a while, then come out with retractions/corrections if you want (e.g. NYT on Russiagate, Hunter Biden laptop, etc). When the other side is a member of your outgroup, it's all "lies" down the line, even if the "lies" are mistaken yet honestly help beliefs. The reasoning here is pretty simple, calling someone a "liar" is a moral attack that us useful to deploy on your enemies.
Contrast Alex Jones to a group that says lots of false things all the time but we don't call a bunch of liars: flat-earthers. For a variety of reasons that may include a mix of genuine belief / trolling / performance, they make even more unlikely claim that the Earth is flat. I highly doubt that the NYT will be condemning them as a bunch of liars anytime soon. Nor do we condemn professional wrestlers, or members of small religious movements even if we think their beliefs are mistaken. You have to be an outgroup member to deserve this treatment.
Also, Jones never said he lied as far as I can tell. Rather, he said that the shooting actually happened which the media ran with as "Alex Jones admits he lied". But if you look at what he actually said at the time of the shooting, it was mostly just his conspiratorial take on the situation that things look fishy. The NYT helpfully compiled a list of Alex Jones "lies" around Sandy Hook. It is here:
https://www.nytimes.com/2022/09/22/us/politics/heres-what-jones-has-said-about-sandy-hook.html
Ironically, if he has said at the time "I think the shootings definitely happened, nothing to see here" , then *that* would have been a straight-up lie as he clearly did not think that. This just highlights the alchemy of converting "my opponent is wrong about something" to "my opponent is a liar"
What about media that are propaganda amplifiers, like Russia Today? Sure, they still prefer 6 to 7 (if only it's generally considered to be more effective), but when the same source claims four contradictory versions in two-to-three days (I think I'm not even exaggerating the time - thinking of the Bucha disaster) with the only thing in common "we did nothing wrong there", one has to wonder whether this is even possible (quoting gwern: "“Possibility” in ordinary language doesn’t include astronomically small odds.") to do on pure 6, without 7.
It is so nice to see discussions of moral theology on here re: "what constitutes a lie? is lying always wrong? is silence, equivocation, or mental reservation the same as lying?" 😁
https://ndpr.nd.edu/reviews/lying-and-christian-ethics/
"In a country with 100 million conservatives, is it really that hard to find a handful of them capable of writing news articles?"
Regarding the liberal bias in newsrooms, this is an old story. Even the media itself (at least back several years) recognised this. *Why* it happens is something that is much debated.
The Public Editor of the New York Times wrote about this in 2016:
https://www.nytimes.com/2016/07/24/public-editor/liz-spayd-the-new-york-times-public-editor.html
And in 2004:
https://www.nytimes.com/2004/07/25/opinion/the-public-editor-is-the-new-york-times-a-liberal-newspaper.html
They ran a guest opinion piece in 2015:
https://www.nytimes.com/roomfordebate/2015/11/11/why-has-trust-in-the-news-media-declined/liberal-news-media-bias-has-a-serious-effect
A former editor in a talk in 2011 admitted they had a liberal, urban lean but denied this slanted political coverage:
https://www.tmatt.net/columns/2011/10/god-and-the-new-york-times-once-again
The Washington Post dipped its toes into controversy in 2013;
https://www.washingtonpost.com/opinions/patrick-pexton-is-the-post-getting-rid-of-the-ombudsman/2013/02/15/fff68282-778f-11e2-8f84-3e4b513b1a13_story.html
And Rod Dreher gave his take on it:
https://www.theamericanconservative.com/wapo-error-has-no-rights/#post-comments
"Over the years, talking to fellow conservatives about media bias, it has usually been my place, as one who worked in mainstream media, to tell conservatives that they’re wrong in some significant way about media bias — not its existence, but the way it works. Most reporters and editors, in my 20 years of experience, do not set out to slant stories, and in fact try to be fair. The bias that creeps into their coverage is typically the result of a newsroom monoculture, in which they don’t see the bias because everybody, or nearly everybody, within that culture agrees on so much. In the case of gay rights and the marriage debate, though, they don’t even make an effort to be fair. I have heard some version of the “error has no rights” claim for years now. They honestly believe they are morally absolved from having to treat the views of about half the country with basic fairness in reporting. And they are shocked — believe me, they really are — that these people view them and the work they do with suspicion, even contempt."
So where are all the conservative reporters? Probably working for newspapers you never heard of because they're local or regional. Why aren't they getting jobs in the NYT or the Post? It's that thing about "cultural fit" in interviews - sorry but we just don't think you'd fit in to the company culture here.
Remember back in 2020 the staff protesting they felt unsafe because the NYT ran an op-ed? Now just imagine the protesting about expecting them to work side-by-side, in the same room with, a bigot and reactionary!
https://www.niemanlab.org/2020/06/this-puts-black-people-in-danger-new-york-times-staffers-band-together-to-protest-tom-cottons-anti-protest-editorial/
The curious thing is that this is a relatively recent development. The left-right division in the media used to be roughly similar to that in the general population. For example, take a look at this fascinating graph of newspaper endorsements for President versus the popular vote:
https://ceo-na.com/editors-choice/editorial-endorsements-in-the-us-election/
In 1984 the fraction of newspapers endorsing Ronald Reagan for re-election was pretty close to the fraction of the popular vote that Reagan won. But the situation skewed increasingly left after that, so that today newspapers are wildly out of step with the general voting population.
So why has newspaper culture become so weirdly exclusive in the last 50 years?
"So why has newspaper culture become so weirdly exclusive in the last 50 years?"
I think it's a combination of things. American journalism liked to present itself as being 'just the facts' fair, impartial, unbiased reporting unlike European advocacy journalism (where every country has its version of "A is the right-wing paper supporting Party D and B is the left-wing paper supporting Party E"), but that was relatively modern. Good old muck-raking and "follow the proprietor's political views" was plenty common in American papers.
So there was indeed an underlying slant, even if the staff did genuinely think they were being fair and impartial. Secondly, there were big conservative as well as liberal papers, and regional/local papers were as influential in their own markets. The New York Times might like to think of itself as 'the newspaper of record' but an awful lot of people around the country were not reading the Times, they were reading the big regional paper and/or their own local papers.
