I love this post and agree with the diagnosis and prescriptions for the most part but I think it's somewhat missing what's happening with the olds. There's plenty of rage incubating in elder gen X and younger boomers that seems to be spurred on somewhat by the internet, but much more by cable news, talk radio/podcasts, or just the watercooler. I was just discussing this with my local middle millennial friends and we came to the conclusion that pretty much all of the political extemism in our families comes from the 45-70 crowd. That's probably because we live in a more rural state, no critical mass of young antifas or groypers here, but I don't think that means this facet of the problem can be dismissed, given how the fractious fault lines of rural blue collar vs urban elite seems to suffuse, and, indeed, drive a large part of our political conflicts. I don't know how to fix that hard problem, short of firebombing Fox, MSNBC and CNN headquarters, lol. Even then, these folks have had their minds poisoned by this stuff for 30 years, I don't know if there's a way of unringing that bell.
This is very much on point. Older people watch Fox and MSNBC, not kids, while You Tube and other platforms host longer form podcasts whose hosts are extreme. Noah, your analysis is otherwise spot on, but on remedies, I see no hope for Twitter or BlueSky moderating their content -- they feed on and react to each other. There is no "moderate" Cronkite social media platform. As Sylvilagus says, I don't think there's a way to unring the bell, put toothpaste back in the container, pick your metaphor. I don't know what the answer is. But if we can't find a way for moderate forces who represent the middle of the country, where most people still live politically, the vicious cycle only will continue
As a Progressive it seems to me there is a false equivalence between the positions of the Right--as reflected by FOX News, X, Newsmax, etc.--and the Left, as reflected in primarily MSNBC. Mainly because of the sheer volume of misinformation and disinformation coming from the Right compared to the Left.
Yes, there is plenty of Trump-is-bad confirmation bias to be had on MSNBC, the Nation, Pacifica, etc. But it is all based--at least mostly--in facts; e.g. DOGE's gutting of IRS personnel will result in a $500bn revenue shortfall in FY2025 (which is extremely significant, btw). This is my opinion, of course, and is open to debate.
But to regain national moderation, we must be able to debate TRUTHS; as in Congressional Budget Office type "truths" based on non-sexy, nonpartisan bean counting. To that end, we must legislate that anything purporting to be "news"--as opposed to opinion--must be based in actual journalistic standards, like 2+ sources, actual links to actual studies, timely retractions when necessary, etc. And perhaps also be subject to relaxed libel standards, to discourage fact-free mudslinging.
To keep our Democracy functional, voting citizens need to know what's actually going on, and ensuring we have a 4th Estate that keeps to the facts is essential to that end.
"But to regain national moderation, we must be able to debate TRUTHS; as in Congressional Budget Office type "truths" based on non-sexy, nonpartisan bean counting. To that end, we must legislate that anything purporting to be "news"--as opposed to opinion--must be based in actual journalistic standards, like 2+ sources, actual links to actual studies, timely retractions when necessary, etc"
So you want Trumps office of "truth" to decide what counts as truth?
Of course not, you want your side to decide. But that's the problem, that's not how politics works. Whatever power you give to the government the other side will eventually control.
Government can't be the arbitrator of truth. Really that's the whole point of the 1st amendment.
All that being said, yes I agree that media should focus more on fact based reporting. But if there hadn't been such a huge liberal tilt in the MSM, companies like Fox wouldn't have had a chance in the first place.
For decades we had the Fairness Doctrine. Which wasn't perfect. But it was something, until Reagan sixed it.
And FYI, FOX News is in no way fringe for GOP voters. As per Ipsos polling last November, it is where 85% of all conservative voters go for some/most/all of their news.
You might disagree, but there is actually something to actual journalistic standards. Like having 2+ sources to corroborate something. Which is a shit ton better than the, "...People are saying..." non-corroborated bullshit you routinely get on FOX News, which is shorthand for "We just made this story up out of whole cloth; but need to pitch it as if it actually has legs."
Maybe the 1st Amendment will forever prohibit the creation of any meaningful "Truth in Reporting" law. But considering that only months ago a Presidential Candidate proclaimed on national TV that: A) "Haitians are eating the dogs and cats in Springfield," and B) "Democrats murder (abort) babies after birth." It's sorely needed.
You're right about false equivalance b/w MSNBC and Fox, but still MSNBC is to the left of the median voter. And I couldn't agree more about CBO. If the scorekeeper/referee is abandoned, the "game" is over
There might not be moderate ones, but there are ones that are at least neutral by default; you probably won't see crazy stuff on Facebook if you don't look for it or have friends that do, and Reddit's siloing of things by topic also helps you avoid far-left or far-right politics where it's not welcome.
Ultimately, if people want to hear what [insert extremist here] has to say, I don't really think the rest of the world should try to tell them they can't - but we also don't have to stick it in the faces of people who don't want to hear about it.
(On a related note, no individual has an obligation to debate Nazis or other people holding terrible positions, but it often helps if someone does - for example, www.talkorigins.org has an extensive archive of rebuttals to various creationist arguments.)
The degree to which retirees are marinating in Fox / Newsmax / OAN is really disturbing. I run into people in the 65+ set all the time who _sincerely believe_ that Trump is only rounding up illegal immigrant criminals, and that the budget will only take benefits away from illegals. Because that's what they're hearing.
There has been research showing that mainstream media coverage is heavily influenced by social media. Fox, MSNBC, etc. are just the last stage in the transmission chain to the “olds.”
It is definitely true that new gods have exerted some nontrivial control over behavior of the old gods, but they aren't fully in the driver seat, imo, otherwise the tech right would have had no issues with their poasting to maintain maga dominance plan, but instead middle America old person maga wrested the reins back from them pretty quickly without much trouble.
CNN is arguably right-of-center. At least on Economic issues. On social issues, like abortion, they and WaPo, NYT trend liberal. All of these press organs still sometimes call Trump out on his lying. Which, unfortunately, is often. As such they're centrist; at least compared to FOX, Sinclair, Newsmax, OANN; where misinformation and disinformation are standard output.
MSNBC is the only national network with liberal programming. Unless you count Late Night comedy shows like John Stewart, Jimmy Kimmel, etc.
Musk likes Twitter the way it is and he is primarily responsible for turning it into the toxic hellhole that it is. Just take the oxygen out of the room. It’s a bad drug, get off of it and use the extra hours of your life to do something productive.
Musk and Zuckerberg /could/ tweak their algorithms. But /why/ will they? Is there something society can reasonably do that would make that in their interest?
This is a point I think Noah is too optimistic on. It’s probable that the Musks of the world enjoy having the power to steer discourse by controlling those platforms and I’m skeptical they can be convinced to intentionally fight the trends that enrich and empower them.
Twitter was always a hellhole. The toxic ingredients have always been there--pseudonymity that removes personal stakes from posting BS, an algorithmic feed that rewards emotional content, and the inability to express nuance due to length restrictions.
I encountered Twitter a matter of months after it was created, and immediately had an allergic reaction to it. The thing I always said about it was: "X spends too much time on Twitter" is a true statement for all values of X. The name change felt ironic.