Thirdly, the liberal reporters became senior and moved up to be editors, as per this opinion piece in the NYT itself from 2015:
https://www.nytimes.com/roomfordebate/2015/11/11/why-has-trust-in-the-news-media-declined/liberal-news-media-bias-has-a-serious-effect
"For instance, Bill Keller, after years as a New York Times columnist taking largely liberal stands, became The Times' executive editor.
Studies consistently show that reporters and editors stand to the left of the American center. More than 30 former journalists now serve in the Obama administration — nothing comparable happens in G.O.P. administrations.
Clustering of left-of-center viewpoints in the newsroom leads to a cloistering, and thus reporters end up unfamiliar with conservative viewpoints. This shows up in the tone of daily coverage (for instance, “property rights” gets put in scare quotes, while “abortion rights” doesn’t)."
This is the former editor Bill Keller who said, in a talk in 2011:
""We're liberal in the sense that ... liberal arts schools are liberal," Keller noted, during a recent dialogue recorded at the Lyndon Baines Johnson Library and Museum. "We're an urban newspaper. ... We write about evolution as a fact. We don't give equal time to Creationism."
Moderator Evan Smith, editor of the Texas Tribune, jokingly shushed his guest and added: "You may not be in the right state for that."
Keller continued: "We are liberal in the sense that we are open-minded, sort of tolerant, urban. Our wedding page includes – and did even before New York had a gay marriage law – included gay unions. So we're liberal in that sense of the word, I guess. Socially liberal."
Asked directly if the Times slants its coverage to favor "Democrats and liberals," he added: "Aside from the liberal values, sort of social values thing that I talked about, no, I don't think that it does."
Next, I do think journalism schools had a lot to do with shaping attitudes, but this is only my own view. Going to college to study journalism, instead of working your way up on a newspaper as in the old days, means exposure to academia, and academia slants left. So you're going to be taught that set of values, and if you remain conservative in your views despite it all, you probably learn to keep your mouth shut and go along to get along.
So there does develop in newsrooms, as per the Keller interview, this culture of "we're all liberal, we have all the right opinions on various topics and are on the right side of history, we have the same social values and beliefs that all good and right-thinking people have". It doesn't mean they think they are reporting in a manner that is *not* fair and impartial, but they don't think to examine their biases anymore than the rest of us think to do so.
And lastly, there has been a push by the new, younger journalists to engage in activist journalism because they really do believe there is only one right side of history and it would be immoral not to fight the good fight, that error has no rights, and there should be no pretence at 'balance' or 'both sides of the story', because all the good right true facts are on one side and all the bad wicked prejudice and bigotry is on the other. So, right back to advocacy journalism of the old days.
I just skipped to your actual thoughts because I get tired of the confirmation bias driven arguments that the press is filled with liars and
once again
I agree.
6 is fuzzy, but you acknowledge that. It's sorta case by case. How egregious is it.
But yeah. This model is good and right, imho.
The National Enquirer story about deliberately >not< checking out their sources' stories resonates for me.
For mainstream, i.e. what people consider "serious", media deliberately not checking: remember Rathergate? Documents (that we now know were fake) presented as real on national news about a political story just before an election. Not authenticated, but Rather said they had been, and continued to say so for weeks (as well as the company backing him).
Now, you can argue that this is rare and therefore memorable.
I say it's common, but most of the time outside people don't go to the effort of actually checking, or else when they check and find it fake they are drowned out by the "new news" or by deliberate suppression. For the few times I have been on the scene for something later reported, I have always found something inaccurate about the report.
[edit]The unusual thing was not that Dan Rather lied. Technically I guess he didn't lie, Mapes lied when she said it had been checked and it had not, while I guess Rather trusted her - although it's interesting that he didn't press her on it when the refutations started coming out and didn't get a second opinion on it.
No the unusual thing was that he got caught at it. He thought he could brazen it out, and it didn't happen.[/edit]
[Further edit]From what I can see (I haven't really followed the multi-year saga) the difference with Infowars seems to be that they got sued fairly quickly. Once they got sued, backing down would have meant losing the case. I guess they also thought they could outlast the people suing them.[end further edit]
I’m a day late, $20 short, but I recall the ‘George Bush amazed by grocery store scanner’ being in a nightly news broadcast. (Probably ABC) where it was presented as he was out of touch with the masses…
YEARS later I encountered the truth in my data-collection tech job. It was brand new scanner tech that weighed and scanned damaged barcodes.
Which ‘lie’ type was this?
Here’s a summary:
https://apnews.com/article/george-hw-bush-north-america-us-news-newspapers-politics-61f29d10e27140b0b108d8e12b64b839
In litigation, the difference between #6 and #7 is the difference between a great lawyer and a terrible lawyer.
Personally I think a very common mode, that is missing from your list is the noble lie and I think a lot of journalists/experts believe they are engaging in the practice. While this is technically congruent with your 7. I think there is a strong association with your phrasing that the person in question is malicious or bad or some such, which doesn't perfectly match up with the noble lie. If you agree with the noble liar it feels weird to call them a liar, but if you disagree that is obviously what they are.
Your definition of lying implies that a person who reasons badly or doesn’t have a good enough model of the world are incapable of lying. This seems like an overstatement.
I encourage people to read the "Nanci Pelosi Hanged at GITMO" article that Scott linked to (https://realrawnews.com/2022/12/nancy-pelosi-hanged-at-gitmo/) thanks to a commenter who wanted to provide an example of actual fake news. I'm sure most would agree that there is no mistaking that particular article for anything but parody fiction (and not particularly good or incisive parody either). I'm not talking about the gist of the content itself (that Pelosi was recently hanged at Trump's order, the way Hillary Clinton was hanged the year before) but the style of narration, which comes across as that of a comedic short story writer trying to amuse themself in a fantasy-fulfillment way. I just don't believe that in any day and age anyone but the bottom-10% stupidest readers among us would mistake a story written this way for a genuine news article, whatever the platform calls itself.
I read your Section 8 (My actual thoughts) as a round-about admission that yes, your definition of lying is unusually high, as many commenters point out, even though you mostly individually refuted them. That is ok, though, it is your choice how strictly to take the definition of lying; if anything, it clearly shows that even you are operating somewhere along that spectrum of truth vs. lie, and your original headline is clearly not a lie given your own definition, but might be viewed as deceptive by others who have a narrower definition of sharing true facts.