Equating Mamdani to right wing conspiracy theorists and hate slingers is extraordinarily off base. The temperament, the tenor, and the message from Mamdani is measured and reasonable. I have seen him speak multiple times about not wanting to disclaim the “globalize the intifada” language and I truly don’t believe he sees it as a call to destroy the state of Israel. He is however an anti-Zionist and believes the state of Israel should be non-religious.
He seems like one of your “normal” and decent Americans. This can’t be said of any of the toxic right wingers you identify in this piece.
Normal, decent Americans don’t go around speaking of getting elected in order to “seize the means of production”. He’s a smooth talking politician with communist and anti Zionist ideas and bad economic thinking with no experience running anything bigger than a local campaign. If he put his economic ideas into practice it would be as bad as what Trump is doing with his ignorant economics.
Speak for yourself. You don’t get to decide what “normal, decent Americans” get to think. If you actually talked with some of them, you’d be surprised with what they actually believe
Trump and Mamdani are alike in believing economic ideas that have been proven wrong by economists and by actual experience: tariffs hurt the country implementing them, rent control causes reduced quality and amount of rentals available, and raise rents in nearby uncontrolled apartments, seizing the means of production leads to economic collapse and death, deporting workers means reduced output and higher prices, making buses free doesn’t improve transportation, getting rid of police results in more crime.
But they both smooth talk a lot and post on social media expounding their belief in these bad assumptions, and both are skilled at getting ignorant voters to believe that they have reasonable ideas. These voters will get what they voted for, and the rest of will suffer along with them. At least Mamdani isn’t as bad as Trump since affected people can leave NYC without an exit tax, unlike leaving the US.
I appreciate your perspective on the economic issues and I too disagree with many of the policy proposals that Mamdani has put forth. But that still in no way equates him with hateful / malicious / greedy MAGA right wing personalities in my book.
I believe Mamdani is a skilled communicator and a genuine person with good intentions. He also seems willing to listen and change his mind, and we have already seen him moderate.
The “other side” just gets more extreme and incoherent with every passing tweet.
It was only a brief statement of what he intends to do, so Politifact says it’s “false”. This is the problem when people say they want to police social media for misinformation, when any politically biased group can claim something a declared socialist says on video supporting implementing socialism is “disinformation”.
All the MAGA fools who didn’t believe what Donald Trump said in his campaign are now regretting their votes as their spouses or friends get rounded up and deported. If a crazy guy says crazy stuff to get elected, you need to expect the government to then do crazy stuff.
I agree. Noah claims that Mamdani "... defended the phrase “globalize the intifada”". Please, Noah, link us to a reference where Mr Mamdani uses and endorses that phrase himself. In the Meet The Press interview I watched those words were put into his mouth and he basically declined to engage, very skillfully evading a direct answer:
"My concern is to start to walk down the line of language and making clear what language I believe is permissible or impermissible takes me into a place similar to that of the president, who is looking to do those very kinds of things, putting people in jail for writing an oped. Putting them in jail for protesting. Ultimately, it's not language that I use. It's language I understand there are concerns about. And what I will do is showcase my vision for this city through my words and my actions."
Agreed. I see Mamdami’s stances as well within moderate centrist values in other democratic societies. The fact that he is considered extreme in the U.S. context is a product of evolutionary rightward skewing of the center of U.S. politics since about 1980. That center is far to the right of most Western democracies at this point. And I realize I’m probably being overly generous in including the U.S. in the remaining group of democratic polities.
Yup. That was my one issue with an otherwise good piece. I like Mamdani, and so many otherwise smart centrist Dems I know are having a full-blown moral panic about him.
This article would be a lot more convincing if it grappled more substantively with the systemic problems that are fueling social/political disorder, instead of simply attributing it to the malign influence of social media (this would be akin to blaming the printing press for the chaos of the Protestant Reformation. Noah’s analysis is weakened by this focus on technological determinism, which presumes that American society was fundamentally healthy prior to the rise of social media (long story short, it wasn’t: social media simply ripped the bandaid off the wound).
One example is with Noah’s weird centrist obsession with the pro-Palestine protesters. I understand the need to punch left, both out of a genuine expression of centrist bonafides as well as a way of staying evenhanded, but there is a consistent aversion in centrist media towards grappling with the humanitarian crisis in Gaza. Even Andrew Sullivan, who is nobody’s idea of a woke scold, has conceded on this point and recognized that Israel is not the liberal-democratic underdog everyone wishes it was. This problem goes beyond the Netanyahu regime and reflects on the fundamental nature of the Israeli project, whose legitimacy is predicated on a two-state solution that is (and likely always was) fanciful. Maybe, just maybe, those immature pro-Palestine protesters are justifiably enraged and upset US military support for a regime that bombs hospitals and weaponizes famine as maliciously as any 20th century autocrat.
I’d even extend this argument to immigration. I’m no nativist, and I find the American right to be a lot more xenophobic and bigoted towards immigrants than I initially expected (I thought the “multiracial working class” coalition Trump assembled in 2024 would’ve solved the matter, but apparently MAGA is even stupider than I previously thought). But it’s clear to everyone that the old centrist regime of liberalizing immigration flows was hopelessly utopian: while immigrants do indeed contribute to the social and economic diversity of America, the social dislocation that mass migration generates (much of it rooted in the bigoted reactions of native-born Americans) is a cost that policymakers and economic theorists have consistently underestimated. And this is doubly true for other core “economic truths” like uninhibited free trade.
Another objection to this otherwise well-written piece is that Noah is assuming that the social stability of the mid-twentieth century is the norm, and that modern instability is an aberration. But it is in fact quite the opposite. Trump’s antics, however authoritarian they may be, are not foreign to the American historical tradition: populist demagogues like Andrew Jackson and Richard Nixon have long relied on such tactics, as well as quasi-genuine egalitarian sentiment, to win elections. Inequality and the paradox of poverty amidst plenty are inherent products of the market economy, as are the rise of politically-powerful oligarchs (Carnegie, Mellon, Rockefeller, etc.). This is not a criticism of America, but rather a humbling observation: some level of sociopolitical instability is inevitable, and efforts to squash it out through paternalistic nannying or indoctrination are bound to fail.
Finally, I’d be a lot more sympathetic to Noah’s arguments if he was more frank about the failures of the centrist gatekeepers whose fall from power he so eagerly laments. One of the reasons why those gatekeepers are increasingly less relevant is because the mass public has been enraged by the perception (real or otherwise) that our social elites are incredibly unaccountable. One reason the rise of Trump is so terrifying is that so many people have simply begun to tune out Trump’s autocratic antics, making it even easier for him to erode norms on corruption and good governance. It would have been much better to nip things in the bud by vigorously enforcing anti-corruption laws, even against our social elites, but now the very idea of good governance is considered laughable.
“One of the reasons why those gatekeepers are increasingly less relevant is because the mass public has been enraged by the perception (real or otherwise) that our social elites are incredibly unaccountable.”
I disagree. I think the centrist gatekeepers lost power because of the social media attack which they can not defend against.
What you have articulated about unfettered immigration and free trade, coupled with increased corruption and lack of accountability, are all symptoms of late stage capitalism. This is the root problem.