I do very much appreciate your article and the discussion though - it makes it very clear that outlawing, or even identifying and condemning false stories is exceedingly difficult. Your point that mainstream media and conspiracy platforms are not fundamentally different, just a different points on a spectrum is the one key and true fact here.
Perhaps we have to take the old adage that "you cannot not communicate" and change it to "you cannot not lie?" Writing a story that is entirely true, based completely on verifiable evidence, and not deceptive at all is very difficult in a complicated, messy and contradictory world.
FWIW, I thought picking apart the ways they give misleading ideas without formally lying was a nice way to teach media literacy and criticism. Kind of like an updated 'How to Lie With Statistics'.
>I asked Jeremy to guess the probability that this was an honest mistake vs. malice. He said (thanks for giving a clear answer!) 60-40 in favor of malice. I think this is pretty high, given that I had to read Jeremy’s comment several times before I realized what the error was supposed to be
Did anyone bring up [JJ's Razor](https://www.sonyaellenmann.com/2018/03/jjs-razor.html), the corollary to Hanlon?
>“The intentionality of an agent with behavior sufficiently indistinguishable from malice is irrelevant.”
I don't think it was malice, but excusing The Washington Post (or The NYT, or whoever that's supposed to be at least occasionally reliable; ie, not Cousin Jimmy's Facebook Rant) for an "honest mistake" of this scale is too generous. It's... what's a good name for it, conveniently advantageous stupidity? What did you call it last year- [too good to check](https://astralcodexten.substack.com/p/too-good-to-check-a-play-in-three)? Dishonesty that tells the right story doesn't get checked, and while it's not technically a lie, neither is it usually an honest mistake. Or rather, we can't know the difference between an honest mistake and a dishonest one, and the distinction is irrelevant for truth-seeking anyways.
Rarely do I think the supposedly-trustworthy outlets are *malicious*, per se. But neither do they make *honest* mistakes. You say all your readers are too paranoid, all your readers say you're insufficiently paranoid, and the truth is somewhere in the middle.
Brian Sicknick was not murdered.
This was known immediately upon his death. The Capitol Police put out a statement about it on Jan 7. (I can't find it, but I can find a Reuters article from Jan 7 that references it https://www.reuters.com/world/us/us-capitol-police-officer-has-died-following-attack-capitol-building-cnn-2021-01-08/ ).
However, the NYT said Sicknick WAS murdered, and so WP:RS applies and that false "fact" got repeated everywhere, including Trumps impeachment.
Now I would think that the NYT might want to explain how/why they pushed this lie (er, untruth) and a good way of doing this might be to burn their source for the "murdered" bit. After all, that's a mistake so egregious that literally nobody should use this person as a source again.*
But they won't.
Which is odd, considering that they've been willing to out bad sources before. Why not in this case?
-They are good friends with the source and don't want to embarrass them?
-The source is politically/socially connected in the journalist community?
-The source has been reliable (in the sense of truthful) in the past and has pinky-sworn that they won't make such a mistake again?
-The source doesn't actually exist, but was inferred from a DM-chain?
-The source has been reliable (in the sense of telling the reporter what they want to write about) and they don't want to lose that resource?
*Of course, I though the same thing about Vox after their "those meanie Israelis shut down the bridge between the West Bank and Gaza" article, but apparently most people disagree with me.
You're all still here? It's over. Go home.
One of the reasons for making the lying distinction as narrowly as Scott does, that I have not seen mentioned here or in the other two columns, has to do with which forms of deception are actionable, and in what ways. (I made a comment about this on DSL.)
Lying in the sense Scott seems to want to use, is libel (or slander), and you can sue for it. It's hard, for multiple reasons, but possible. If a journalist makes a claim, and they know it is false when they make it, and it causes damage to you, and you can measure that damage in dollars, and you can convince a jury of all this, then a judge can punish that journalist in a way that journalist will feel, and make that type of lying less likely to happen again.
Lying in the sense a lot of people here seem to want to use, is practically impossible to bring a suit against. Karl Gallagher, in another comment here, mentions "sins of commission" and "sins of omission". Commission is the above thing, and an easy call. Omission, though, can happen for lots of reasons, some innocent, and is technically unavoidable unless every article comes with a bolus of every known true claim in the universe. "Oh, you claimed Saddam was shopping for nukes, but you left out the bit where OBL was Saudi Arabian!" Is the latter relevant? To some people, in some contexts, it is. In others, not. In yet other contexts, it's assumed, and clutters the article if included. Clutter matters. If a journalist makes an important claim in an article, but has to attach megabytes of exposition to avoid any possible charge of omission, readers will glaze over and skip to the next article, and that claim may as well never been made, and so is therefore unmakeable. By extension, every interesting (and possibly true enough) claim is unmakeable. I don't consider that situation preferable.
So omission can't warrant a suit (AFAIK). The defendant can always argue that the omitted claim was irrelevant, or assumed known but not important enough, or out of scope or context, or even itself questionable enough to be un-includable without even more justification. If we ruled these excuses insufficient and we allowed such suits anyway, we would quickly flood the court with them, and nearly any article could be found punishable, and we'd find the suits correlating with "someone had an agenda opposed by the claim" much more than with "journalist was being deliberately deceptive". Ultimately, every journalist would clam the hell up (or find a black market), and there'd be little news to be found anywhere, even the stuff that's true enough to act on.
In a world where one type of deception can bring a formal response by secular law, and another type is effectively unassailable via the same means, it's worth making the distinction. I contend Scott has done so.
Notice I'm saying "via the same means". There are ways to deal with deception by omission, by misleading, etc., and there are even institutional pressures that arbitrarily block those ways, and I think there's changes worth making in those areas. But those ways still require acknowledging the distinction.
As a corollary, acknowledging that distinction will also mean acknowledging that both InfoWars and NYT commit sins of omission, because only then can one group of people with evidence of InfoWars making a sin of commission that NYT isn't, convey that information to the group of people who don't yet have it. The alternative is to yell "one's lying, the other isn't!". That language is so low in resolution, that it makes the former group of people look like they're being deceptive as well. In other words, for example, I see no way to make the case that InfoWars is worse than NYT (or WeeklyWorldNews is worse than WaPo, etc.), without also copping to the charge that NYT cannot be implicitly trusted.
I know I'm late to this conversation, and I haven't read all the comments to see if anyone has made this point yet, but I was reminded of this story from <a href="https://www.npr.org/sections/alltechconsidered/2016/11/23/503146770/npr-finds-the-head-of-a-covert-fake-news-operation-in-the-suburbs">NPR</a> from 2016. NPR actually went to the trouble of tracking now a Fake News (under the strictest definition) purveyor, and found this guy:
<blockquote><p>[Jestin] Coler is a soft-spoken 40-year-old with a wife and two kids. He says he got into fake news around 2013 to highlight the extremism of the white nationalist alt-right.</p>
<p>"The whole idea from the start was to build a site that could kind of infiltrate the echo chambers of the alt-right, publish blatantly or fictional stories and then be able to publicly denounce those stories and point out the fact that they were fiction," Coler says.</p></blockquote>
Many more fascinating details at the link.
It occurs to me that fake news isn't just run for the clicks (as in the Macedonians) or however you want to characterize what InfoWars does. Some of it is a pretty straightforward effort at embarrassing your opponents, with a potentially large payoff. Let's say the FBI agent murder story (as detailed in this article) infiltrated a congressional candidate's talking points because some low-level researcher tasked with showing that Vince Foster was part of a weird and disturbing pattern found it and didn't vet the source, and his notes went to a speech-writer who then didn't double-check the researcher, and then the congressional candidate is saying something that isn't actually true, and then the mainstream media spend the rest of the campaign talking about it.
If anything, I'm disappointed that you're making this out to be a technical nit-picky point.
Most people seem to care passing judgement on media outlets for their morality, and If that is your goal, then sure, the a strictly literal definition of "lying" might be nit-picky.
But if your goal is to be able to be able to make sense of the world, then being able to tell the difference between outright lies, shoddy speculation, and misleading framing is a crucial skill.
We can reason through own analysis and model our own framing, but we're more or less reliant on outside sources for our raw facts, and the knowledge that we can _mostly_ trust those raw facts is extremely useful.
"I think this is pretty high, given that I had to read Jeremy’s comment several times before I realized what the error was supposed to be"
You must have aced calculus.
The Bulletin of Atomic Scientists thing appears to be a genuine miscommunication, not even a motivated misreading of facts or the like. The Bulletin talked to a bunch of the scientists MacAskill cited and five of them really did say that they didn't remember talking to him. This is because they were contacted by a member of his research team who didn't mention that the questions were for his book: they really gave the quotes, but didn't think they were giving them to him, for that.
Scott: <writes a third post in a series trying to explain the point better>
People: CA psychiatrist says that "the facts" used to spread MISINFORMATION were ACTUALLY TRUE
> Regardless of any of this, the point I’m trying to make with these posts is that the media, while doing all of 1 through 6 pretty often, very rarely does 7.
Why did you want to make this point though? I don’t think a significant number of readers learned something important or changed their minds about anything?
Regardless of the exact definition of lie used, this article has caused me to realise something I hadn't properly internalised before.
Most people - even otherwise very dishonest people - do actually have an subconscious aversion to just making things up completely.
Sadly, even the tiniest shred of evidence is enough to overcome this.
>Could I have saved myself some trouble if I had titled the post “The Media Very Rarely Says False Things”? Or “The Media Very Rarely Makes Up Facts”? I think people would have been equally annoyed that I was using “false things” or “make up facts” in a way that excludes technically-true-but-misleading statements.
That's because people are mainly complaining about your category, and only secondarily about your definition. "Makes up facts" is a category that doesn't carve reality at its joints very well--people find such a category much less useful than they find the category "lies". They are upset that you are applying a word for a useful category to a less useful category, but they are also upset that you are using a less useful category at all. Just changing the word you use doesn't change most of the complaint.
>Oh god, if saying there’s “no evidence” for something counts as a lie, then every media source in the country stands hopelessly condemned.
You seem to be saying "this example makes media lies common, so it doesn't count". A better conclusion is "media lies are common; this example demonstrates it."
(This is especially so since you've emphasized literal truth. If literally true means it's not a lie, and context doesn't count, "literally false" *must* mean "lie" and context can't count for that either. And it's literally false that there's no evidence.)
Reason a couple of hours ago posted something which connects to some of these comment threads (one in particular actually but Substack is being stupid and not letting me find it right now):
"How Do We Solve a Problem Like George Santos? The slippery slope of political fabulism, from the "Jew-ish" freshman representative to the president of the United States."
https://reason.com/2023/01/12/how-do-we-solve-a-problem-like-george-santos/
"numerous comments getting angry at me for saying that I was calling NYT exactly as bad as Infowars, and saying I was being deceptive / lying because of this. " - This is a clear sign of a) so. reading too many of this comments - ACX is a no lies-place, that is one reason we are here
b) having done a post that triggers sub-par comments (this one included). Which may happen. - About the NYT: was their hit-piece about Scott a lie (by his definition)?
A lie requires intent. "Intent" of pure talk - is usu. a hard tell. Why do we talk? Why do I write this comment?!? - I want media to be mostly right, not wrong. I hope.
Bigfoot-tabloids run on interviewing cranks? Disagree. We have (had when me young) one in Germany, too. (No, not BILD.) And we do not have enough people believing in alien-abduction/bigfoot/third eye (as in: a real third eye at the back of one's head). Still that tabloid brought out those stories every month. They were not lying but in the business of writing and selling a regular bigfoot-tabloid. Fiction. Same as Harry Potter. Same as the dozen or so of royalty/celeb-journals that need a dozen "stories" every fortnight; same as Barbara Cartland or Zane Grey. They do not do "research" or interviews, lol, grow up.
I don't understand how this point is confusing at all. Of course there's a meaningful distinction between "lying" and "saying true things in misleading ways". I get that it's pedantic but isn't a rationalist blog the place you're supposed to go for constructive pedantry?
> the point I’m trying to make with these posts is that the media, while doing all of 1 through 6 pretty often, very rarely does 7.
Fair enough, but does this distinction make enough of a difference ? Let's imagine two Bugmaster clones: one agrees with you, and the other believes that the media does #7 pretty often, as well. How will their behaviour differ (with respect to the media) ?