Noah's presumption is that "technology can solve everything", so don't be surprised when he doesn't include any ideas that "we need to solve underlying social issues first". https://x.com/Noahpinion/status/1941260882224480439
I think you misunderstand some of Noah's thought. It would be more accurate to say he thinks technology determines everything, and the everything would still be a bit of a stretch.
"Every technology changes the social contract. Cars changed the social contract. TV changed the social contract."
I think you might have to demonstrate how you "solve underlying social issues" to be convincing about doing them first. I can't think of many a where someone set out to "solve underlying social issues first" and succeeded. The best examples I can think of are the MLKs approach to civil rights, but such examples are few and far between, and even then there's a significant argument that this was dependent on changes in society that were started by technological changes.
Re: "It would be more accurate to say he thinks technology determines everything"
He did mention in his "about" part of his blog post, when he mentions why his philosophy is "techno-optimism" (https://www.noahpinion.blog/about):
"A big example is technology. Noahpinion is explicitly a techno-optimist blog, meaning that I believe technological innovation is going to be more important and transformative — and usually, though not always, better for the human race — than most people realize. The recent advances in vaccines that helped us get through the Covid pandemic, and the advances in green energy that promise to help us avoid the worst of climate change, are cases in point."
But maybe your point of view is more fit with the ideas of Noah though.
Re: I think you might have to demonstrate how you "solve underlying social issues" to be convincing about doing them first.
In retrospect, I was rehashing a bit some criticisms of the Industrial Revolution by Ted K, that rather finding underlying issues to solve first, scientists and business leaders just innovated out of it - obesity is one example, Americans are now less affected from it not through government campaign for healthy eating, but through Ozempic; which benefits drug companies like Novo Nordisk the most!
I agree with you about the problem, but the solutions that you propose will not succeed. We can take measures to reduce the amount of social media in children's lives, but what about adults? There are two problems— How would you incentivize social media companies to change their algorithms? Second, any government interference in algorithms is a clear violation of the First Amendment. To summarize, this is an important issue that needs much more thought.
To get Americans out in the real world interacting with each other positively, one could require a year of public service from every young citizen and also facilitate supported full time volunteer work by older Americans.
Face to face interaction is essential to human communication. Non-verbal cues and communication are entirely missing from the online forums and those subtle and sometimes not so subtle bits of social information help tremendously in moderating human social interaction. Anthony Bourdain put it thus - "Eat at a local restaurant tonight. Get the cream sauce. Have a cold pint at 4 o’clock in a mostly empty bar. Go somewhere you’ve never been. Listen to someone you think may have nothing in common with you. Order the steak rare. Eat an oyster. Have a negroni. Have two. Be open to a world where you may not understand or agree with the person next to you, but have a drink with them anyways. Eat slowly. Tip your server. Check in on your friends. Check in on yourself. Enjoy the ride." Sometimes, having been on the planet for a good long while and having seen some of the world brings a wisdom which is lacking in the young, energetic but still undeveloped mind. I should know.
"How would you incentivize social media companies to change their algorithms? Second, any government interference in algorithms is a clear violation of the First Amendment."
IMO, that's not so simple and this is where Section 230 reforms could possibly come into play. Media companies are clearly protected from liability if they're just hosting third party content. OTOH, the use of algorithms to curate content could subject platforms to liability for clearly harmful content if Section 230 did not exist.
I mostly agree with Mike Masnik's argument that Section 230 is about clarifying and codifying free speech protections that already exist in the Constitution rather than creating new protections, but I'm not quite as absolutist about this as he is.
I'm sympathetic to Masknic's more expert critics (eg Kosseff) who say 230 is overbroad and creates protections above and beyond what's in the constitution, particularly when platform owners are reckless and intentionally create algorithms they know or should know will cause harm just for the purpose of increasing profit.
As I understand it, Kosseff is largely aligned with Masnik and is no crank, but he believes adjusting 230 to require a bit more transparency and accountability in exchange for 230's liability protection would result in a better overall result for society and would still pass constitutional muster.
Algorithm reform instantly becomes highly technical. A reason it is better to just repeat Section 230 outright.
That said I think algorithm is easy to define in this context: They use an individual's history (search, sites visited, etc.) as information to feed content.
Take away the individual customization feed and the problem is far smaller, as are the profits.
Not only is it clearly against the First Amendment, but the public will reject it. We saw what happened when the Biden administration tried to do this to suppress the embarrassing laptop story and to promote its COVID propaganda.
That was poorly written. When I said "promote its COVID propaganda" what I was referring to was actually when the administration pressured social media to censor people who opposed their messaging on COVID.
You should listen to the Ezra Klein podcast with Chris Hayes about Mamdani as an early example of a political campaign native to the vertical video format. They seemed to agree that as a medium TikTok/Reels etc don't reward outrage bait quite so much as Twitter/X do.
For what it's worth, they seemed to be relatively sanguine about Mamdani himself; and I think one of their points was roughly that his cautious defense of the "infatida" phrasing actually worked well in TikTok style media, even though it's obvious fodder for attack ads and negative memes in other media forms like TV and Twitter.
Good piece, and I very much agree with the diagnosis. I'd like to propose an alternative solution, though, one that doesn't rely on the goodwill and responsible moderation of actors like Elon or Zuck whose profits are built on "engagement" and have a very strong financial incentive to platform and spread controversial content that attracts eyeballs.
Simply put, we should repeal section 230 of the Communications Decency Act. Not reform it, completely repeal it. Many of the people who attack § 230 are interested in reforming it for partisan ends and think they can keep the current internet more or less the same, while simply advancing their worldview and suppressing their opponents'. I see it very differently: a full repeal of § 230 is a deathblow for social media and user-generated content on the internet, one which cannot easily be recovered from or worked around by making superficial tweaks to comply with the letter of laws and playing an endless game of whack-a-mole with regulators.
Platforms will not be able to function if they're held legally liable for whatever their users post; the litigious nature of the US legal system will crush them in short order. It means a return to a world where the ability to publish and widely disseminate information is gatekept by having skin in the game: you need financial resources to set up your own publishing and marketing infrastructure, or you need to persuade someone who does have those resources that you're trustworthy enough that they're comfortable taking responsibility for the consequences of your output. No more of the current world where any anonymous troll can sign up a free account and get access to a world-class broadcasting system.
The flaw I see with this proposal -- as with so many other proposals to reform our dysfunctional country -- is simple, though: the politicians we have elected like the division. They like the discord. They like the fear, the loathing, the utter hatred towards our fellow citizens, because it's so much easier to motivate people to support you by giving them villains to hate and boogeymen to fear than by articulating a positive vision someone might actually expect you to deliver on. The massive amounts of money that social media companies make and their ability to buy politicians is also a significant problem, perhaps never demonstrated more clearly than by Trump's abrupt about-face on TikTok. But I think you could take the money completely out of the system and the bloc voting paradigm would still be enough that Congress would never act on this.
This comment and your reply below are the best in this thread. I would really appreciate it Noah and others with big megaphones would take up whether the benefits of Section 230 even remotely approach the harms.