Based on https://twitter.com/Kantrowitz/status/1613168223054188545
(ChatGPT said that Musk was a Twitter CEO since 2021)
ChatGPT :
- very rarely persistently lies,
- but it is very obsequious, and so
- it makes up a story to show it is a good bot and
- has a lot of motivated cognition,
- it is not aware of where its own data comes from but
- it is good at rationalizing after the fact
Basically, it has a bunch of very human failings.
I had a sudden brilliant hypothesis in relation to this: "The media very rarely asserts anything (true or otherwise) other than quoting people." This explained exactly why they never lie, as the only lie would be literally making up quotes. Having concluded this, I opened today's Times to check my hypothesis...
And it was half write. The media makes factual assertions about events (as opposed to what people say about events) all the time. However, there is a pattern to the sorts of stories they do and don't report like this:
(I'll call statements such as "John Smith shot his wife" Direct Reports and "John Smith shot his wife, according to experts" Indirect Reports. All examples are from today's (London) Times.)
It's completely standard for Science/World News articles to use Direct Reports. For example, "President Putin is trying to pin the blame for his faltering Ukraine campaign on rogue Wagner mercenaries as he seeks to consolidate his power within a dwindling band of loyalists." There's no source cited either named or anonymous.
Articles about people being "told" to do things (e.g. "Hand out statins on demand, NHS told") aren't really news stories at all, their editorials printed in the "News" section by means of a friendly think tank/rent-a-quote.
Articles about politics are a mixture of Direct Reports of easily verifiable facts printed directly (eg. what's happening in the legislature), and Indirect Reports of everything else. There are few or no Direct Reports that you couldn't easily work out the source of.
Crime/Court Proceedings articles are mostly Indirect Reports, with Direct Reports interspersed.
Culture War articles are mostly Indirect Reports. Direct Reports only occur where they represent very simple analysis, e.g. "The shortlist for artist of the year is all male and four out of five nominees for album of the year are men."
Finance/Economics articles are all Indirect Reports, other than stock prices and indices.
In any articles, background information will often contain Direct Reports in places which violate the above rules. Looking at my own archive of "Newspaper is wrong" screenshots, these are often incredibly sloppy (eg "The poet participated in the feuding between the Guelphs, a mercantile family, and the Ghibellines, who were predominantly agricultural" or referring to Cecil Rhodes as a "19th Century slaveowner"). I'd give it 70% that these are just thick journalists, 30% that they consider this to be simplification as opposed to falsehood.
This broadly makes sense when you think about it - unless an article is written by an eye witness, the information in it must have come from somewhere. There's no reason not to tell you where it comes form, other than saving space.
I think the imporance of it is that if we accept (7) as the definition of a lie (which I'm happy with), newspapers really do very rarely lie. But the reason for this is that (7) can only apply to Direct Reports or inventing quotes, and Direct Reports just aren't a big feature of domestic journalism, and especially the sort of journalism where anyone would want to lie.
TL;DR - The media very rarely lies because the media very rarely says anything
"The poet participated in the feuding between the Guelphs, a mercantile family, and the Ghibellines, who were predominantly agricultural"
My jaw is dropping, where did you get that nugget of Dante explaining from? 😀
https://www.thetimes.co.uk/article/divine-comedy-as-florence-and-ravenna-vie-for-dante-s-bones-0qwbqsgpx
In other words, the paper of record of a country that was ruled by the Guelphs for over a century, and who's largest constituent has the Guelph cross as their flag.
“If you’re 71% confident, and you falsely say you’re 72% confident, then you are lying.”
If that’s not just an offhand comment, and you really mean it, then by that definition every conscious agent who has ever lived or ever will live, aside from perhaps certain AI agents that may come to be, have lied and will lie again. And being in a media position would probably incentivize this kind of lying a lot; in fact many of your examples of media not lying but being deceptive or sloppy fit that category. Overstating one’s confidence is a common occurrence, hell I did it at the top of this post. People don’t usually provide artificially precise Bayesian confidence statements, but if pushed how many times will someone who says they’re “certain” about something back off from their certainty? From my own experience, I know a lot of those times I am in fact ginning up some kind of epistemic courage or feeling pressured to firmly state a belief that in my gut if I attend to it I will realize I have less confidence in that I’ve presented. And after that reflection, if it’s say a job interview, there’s a good chance I will stick to my inflated confidence.
Of course at this point this gets into the qualia of lying and deception, and from an outside perspective if you question someone’s confidence and they back down, it’s an ambiguous judgement call whether you think their initial confidence was on some level a lie or just a lack of reflection on the part of the overconfident one. Any given instance of this may be undetectable but it seems like the sort of thing that happens all the time.
It seems really predictable and obvious that the #7 type of lie would be rare in media reporting, since it's so incredibly easily avoidable. If you can just find literally any idiot to tell you exactly what you want to report, and then you can report it as the truth without technically "lying", then why would you NOT use that super convenient defense? Any 'evidence', any 'witness' is sufficient, and you can just make up a QED at the end. Wham, bam, now you're telling the truth.
I'm not really sure how this is a helpful distinction though; I don't think many people are surprised by the idea that most media misleads without committing to full-on #7 lies. You would have to have some pretty motivated anti-censorship cognition going, though, to conclude "technically they aren't lying in the most stringent sense possible, therefore censorship may not be as easy/effective as most people think". For those considering what sorts of 'news' should be censored or moderated away from certain places based on untruth, I imagine very few would consider #7 lies to be a unique subclass of offense, while everything else gets a pass. Outright lies are not a requirement for misinformation; I can find a source to tell me anything I want you to hear.
There's also use of sources, and while I don't think it's lying as such, it's easy to have the 'reputable' sources that you use regularly be the ones with liberal opinions, or at least ones that fit in with the reporter's/media's biases. 'All right thinking and good people of course think X is a human right, so covering this story I will ask Dr. Y and Professor Z for quotes and they will confirm that X is a human right, then I put in a bit at the end from those kooks and bigots wanting to ban X' style reporting.
For instance, Fr. Thomas Reese gets regularly quoted by media because he's a reliable source, and he's on the liberal side so he'll tell them that the Church needs to modernise for the world or this new encyclical is the same old homophobia or whatever. And that's great, and that fits with their own opinions, and they feel like they are being fair and impartial because after all, they asked a real live Roman Catholic priest for an opinion so he must know! If you said they were slanting their reporting they'd be indignant because how dare you, they went to the proper guy and everything for his views and he's the expert here! It never dawns on them that going to the same reliably liberal guy for the same reliably liberal interpretation every time, and not searching for the other side (unless they're doing a bit on 'unreasonable stick-in-the-muds try to hold back progress') is being slanted.
De gustibus non est disputandum sed solitarius postulatio rigoris.
Why I don't think this very strictest definition works:
Let's say I make up a fact and publish it. Technically, I don't *know* that it's false since I also made sure not to check - perhaps it just accidentally happens to be true. But it's ridiculous to claim that I'm not lying here. If *this* is the standard, then *of course* the media almost never "lies".
Speaking briefly to the John Buridan point about how we develop anti-conspiracy-theory antibodies, and inform our priors... the decisive moment of that for me was reading Michael Shermer's book "Why People Believe Weird Things", which explained at a level my high-school self could digest, all about cults and creationism and holocaust denial and whatnot - and then devoted several chapters to debunking them simply and effectively. The thought patterns that encouraged have been invaluable throughout life, because "knowing what the right questions to ask" in such situations is very much an art. It's a teachable art, but few people blunder upon it all on their own without help.
I heartily recommend the book to everyone's attention. Some things hold up really well over time.
great piece, I love this analysis of the press, lying, etc.
Are people doing some kind of emotional thing like... "I hate this #5 so much, more than a normal #5... so maybe it's really a #7"?
Getting rid of #7s from the media would leave the vast majority of the problem still in place. Does anyone really doubt that?
Yeah the basic tension in this is the maximal point that it is by no means clear that Infowars is telling lies is being reacted to in a binary way to defend the basic principle that you can identify falsehoods in some or many cases. I think part of this is to defend the utility of honesty in personal life which feels threatened by the maximal point but is really a different kettle of fish than assessing honesty in reporting about a broad range of subjects that are often filtered through a world model that motivates biases to protect it without the reporter knowing it consciously
Part of the problem here is that "lie" is considered a bad thing, so when someone says something untrue and the listener hates them, said listener can describe the statement using the single-worst word in the "falsehood" Russell conjugation.
The example I had in mind is the large collection of falsehoods that the Nazis stated about the Jews. (Sorry for invoking Godwin's Law here). According to Scott's analysis, the Nazis were not lying about the Jews, since the Nazis believed every one of their antisemitic canards and conspiracy theories. They technically avoid definition #7. Yet we still commonly say that the Nazis lied about the Jews. Why? Because lying is bad and the Nazis were bad, and their falsehoods about the Jews were really egregious and defamatory. It might also be because we have actual, definition #7 examples of the Nazis lying (ie, Hitler promising that he had no further territorial claims on Czechoslovakia beyond the Sudetenland), but careful readers should still note that "The Nazis were liars" and "The Nazis lied about the Jews" are two different statements.
The expression "believe their own lies" shows that the word "lie" is commonly used outside of the strict definition #7 sense. I think that the common definition of "lie" is the union of definitions #6 and #7, and the most negative Russell conjugation of "falsehood" under definitions #2 - #5. When someone says "lie" under definitions #6 or #7, it can involve morally positive examples of lying (ie, a resistance member hiding Jews and lying about it to the Gestapo officer checking in on her house), but when someone says "lie" to describe something that fits into definitions #2 - #5, it is always to denounce.
I think people are generalizing from personal life where an individual has very specific knowledge about events and actions in their life making it hard to say something untrue without being more on the definite lie part of the spectrum to journalistic reporting and general worldview or model dependent ways of making sense of the world
There are two factors required to be able to report accurately. One is the skill to do so and another is the desire to do so. You often need a lot of both and it's easy to fall short one way or the other without consciously intending to. While telling the truth in personal life is rightly strongly held and therefore appealing to transfer on a one to one basis to people who are reporting on a wide variety of issues it just doesn't go one to one
Approximately nobody in the general media is motivated to lie when reporting about fusion experiments yet it's perfectly normal for me to see basic mistakes in fusion articles. You've heard net energy was achived? Big lie. The total energy balance was 1%. You've heard that and therefore fusion is clearly a dead end? Big misrepresentation. Anyone who knows the facts is poorly justified in saying that. The laser technology itself may be capable of 40x greater efficiency and there are other more readily commercialized designs that may actually reach a full net energy gain by 2030. In other words both of the primarily reported facts about the fusion energy gain experiment would be pretty easy to call lies if you thought they were wrong and thought that the reporters had a special bias towards giving inaccurate information about fusion
So now maybe you want to agree with that but also maybe you don't because it takes away the "Just Stop Lying" button that you could just keep pressing hard enough to escape from the Late Pre-truth Era. I have an answer for that. On Metaculus the current prediction for a 100 MW fusion reactor to come online is 2035. That's probably a very accurate piece of information that you won't hear in many other places
> In a perfect world, we would have separate words for all of these. In our own world, to save time and energy we usually apply a few pre-existing words to all of them.
After being distracted for entirely too long with this, I propose the following words for those 7 things:
1. veracinate - from latin verus (truth) and ratio (reason). reasoning truly.
2. misveracinate - now you've fucked up - but you were at least trying to reason truly
3. idioracinate - you're reasoning like an idiot or common man (latin roots)
4. karsioracinate - from the greek karsios (oblique) aka skewed reasoning
5. karsiocept or karsioceive - from latin capere (to take), it's unintentional deception via bias
6. veracept or veraceive - you're taking the truth from people, subverting them with facts
7. deceive - we dont need a new word for this.
But I bet Scott doesn't read the comments 2 days later...
Your 7 categories do not seem to include the most common of them all:
8. Being so certain of the Higher Truth that precise details don’t matter when they obscure that higher truth.
I raise this point because I suspect it drive almost all of this nonsense. Who CARES whether the police kill 15 or 5000 black men every year, “we all know it’s too many”. Who cares whether 1 in 3 or 1 in 300 women is raped, “we all know that all men are potential rapists”.
If you know the TRUTH then you are doing God’s work by limiting the ability of your readers to be distracted by irrelevant details that confuse the point.