If section 230 was repealed, then Twitter and Instagram might block everything except cat videos to avoid getting sued, but all the political shit posting would just move to TikTok since the owners are in China and can’t be successfully sued for liable. You can say we could just outlaw it, but we already did and Trump is just ignoring the law.
TikTok is an American company that has very significant technical operations in America. They're absolutely vulnerable to being sued in US courts. Bytedance could try to address this by simply withdrawing all presence from the US, but that would render the service very slow, kludgy and difficult to use; modern video services need CDNs and globally distributed datacenters to offer a high-volume, performant service. US courts could also just order the service blocked by American ISPs, a remedy that's been used before in cases where a foreign site is hosting something in violation of US law. Yes, people can still use VPNs and the like to access banned services, but it's slow, clunky and most people won't bother. And once you start depriving a social network of its audience, it will collapse very quickly since network effects are what sustain them.
The vast majority of speech on social media would be protected under the first amendment regardless of whether platforms are liable for unprotected speech on them.
Yes, it would. But that's not the point. The point is that defending that speech as protected is expensive. Very expensive. And with the sheer volume of stupid things people post on the internet, you're going to lose a lot of those cases, too, because it's not protected speech: it's easily-provable libel.
Many of these individual posters could be sued right now, successfully, but they aren't because the likelihood of recovering enough money from them to make the suit worthwhile is nearly zero. But if a multi-billion-dollar corporation is the one held legally responsible for the speech, the lawyers will come out like a swarm of piranhas, because winning a jury judgment against Facebook could set you up for life.
At the moment, this doesn't happen because § 230 makes these lawsuits utterly futile. The people disseminating the speech who have actual money to take are completely and unconditionally immunized from responsibility for it, so lawyers know better than to bother, and the only lawsuits you see are the purely performative ones where some rich person wants to make an example of an individual and doesn't care about the cost. Take that shield away, though, and suddenly these lawsuits become very viable. Even if the vast majority fail because they're going after protected speech, it doesn't matter: companies can't afford to defend against all of them, and they really can't afford to lose many of them. It's about risk calculus on the balance sheet: you can't afford to take responsibility for the speech of every random person with an email address.
Edit: the follow-up to note here is that what I'm proposing is literally how it works in every medium _except_ the internet. If you start publishing every random thing someone sends you in a newspaper, no matter how crazy or false, you'll get sued over it. This is not a First Amendment issue because people do not have an unconditional right to say anything they want while being totally immunized from consequences. They never have in the history of the country. This is about a special legal carve-out immunizing a specific class of communications platforms from legal liability in a way that we don't give to any other form of communication.
"you can't afford to take responsibility for the speech of every random person with an email address." Even Musk and Zuckerberg don't have pockets that deep!
Occasionally I try to imagine a world where § 230 did not pass. Every time it looks like a better world.
The intentionally addictive algorithms hurt us. The lack of accountability hurts us. The obscenely rich and arrogant social media oligarchs hurt us. We are ruining our children, we advantage the worst politicians, we may yet ruin the nation. And for what?
I’m a retired physical scientist so the depth of my historical knowledge isn’t what many of my Liberal Arts college classmates is. But my guess is each change in communication technologies was disruptive to existing social structures. Most of these changes likely significantly reduced the power/impact of those who were at the heads of aging structures of information dissemination and there were disruptions to society. Even what some may consider trite examples today likely had outsized impacts at the time. Some trite examples include: The only tool we know that Jesus had was speaking without amplification which naturally limited the spread of what he had to say. Diffusion of information, in the information theory sense, was limited by multiple factors for much of human existence, travel was limited to foot, then horse in some places but not others, and water in some places but not others. At various points in time and space writing and reading were developed but typically limited to a tiny subset of humanity. We all learned about the invention and impact of the printing press in elementary school and no doubt it was a highly disruptive technology, even when a majority of people couldn’t read. Catholic mass was in Latin until ca. 1965 yet only educated people in most of the world couldn’t read or write Latin. Radio was a disruptive technology. Would Hitler or FDR achieved power in the ways they did without radio? Then TV and then the internet. Today some are studying and discussing how AI is measurably changing neural pathways in the brains of people who use the most of it (e.g. recent David Brooks article in NYT, but I’ve read others). But are these changes greater or more important than the transition from entirely verbal communication and history transmission? I don’t read as fast as Brad DeLong but reasonably fast with some material, e.g. 150-200 pages/hour for light mystery type reading. But today when I read papers with lots of math my speed has decreased significantly from when I was in my 20’s to 40’s. My ability to conceptualize and visualize something like a Greens function is highly diminished. But I know how to get what I need on the internet (if it isn’t behind paywalls which wasn’t an issue until I retired). My internet search skills were obviously zero before the internet, when you spent hours a week in the library just reading Chemical Abstracts, which now would even be foreign to me. For instance I don’t understand why people watch videos or listen to podcasts for non-visual content. The transmission rate of information is a fraction of reading. I’ve tested myself with video podcasts where transcripts are available (most are on NYT, not all are on Substack or many other sources) and I’m at least 3x faster reading even when the speakers are good and don’t spend a lot of time in conversational habits with little information content. My kids have shown me parts of Tik Tock videos but I’ve never watched one on my own, nor have I ever used X/Twitter. So man has gone through a lot of disruptions in information transmission and a lot of it has always been junk. The current disruption is occurring faster than before, but I suspect every technological change in information transmission has been faster than earlier ones, in part because each bootstraps using the earlier technology. My parents didn’t let us watch “junk” TV growing up but I would see some at friend’s houses. They thought the 3-stooges disruptive to development, and I suppose I still read more and faster than most of my childhood contemporaries who had a steady diet of TV, but there is no evidence today that it shaped their views or response to the real world, or even at the time. Taking phones away from kids in school makes sense, but figuring out how to change internet behavior of adults is a difficult problem. There have always been people prone to conspiracy theories or similar things. Put bluntly there have always been, and will always be a lot of stupid people. Objectively the greatest, most powerful, and disruptive conspiracy theories of all time, organized religion, are still with us. The country is currently in the grips of the pure evil of Calvinist racism, but that has been with us since the Puritans, and certainly in the US South it had, and still does have, a huge impact, but is the impact greater today than the long history of US slavery for which most of the information to ordinary people was transmitted in person verbally at church. There are still people who believe in Noah’s Ark and other ludicrous fables. Has the internet changed that?
> The devastation of the America I grew up in is reversible; the spread of social media technology is not. The key is to use technological solutions to solve the technological problem that we’ve inadvertently created — to yoke the shoggoths and make them work for us once more, instead of against us.
Overly online, antisocial miscreants have existed since before the dawn of the world wide web, let alone modern social media. Eg, USNET and Richard Stallman-adjacent open source activism, not to mention various Libertarian and Communist communities. Moreover, we didn’t even need the internet to create insular, long distance communities as demonstrated in both right wing and left wing organizations throughout American, European, and global history; commonly centered around physical publications, exchange of letters, and occasional in person meetups. Eg, Vladimir Lenin and his network of nihilists, anarchists, and other socially destructive revolutionaries. See also the more recent networks of domestic terrorists (eg, Timothy McVeigh) and international ones (eg, Al Qaeda and Bin Laden).
To me this all comes down to the age-old questions of meaning and the challenges many face in accepting the banality of modernity as well described by Nietzsche's “The Last Man”. The internet and associated platforms are just our modern instantiating mechanisms and it is foolish to be overly focused on the tech aspect, let alone to believe that humanity can find the philosophical, dare I say spiritual, solution to the real metaphysical challenges of “man’s quest for meaning” and the multitude of pathetic, even at times destructive, behaviors that entails in modern world.
Instead, I believe we should be excited for the ever more fragmenting national and global social reality. The real dangers came from people coordinating on increasingly asinine right and left wing populist thought, facilitated by both the internet and broadcast media. Yet that world peaked in the mid 2010’s and we’ve been dealing with ever more fragmented media spaces. Eg, the demise of Twitter and the failure of BlueSky, Threads, Mastadeon, etc to recapture that global community aspect. Ideally our modern antisocial individuals can increasingly find small, niche communities for diversion. Many may even captivate themselves with artificial companions. Over time that should weaken the strength of antisocial political movements like MAGA and The DSA. Hence, reality can return to those actually living in it.
I think your proposed solutions here are simple, effective, and impossible.
> parents need to be encouraged to severely limit their kids’ use of phones at home.
Good luck with that. I don't see a good way around "but you're on your phone all day!" And even if you could keep everyone off social media until they're 18 you haven't solved this. Boomers were off social media until they were 50 and yet still lots of them get sucked in.
> the people who run social media platforms — Elon Musk, Mark Zuckerberg, etc. — can implement algorithmic tweaks that will direct people toward more moderate, reasonable content and away from extremist, incendiary content
We see how that's going already. Social media platforms face a powerful financial incentive not to do this. That's been enough so far to prevent Zuckerberg from taking any action. And that's when the CEO himself hasn't drunk the political cool aid...
I am highly sceptical that social media companies will ever want to filter content and downrank inflammatory content. The reason? It makes them more money. As Noah has pointed out in his article, studies show clearly that negativity draws more attention and retains it longer. Well, this is what the business model of these companies is built on. Why would they ever give up profits?
The biggest tragedy is that what we witness nowadays, and not only in the United States, mind, is that social media companies did not cause deliberately. They just sought a strategy that will make them the most money, and then made sure they will not be sued for it, it's that simple. Even if LLMs make it easier to filter negative content, I don't expect anyone will use them that way. All that investment in advanced algorithms needs to be recouped in some way, doesn't it?
So unfortunately, it is up to us to deal with the issue. Restricting children from being overly exposed to social media is a good step. Yet, it will initially lead to social stratification, as well-to-do parents are more likely to pay attention to their children's time spent on social media than those poor folks who need to work double shifts to make ends meet. It is the same old story as with education. Parents who pay attention to how their children do at school are more likely not to have serious financial problems than those who neglect education, and it reflects on their children's adult lives.
I appreciate your analysis, but I think you are too pessimistic when you conclude only private action can counter the ills of social media. Yes, social media companies will not police themselves, but we, the public, can police them. There are political solutions as well as private acts.
Section 230 enabled this cancer, Section 230 can be repealed. If you doubt repeal would much matter read Dan's comments in this thread, he makes the case so well there is no point repeating here.
This piece doesn't really address the underlying problem: companies found that due to the way people like to engage with technology they're more likely to engage with negative content than positive, so they engineer algorithms to maximize outrage and thus engagement.
Who tiktok's owners are will not change that basic fact. If anything as a private company it is less beholden to the growth mindset of Meta.
Not sure that there is a solution other than breaking up with social media. The techno optimists were wrong about it.
This article is excellent. The thesis is consistent with E.O. Wilson's observation that modern humanity is a mixture of "paleolithic emotions, medieval institutions, and god-like technology." Human nature remains vulnerable to manipulation through stoking the emotions of fear and anger and our norms and institutions have not yet adjusted to prevent bad actors from misusing the latest information technologies. Language is humanity's superpower and we need to protect our ability to engage in social thinking and reasoning; otherwise what makes our species unique and special (our ability to cooperate) will continue to erode.
Section 230 and the DMCA carve out a new, unique protection for publishers: they can publish and profit off of slander/libel/copyright infringement/lies, and they are fully protected. It's like eliminating slander/libel laws, and then being shocked that tabloids are destroying people's lives. The solution is to add a few extra words to qualify the protection: if the publisher is making money through attention (usually through ads), and it is choosing what to show users, then DMCA and/or Section 230 should not apply. Websites which just show content in chronological order, or stuff that's upvoted, or any other transparent way, would be completely unaffected. Basically, companies can't profit by shoving lies and outrage at us. It would have 0 effect on most websites, but would force all the major social media sites to change for the better.
I agree excluding algorithmic content promotion from Section 230 protection would do a lot of good, but I'm afraid it is too abstract to catch on politically. Repeal is much easier to explain, it does not require voters to understand Meta's business model. It only requires them to agree unaccountable social media is not their friend and the richest people in America do not deserve protection we do not give their local newspaper.
"The same rules for all publishers" is easy to understand, fair, and the best way to counter inevitable 1st amendment propaganda. "We had free speech before Section 230, we'll have free speech after it" cuts nicely through the BS.
I understand your concern about non-algorithmic bulletin boards, a legitimate but second-order problem. In the absence on Section 230 courts would decide if internet bulletin boards or forums are publications like newspapers or more similar to directories such as physical bulletin boards or the phone book. Criteria you mention "Websites which just show content in chronological order, or stuff that's upvoted, or any other transparent way" would later be relevant in court but first: Repeal the law.
Dean Baker proposes a new Defamation Takedown protocol like the DMCA. Basically, if someone reports a post as defamatory, the hosting site would need to take it down, just like it does with copyright infringement notices today. That might be more acceptable to the giants (since it creates yet another barrier to entry for small players, like the DMCA does), but it is far less effective than my proposal, which is simpler and harder to game. It also requires the person being defamed to monitor the entire internet and request takedowns, which can be days or weeks later. Just like Youtube makes reporters jump through hoops to prove they own the copyright (unless you are another giant corporation, then they just trust you), I think this wouldn't really be very effective unless crafter very carefully. The DMCA works since giants on both sides of the problem worked out a system they felt was fair--there's no giant representing the people being defamed, so the law will be gutted as it is being created. And my approach halts most disinformation and misinformation as well (it can't get amplified) even if it's not defamatory.
I view algorithmic curation as the problem, not the solution. When Facebook was just things your friends posted shown in chronological order, it was fine. Twitter becoming the "global commons" and putting everyone in the same "room" is what caused the current escalation and polarization. Humans can't handle interacting with so many people/information sources; algorithms give us the illusion of being able to surmount this limitation, while actually taking advantage of it by selecting for "engagement". Algorithmic curation should be regulated into the ground like cigarettes. Turn the internet back into a bunch of separate "rooms", and things will go back to normal.
Agree! My concern is political: "algorithms" is abstract. Meta's business model is not widely understood.
I wonder if it is more feasible to go after special Section 230 protections big platforms enjoy. "Elon Musk should be just as accountable as your local paper" seems an easier political sell and would have similar effect.