The difference between 8 and 7 is that you don’t even believe you are making up something; you are simply reporting the reality and if facts say otherwise, well, soon enough we’ll realize those facts are in error.
I think what you’re describing as 8 is just an unusually intense manifestation of the more general way in which we see human beings react emotionally to certain politically charged topics and then reliably rationalize the moral necessity of the various shades of deception delineated in 5/6/7.
I’d highly recommend HotelConcierge’s blogpost ‘The Tower’ for a far better explanation of this phenomenon and a deeply illuminating treatment of among other things, how our current hyper-partisan, depressingly cynical media climate might be the logical, inevitable outcome of unlocking forces such as the inexorable free flow of information and mixing it with the zero-sum, my-way-or-the-highway instincts of human nature.
I think the difference is that you view people engaged in 5, 6, 7 as thinking "OK, this is a lie, but it's justified". Whereas I view people engaged in 8 as not even believing that it's a lie.
If a Christian talks about the Resurrection, are they "lying"?
OK, now if a Democrat talks about how Trump was a tool of the Russian Government are they "lying"?
In both cases there's zero evidence for the claim that's convincing to an outsider; what does exist is a pattern of being told a claim enough times that the listener accepts the claim as so true that it's now foundational; it doesn't require evidence, rather it IS evidence in terms of which new data should be interpreted. Because we KNOW that Trump is a tool of the Russian Government, therefore such and such a new deal with Putin is obviously structured in some way so as to damage America...
I think different people are arguing aspects of this based on different goals.
Some are arguing with an "ethics" goal in mind (it's not as morally culpable if you didn't make up the statement, or whatever).
Some are, I guess, arguing some sort of "legal" theory of what can you get away with.
But a few (like me) are arguing with an "understanding" goal in mind (what do these people imagine they are doing when they flood the field with these crazy claims month after month, year after year?) It's in this 'understanding' context that 8 is a different sort of phenomenon from 7 or earlier.
We're a community with a self-selection bias that sorts for folks who fancy themselves smart (you know it's true, c'mon), and I hate to break it to you fellas but we have more specific words in English to cover all of these edge cases without blurring definitions. The first two that come to mind are duplicity and subreption.
I think that if we're going to strive towards goals like: cultivating sophistication, encouraging and preserving nuance and enhancing individual sovereignty, we've got to recognize when lazy colloquial language is doing harm to our ability to discuss a particular subject.
I can never keep up with all the comments on your great posts, so apologies if this was already posted as an example, but here's a clip from a CNN report where they are not just lying, but they are deliberately saying the exact opposite of what's actually happening. https://www.youtube.com/watch?v=zfOcz--2AjE
The context is a gathering at a neighborhood vigil after a police shooting of a young man. There was rioting in response to the shooting. The deceased's sister is speaking to the crowd, and the CNN voiceover describes it as, "Smith's sister calling for peace...". The reality? She was encouraging the rioters to go riot in white neighborhoods. (They actually issued a mealy-mouthed apology for it: https://www.buzzfeednews.com/article/davidmack/cnn-sorry-for-not-airing-full-clip-of-milwaukee-sister).
I don't see how this can be interpreted as anything but a deliberate lie, and not just by selectively showing only one part of her words, which on its own would count as seriously misleading, but by characterizing it as they did, it shows a willingness to unabashedly lie to promote a desired narrative.
Scott, I think you are equivocating here. In section IV of your original post you said the point of this was to show that fighting misinformation was hard. How (for the most part), you couldn't get very far simply by removing posts from "liars saying completely false things".
However, if your threshold for labeling something "misinformation" and removing it is not only that what it says (or maybe even clearly implies) is blatantly false, but you also require that the author of the article didn't believe what they were saying at the time, it would be impossible to identify misinformation even if it were prevalent. I mean what do you expect Facebook to do, bring Alex Jones in to take a lie detector test every time InfoWars says something blatantly false?
Any reasonable argument in favor of censoring obvious misinformation is going to label something "misinformation" if it contains prominent claims that are blatantly false or maybe even if it is clearly designed to convince the reader of an obviously false claim. But in any case, your censor probably isn't going to care about the state of mind of the person who wrote the article. So if the ability of censorship to fight misinformation is what you really care about here, you shouldn't care about whether Alex Jones actually believes Obama's birth certificate is fake either.
I think the concept of intent may be part of the puzzle here. Notably, if you assume intent then probably you should assume positive intent, and then the test for lying becomes difficult to fail.
I wrote a related post here: https://pelorus.substack.com/p/does-intent-matter
I am still going to push back on this framing.
Tl;DR Making false facts is Thing 7 and misreported confidence is Egregious Thing 6, but only if a person is reasoning well. A person reasoning poorly can commit a Thing 4 at worst. But a poorly reasoning person can intentionally make false facts and overstate their confidence, recognizing that they are doing so. Merely noting someone's unreasonable state does not guarantee a lack of intentional deception. Furthermore, a deceiver can often exaggerate their unreasonableness to escape consequences.
You say in Section 8:
|3) Reasoning badly, because you are dumb.
|4) Reasoning badly, because you are biased, and on some more-or-less subconscious level not even trying to reason well.
|5) Reasoning well, having a clear model of the world in your mind, but more-or-less subconsciously and unthinkingly presenting technically true facts in a deceptive way that leaves other people confused, without ever technically lying.
|6) Reasoning well, having a clear model of the world in your mind, but very consciously, and with full knowledge of what you’re doing, presenting technically true facts in a deceptive way intended to make other people confused, without ever technically lying.
|7) Reasoning well, having a clear model of the world in your mind, and literally lying and making up false facts to deceive other people.
|...I prefer to reserve lying for 7 and the most egregious cases of 6
And in Section 2:
|I don’t want to say you’re allowed to sound more confident than you are. If you’re 71% confident, and you falsely say you’re 72% confident, then you are lying. But if you are very dumb, and seeing a random piece of toast makes you 100% confident that Obama’s birth certificate is false, and you vomit some random words to that effect onto a page, then you’re an idiot but not a liar.
I object to this because:
A) Many people simply do not reason well and have a clear model of the world in their mind when discussing controversial topics. Under this framing, a person who is sufficiently mindkilled cannot lie, even if creating a false claim from thin air.