I love this post and agree with the diagnosis and prescriptions for the most part but I think it's somewhat missing what's happening with the olds. There's plenty of rage incubating in elder gen X and younger boomers that seems to be spurred on somewhat by the internet, but much more by cable news, talk radio/podcasts, or just the watercooler. I was just discussing this with my local middle millennial friends and we came to the conclusion that pretty much all of the political extemism in our families comes from the 45-70 crowd. That's probably because we live in a more rural state, no critical mass of young antifas or groypers here, but I don't think that means this facet of the problem can be dismissed, given how the fractious fault lines of rural blue collar vs urban elite seems to suffuse, and, indeed, drive a large part of our political conflicts. I don't know how to fix that hard problem, short of firebombing Fox, MSNBC and CNN headquarters, lol. Even then, these folks have had their minds poisoned by this stuff for 30 years, I don't know if there's a way of unringing that bell.
Talk radio has been screwing with Republican minds since at least Rush Limbaugh and the 1990s...
Surely it dates back to Father Coughlin in the 1930s when he had a radio audience of 25% of America .
This is very much on point. Older people watch Fox and MSNBC, not kids, while You Tube and other platforms host longer form podcasts whose hosts are extreme. Noah, your analysis is otherwise spot on, but on remedies, I see no hope for Twitter or BlueSky moderating their content -- they feed on and react to each other. There is no "moderate" Cronkite social media platform. As Sylvilagus says, I don't think there's a way to unring the bell, put toothpaste back in the container, pick your metaphor. I don't know what the answer is. But if we can't find a way for moderate forces who represent the middle of the country, where most people still live politically, the vicious cycle only will continue
As a Progressive it seems to me there is a false equivalence between the positions of the Right--as reflected by FOX News, X, Newsmax, etc.--and the Left, as reflected in primarily MSNBC. Mainly because of the sheer volume of misinformation and disinformation coming from the Right compared to the Left.
Yes, there is plenty of Trump-is-bad confirmation bias to be had on MSNBC, the Nation, Pacifica, etc. But it is all based--at least mostly--in facts; e.g. DOGE's gutting of IRS personnel will result in a $500bn revenue shortfall in FY2025 (which is extremely significant, btw). This is my opinion, of course, and is open to debate.
But to regain national moderation, we must be able to debate TRUTHS; as in Congressional Budget Office type "truths" based on non-sexy, nonpartisan bean counting. To that end, we must legislate that anything purporting to be "news"--as opposed to opinion--must be based in actual journalistic standards, like 2+ sources, actual links to actual studies, timely retractions when necessary, etc. And perhaps also be subject to relaxed libel standards, to discourage fact-free mudslinging.
To keep our Democracy functional, voting citizens need to know what's actually going on, and ensuring we have a 4th Estate that keeps to the facts is essential to that end.
"But to regain national moderation, we must be able to debate TRUTHS; as in Congressional Budget Office type "truths" based on non-sexy, nonpartisan bean counting. To that end, we must legislate that anything purporting to be "news"--as opposed to opinion--must be based in actual journalistic standards, like 2+ sources, actual links to actual studies, timely retractions when necessary, etc"
So you want Trumps office of "truth" to decide what counts as truth?
Of course not, you want your side to decide. But that's the problem, that's not how politics works. Whatever power you give to the government the other side will eventually control.
Government can't be the arbitrator of truth. Really that's the whole point of the 1st amendment.
All that being said, yes I agree that media should focus more on fact based reporting. But if there hadn't been such a huge liberal tilt in the MSM, companies like Fox wouldn't have had a chance in the first place.
For decades we had the Fairness Doctrine. Which wasn't perfect. But it was something, until Reagan sixed it.
And FYI, FOX News is in no way fringe for GOP voters. As per Ipsos polling last November, it is where 85% of all conservative voters go for some/most/all of their news.
You might disagree, but there is actually something to actual journalistic standards. Like having 2+ sources to corroborate something. Which is a shit ton better than the, "...People are saying..." non-corroborated bullshit you routinely get on FOX News, which is shorthand for "We just made this story up out of whole cloth; but need to pitch it as if it actually has legs."
Maybe the 1st Amendment will forever prohibit the creation of any meaningful "Truth in Reporting" law. But considering that only months ago a Presidential Candidate proclaimed on national TV that: A) "Haitians are eating the dogs and cats in Springfield," and B) "Democrats murder (abort) babies after birth." It's sorely needed.
The fairness doctrine wouldn't apply to Fox news or the internet anyway, it only applied to government owned airwaves.
And even then it was still of questionable constitutionality.
But the point remains, if you did have that government department, TRUMP would be in charge of it now!!!
You're right about false equivalance b/w MSNBC and Fox, but still MSNBC is to the left of the median voter. And I couldn't agree more about CBO. If the scorekeeper/referee is abandoned, the "game" is over
There might not be moderate ones, but there are ones that are at least neutral by default; you probably won't see crazy stuff on Facebook if you don't look for it or have friends that do, and Reddit's siloing of things by topic also helps you avoid far-left or far-right politics where it's not welcome.
Ultimately, if people want to hear what [insert extremist here] has to say, I don't really think the rest of the world should try to tell them they can't - but we also don't have to stick it in the faces of people who don't want to hear about it.
(On a related note, no individual has an obligation to debate Nazis or other people holding terrible positions, but it often helps if someone does - for example, www.talkorigins.org has an extensive archive of rebuttals to various creationist arguments.)
The degree to which retirees are marinating in Fox / Newsmax / OAN is really disturbing. I run into people in the 65+ set all the time who _sincerely believe_ that Trump is only rounding up illegal immigrant criminals, and that the budget will only take benefits away from illegals. Because that's what they're hearing.
There has been research showing that mainstream media coverage is heavily influenced by social media. Fox, MSNBC, etc. are just the last stage in the transmission chain to the “olds.”
It is definitely true that new gods have exerted some nontrivial control over behavior of the old gods, but they aren't fully in the driver seat, imo, otherwise the tech right would have had no issues with their poasting to maintain maga dominance plan, but instead middle America old person maga wrested the reins back from them pretty quickly without much trouble.
Angry CNN retirees? Is that a thing?
Don't the angry liberal retirees watch MSNBC rather than CNN?
CNN is arguably right-of-center. At least on Economic issues. On social issues, like abortion, they and WaPo, NYT trend liberal. All of these press organs still sometimes call Trump out on his lying. Which, unfortunately, is often. As such they're centrist; at least compared to FOX, Sinclair, Newsmax, OANN; where misinformation and disinformation are standard output.
MSNBC is the only national network with liberal programming. Unless you count Late Night comedy shows like John Stewart, Jimmy Kimmel, etc.
My dad has definitely become an angry liberal at this point...
Musk likes Twitter the way it is and he is primarily responsible for turning it into the toxic hellhole that it is. Just take the oxygen out of the room. It’s a bad drug, get off of it and use the extra hours of your life to do something productive.
You're suggesting we expose all Twitter users to vacuum? :D
Musk and Zuckerberg /could/ tweak their algorithms. But /why/ will they? Is there something society can reasonably do that would make that in their interest?
The answer is probably no, because it's less profitable.
Metaphorically speaking. In space no one can hear you tweet.
No. That's a bad solution. Much more feasible and equally effective to displace oxygen in the room with a heavier gas like CO2.
Dem na kepe da air welwala. (In Belter)
This is a point I think Noah is too optimistic on. It’s probable that the Musks of the world enjoy having the power to steer discourse by controlling those platforms and I’m skeptical they can be convinced to intentionally fight the trends that enrich and empower them.
I was very surprised to see this as a solution. I didn't think Noah was this naive.
Twitter was always a hellhole. The toxic ingredients have always been there--pseudonymity that removes personal stakes from posting BS, an algorithmic feed that rewards emotional content, and the inability to express nuance due to length restrictions.
I encountered Twitter a matter of months after it was created, and immediately had an allergic reaction to it. The thing I always said about it was: "X spends too much time on Twitter" is a true statement for all values of X. The name change felt ironic.
Equating Mamdani to right wing conspiracy theorists and hate slingers is extraordinarily off base. The temperament, the tenor, and the message from Mamdani is measured and reasonable. I have seen him speak multiple times about not wanting to disclaim the “globalize the intifada” language and I truly don’t believe he sees it as a call to destroy the state of Israel. He is however an anti-Zionist and believes the state of Israel should be non-religious.
He seems like one of your “normal” and decent Americans. This can’t be said of any of the toxic right wingers you identify in this piece.
Normal, decent Americans don’t go around speaking of getting elected in order to “seize the means of production”. He’s a smooth talking politician with communist and anti Zionist ideas and bad economic thinking with no experience running anything bigger than a local campaign. If he put his economic ideas into practice it would be as bad as what Trump is doing with his ignorant economics.
Speak for yourself. You don’t get to decide what “normal, decent Americans” get to think. If you actually talked with some of them, you’d be surprised with what they actually believe
https://www.politifact.com/article/2025/jul/03/seizing-means-production-Zohran-Mamdani/
PolitiFact | In Context: Zohran Mamdani’s use of phrase ‘seizing the means of production.’ What’s it mean?
Again, I don’t see any comparison whatsoever between Trump and Mamdani. I guess you do but where’s the evidence?
Trump and Mamdani are alike in believing economic ideas that have been proven wrong by economists and by actual experience: tariffs hurt the country implementing them, rent control causes reduced quality and amount of rentals available, and raise rents in nearby uncontrolled apartments, seizing the means of production leads to economic collapse and death, deporting workers means reduced output and higher prices, making buses free doesn’t improve transportation, getting rid of police results in more crime.
But they both smooth talk a lot and post on social media expounding their belief in these bad assumptions, and both are skilled at getting ignorant voters to believe that they have reasonable ideas. These voters will get what they voted for, and the rest of will suffer along with them. At least Mamdani isn’t as bad as Trump since affected people can leave NYC without an exit tax, unlike leaving the US.
I appreciate your perspective on the economic issues and I too disagree with many of the policy proposals that Mamdani has put forth. But that still in no way equates him with hateful / malicious / greedy MAGA right wing personalities in my book.
I believe Mamdani is a skilled communicator and a genuine person with good intentions. He also seems willing to listen and change his mind, and we have already seen him moderate.
The “other side” just gets more extreme and incoherent with every passing tweet.
It was only a brief statement of what he intends to do, so Politifact says it’s “false”. This is the problem when people say they want to police social media for misinformation, when any politically biased group can claim something a declared socialist says on video supporting implementing socialism is “disinformation”.
All the MAGA fools who didn’t believe what Donald Trump said in his campaign are now regretting their votes as their spouses or friends get rounded up and deported. If a crazy guy says crazy stuff to get elected, you need to expect the government to then do crazy stuff.
Tenor is important! Mamdani is not whipping people into an angry frenzy. Confusing him with Lenin is a category error.
I agree. Noah claims that Mamdani "... defended the phrase “globalize the intifada”". Please, Noah, link us to a reference where Mr Mamdani uses and endorses that phrase himself. In the Meet The Press interview I watched those words were put into his mouth and he basically declined to engage, very skillfully evading a direct answer:
"My concern is to start to walk down the line of language and making clear what language I believe is permissible or impermissible takes me into a place similar to that of the president, who is looking to do those very kinds of things, putting people in jail for writing an oped. Putting them in jail for protesting. Ultimately, it's not language that I use. It's language I understand there are concerns about. And what I will do is showcase my vision for this city through my words and my actions."
https://www.nbcnews.com/meet-the-press/meet-press-june-29-2025-n1312337
It was a gotcha join-me-in-the-pigpen question and he dodged it.
What's gotcha about it? It should be really easy to denounce a phrase like that. His refusal to do so is REALLY troubling
Agreed. I see Mamdami’s stances as well within moderate centrist values in other democratic societies. The fact that he is considered extreme in the U.S. context is a product of evolutionary rightward skewing of the center of U.S. politics since about 1980. That center is far to the right of most Western democracies at this point. And I realize I’m probably being overly generous in including the U.S. in the remaining group of democratic polities.
Yup. That was my one issue with an otherwise good piece. I like Mamdani, and so many otherwise smart centrist Dems I know are having a full-blown moral panic about him.
This article would be a lot more convincing if it grappled more substantively with the systemic problems that are fueling social/political disorder, instead of simply attributing it to the malign influence of social media (this would be akin to blaming the printing press for the chaos of the Protestant Reformation. Noah’s analysis is weakened by this focus on technological determinism, which presumes that American society was fundamentally healthy prior to the rise of social media (long story short, it wasn’t: social media simply ripped the bandaid off the wound).
One example is with Noah’s weird centrist obsession with the pro-Palestine protesters. I understand the need to punch left, both out of a genuine expression of centrist bonafides as well as a way of staying evenhanded, but there is a consistent aversion in centrist media towards grappling with the humanitarian crisis in Gaza. Even Andrew Sullivan, who is nobody’s idea of a woke scold, has conceded on this point and recognized that Israel is not the liberal-democratic underdog everyone wishes it was. This problem goes beyond the Netanyahu regime and reflects on the fundamental nature of the Israeli project, whose legitimacy is predicated on a two-state solution that is (and likely always was) fanciful. Maybe, just maybe, those immature pro-Palestine protesters are justifiably enraged and upset US military support for a regime that bombs hospitals and weaponizes famine as maliciously as any 20th century autocrat.
I’d even extend this argument to immigration. I’m no nativist, and I find the American right to be a lot more xenophobic and bigoted towards immigrants than I initially expected (I thought the “multiracial working class” coalition Trump assembled in 2024 would’ve solved the matter, but apparently MAGA is even stupider than I previously thought). But it’s clear to everyone that the old centrist regime of liberalizing immigration flows was hopelessly utopian: while immigrants do indeed contribute to the social and economic diversity of America, the social dislocation that mass migration generates (much of it rooted in the bigoted reactions of native-born Americans) is a cost that policymakers and economic theorists have consistently underestimated. And this is doubly true for other core “economic truths” like uninhibited free trade.