For example, I recently learned that a friend, devastated by a breakup and trying to express her distress, uttered the words, "He raped me". Another friend took her at her literal word, and started to file a report. She immediately realized her mistake and tried to stop him. (There were no charges, but I don't know if she successfully prevented the initial filing.)
Suppose she had decided that her treatment was morally equivalent to rape and should be prosecuted, and had proceeded with the report. It would have been a mere "Thing 4", because her reasoning was compromised by emotion.
I think most intentional falsifications about the world involve a highly unreasonable individual. It can be very useful to distinguish a cold-blooded and hot-blooded deceptions, but unreasonable people can still intentionally deceive. If internal states matter, then the person's consciousness of their deception should be a major factor, even if they are also unreasonable.
B) Under this framing, a person can vastly reduce the apparent badness of a falsehood by merely pretending to have been reasoning poorly. Because we don't have access to peoples' thoughts we can rarely distinguish this. I think this is a major source of contention by the comment section, especially regarding InfoWars.
C) Unless I grievously misunderstand, the original framing was about moderating false reports such that only "lies" are removable. However, it is extremely difficult to determine whether someone is exaggerating their confidence by 1% (Egregious Thing 6). It takes very little bias for a person to overstate their confidence, and (as in B) even if somehow caught, one can escape moderation by blaming poor reasoning. Even in the case of explicit Bayesian reasoning, you could nudge the priors or introduce a subtle math error, and if needed pretend it was because you made a mistake (Thing 3). This form of lie is effectively unaccountable.
I wonder if there should be a term for the absence of Gell-Mann Amnesia. I just read an article about a subject I had personal knowledge of, and it was surprisingly fair and accurate.
Although what's even weirder is one time last year when I read an article on a topic close to me and it was overall fair and accurate but had one offhand statement that was completely false. It was kind of baffling how that could even happen, and it wasn't ideological or bias or anything either. Presumably, the source for the article just misremembered something and they didn't bother to check.
I found an example of Scott using "lie" to mean exactly what he says isn't a lie here. :) From an old SSC post, discussing the media: https://slatestarcodex.com/2015/09/11/book-review-manufacturing-consent/
==
As per Chomsky, this was rarely done by direct lies, in the form of front page “EVERYTHING FINE IN GUATEMALA, SAY SOURCES”
==
By Scott's current definition, if there are indeed sources who are willing to say "Everything fine in Guatemala," that would not constitute "direct lies."
I think the fundamental issue is sentences of the following format:
--
"[completely false assertion]," said [totally dishonest person], who studies [general field of knowledge that the dishonest person told reporter he studies and which relates to the completely false assertion].
--
Under Scott's current definition, as long a newspaper accurately quotes a source, it's never a lie. So you could have a lengthy article consisting 100% of false assertions, but so long as they are attributed to the people who said them, the article has not lied.
The problem is, there are 8 billion people on the planet, and you can find someone to say *anything*. And even Scott in 2015 called things like this "lies."
>This is why I’m so convinced people are erring on the side of too mistrustful - you can fill your articles with sentences about how you’re not claiming X, and people will still find ways to accuse you of lying because you said X.
The problem is that you wrote a post strongly implying it, and that still holds true even if you were careful to insert a couple disclaimers saying you didn't actually think so.
"My main argument against this is that millions of people believe conspiracy theories - if they didn’t, we wouldn’t care so much about them!" This seems like a bit of a tautology to me.
Technically speaking, we might care about them because we *think* that millions of people believe in them...not because that phenomenon actually exists.
And if anything, I also think we're always somewhere between intrigued, horrified & entertained by the idea of people who believe totally ridiculous things. So much so that there there is a kind of psychological bias to *want* to believe that conspiracy theories exist & have loyal believers (which, by the way, I think is part of the actual audience for InfoWars as well...namely all the people constantly pointing at it with some judgemental lens).
So, all that is to say is just that the logic here doesn't totally hold up for me and I remain unclear about how many *actually* believe in any of these things.
From what I've read, the Weekly World News in particular did totally make up a significant portion of its content, but it also went through a lot of local and other newspapers for weird-but-supposedly-true stories and republished them. (The employees thought of themselves basically as entertainers, I think.)
Is part of the thesis "lies in headlines don't count as long as the article doesn't explicitly lie?" Because you lost me in the very first post of this series when the very first example of "not a lie" was "New Vaccine Data Shows Alarming Number Of Stillbirths And Miscarriages Caused By Covid Shot." That is a lie of at least grade 6, probably 7. The data do not (and cannot) show causation. So I stopped reading at that point, because that's what everyone does when an article seems like obvious bullshit within the first few sentences (based on their own individual biases, of course).
Any discussion of lying in the media that ignores headlines is irrelevant.
A more accurate headline for your article would be "The Media Very Rarely *Technically* Lies," but you probably didn't go with that because:
1. Not as good clickbait
2. Obviously true
3. So what?
And you'd still need a better first example.
To address the 1-7 taxonomy you presented at the end: I think 6 is definitely lying. I understand that you're making a narrow point about 7 in these articles, but 6 seems to me to fit entirely within the accepted understanding of the word "lying." If I come home after a late night and tell my wife, when she asks where I was, "I was out with the guys, we got a few drinks," but I intentionally omit the context that the drinks were at a strip club and I also paid for a private dance, I think most people would say that I lied to my wife about where I was.
Also, I think we're actually pretty close to having separate terms for each of these numbers. 1 is just being correct, and 2 is just being incorrect in the most straightforward way. 3 is also being incorrect, although you might tack on a pejorative like "stupidly". I think 4 would rightly be called "willful ignorance," although we could certainly get into a debate about the implications of the word "willful." 5 is the only one I don't think there's a good concise term for. (And of course I think both 6 and 7 would be "lying," although 6 contains kinds of lies that have more specific names, like "lies of omission," and "statistical lies.")
How does this sort of thing rate on the "degrees of lie" scale? https://donmoynihan.substack.com/p/anatomy-of-a-fake
I didn't really know about substack until I stumbled on a link to the original post of this series. No idea how many blogs are of this quality or if it's just a few, but I feel like I've warped into a place on the internet where people are not only smart and reasoned, they make me feel downright dumb. I legitimately thought I was smart before.