Another objection to this otherwise well-written piece is that Noah is assuming that the social stability of the mid-twentieth century is the norm, and that modern instability is an aberration. But it is in fact quite the opposite. Trump’s antics, however authoritarian they may be, are not foreign to the American historical tradition: populist demagogues like Andrew Jackson and Richard Nixon have long relied on such tactics, as well as quasi-genuine egalitarian sentiment, to win elections. Inequality and the paradox of poverty amidst plenty are inherent products of the market economy, as are the rise of politically-powerful oligarchs (Carnegie, Mellon, Rockefeller, etc.). This is not a criticism of America, but rather a humbling observation: some level of sociopolitical instability is inevitable, and efforts to squash it out through paternalistic nannying or indoctrination are bound to fail.
Finally, I’d be a lot more sympathetic to Noah’s arguments if he was more frank about the failures of the centrist gatekeepers whose fall from power he so eagerly laments. One of the reasons why those gatekeepers are increasingly less relevant is because the mass public has been enraged by the perception (real or otherwise) that our social elites are incredibly unaccountable. One reason the rise of Trump is so terrifying is that so many people have simply begun to tune out Trump’s autocratic antics, making it even easier for him to erode norms on corruption and good governance. It would have been much better to nip things in the bud by vigorously enforcing anti-corruption laws, even against our social elites, but now the very idea of good governance is considered laughable.
This comment has to be Ai. "Another objection to this otherwise well written piece..." This voice sounds neither American nor human.
That is literally the tone and phrasing that Noah Smith uses often himself.
Though the destruction of online credibility due to AI as evidenced by this response complicates Noah's thesis here.
Damn; wrong on both counts.
That doesn't make it wrong.
“One of the reasons why those gatekeepers are increasingly less relevant is because the mass public has been enraged by the perception (real or otherwise) that our social elites are incredibly unaccountable.”
I disagree. I think the centrist gatekeepers lost power because of the social media attack which they can not defend against.
What you have articulated about unfettered immigration and free trade, coupled with increased corruption and lack of accountability, are all symptoms of late stage capitalism. This is the root problem.
Noah's presumption is that "technology can solve everything", so don't be surprised when he doesn't include any ideas that "we need to solve underlying social issues first". https://x.com/Noahpinion/status/1941260882224480439
I think you misunderstand some of Noah's thought. It would be more accurate to say he thinks technology determines everything, and the everything would still be a bit of a stretch.
"Every technology changes the social contract. Cars changed the social contract. TV changed the social contract."
https://x.com/Noahpinion/status/1883250670343746010
I think you might have to demonstrate how you "solve underlying social issues" to be convincing about doing them first. I can't think of many a where someone set out to "solve underlying social issues first" and succeeded. The best examples I can think of are the MLKs approach to civil rights, but such examples are few and far between, and even then there's a significant argument that this was dependent on changes in society that were started by technological changes.
Re: "It would be more accurate to say he thinks technology determines everything"
He did mention in his "about" part of his blog post, when he mentions why his philosophy is "techno-optimism" (https://www.noahpinion.blog/about):
"A big example is technology. Noahpinion is explicitly a techno-optimist blog, meaning that I believe technological innovation is going to be more important and transformative — and usually, though not always, better for the human race — than most people realize. The recent advances in vaccines that helped us get through the Covid pandemic, and the advances in green energy that promise to help us avoid the worst of climate change, are cases in point."
One of his blogposts was also about denying pushbacks against technological progress, and he used "decel" (a term from effective accelerationists: https://www.noahpinion.blog/p/dont-be-a-decel?utm_source=publication-search)
But maybe your point of view is more fit with the ideas of Noah though.
Re: I think you might have to demonstrate how you "solve underlying social issues" to be convincing about doing them first.
In retrospect, I was rehashing a bit some criticisms of the Industrial Revolution by Ted K, that rather finding underlying issues to solve first, scientists and business leaders just innovated out of it - obesity is one example, Americans are now less affected from it not through government campaign for healthy eating, but through Ozempic; which benefits drug companies like Novo Nordisk the most!
I agree with you about the problem, but the solutions that you propose will not succeed. We can take measures to reduce the amount of social media in children's lives, but what about adults? There are two problems— How would you incentivize social media companies to change their algorithms? Second, any government interference in algorithms is a clear violation of the First Amendment. To summarize, this is an important issue that needs much more thought.
To get Americans out in the real world interacting with each other positively, one could require a year of public service from every young citizen and also facilitate supported full time volunteer work by older Americans.
Face to face interaction is essential to human communication. Non-verbal cues and communication are entirely missing from the online forums and those subtle and sometimes not so subtle bits of social information help tremendously in moderating human social interaction. Anthony Bourdain put it thus - "Eat at a local restaurant tonight. Get the cream sauce. Have a cold pint at 4 o’clock in a mostly empty bar. Go somewhere you’ve never been. Listen to someone you think may have nothing in common with you. Order the steak rare. Eat an oyster. Have a negroni. Have two. Be open to a world where you may not understand or agree with the person next to you, but have a drink with them anyways. Eat slowly. Tip your server. Check in on your friends. Check in on yourself. Enjoy the ride." Sometimes, having been on the planet for a good long while and having seen some of the world brings a wisdom which is lacking in the young, energetic but still undeveloped mind. I should know.
"How would you incentivize social media companies to change their algorithms? Second, any government interference in algorithms is a clear violation of the First Amendment."
IMO, that's not so simple and this is where Section 230 reforms could possibly come into play. Media companies are clearly protected from liability if they're just hosting third party content. OTOH, the use of algorithms to curate content could subject platforms to liability for clearly harmful content if Section 230 did not exist.
I mostly agree with Mike Masnik's argument that Section 230 is about clarifying and codifying free speech protections that already exist in the Constitution rather than creating new protections, but I'm not quite as absolutist about this as he is.
I'm sympathetic to Masknic's more expert critics (eg Kosseff) who say 230 is overbroad and creates protections above and beyond what's in the constitution, particularly when platform owners are reckless and intentionally create algorithms they know or should know will cause harm just for the purpose of increasing profit.
As I understand it, Kosseff is largely aligned with Masnik and is no crank, but he believes adjusting 230 to require a bit more transparency and accountability in exchange for 230's liability protection would result in a better overall result for society and would still pass constitutional muster.
Use algorithms and lose Section 230 protection appeals to me.
Or at least sufficiently complex and non-transparent algorithms. "Display stuff in reverse chronological order" is technically an algorithm, too.
Most importantly, spam filters are an algorithm and you need those.
Banning algorithms outright would not clearly violate the first amendment.
That's true and it's a great idea.
Surely you can't have social media without an algorithm?
What would be reasonable though would be to require platforms to make their algorithms public knowledge.
Algorithm reform instantly becomes highly technical. A reason it is better to just repeat Section 230 outright.
That said I think algorithm is easy to define in this context: They use an individual's history (search, sites visited, etc.) as information to feed content.
Take away the individual customization feed and the problem is far smaller, as are the profits.
Not only is it clearly against the First Amendment, but the public will reject it. We saw what happened when the Biden administration tried to do this to suppress the embarrassing laptop story and to promote its COVID propaganda.
I grant interference in algorithms raises first amendment questions, outright banning algorithms is less clear.
Today the public would reject it. Tomorrow's may not.
This would, in fact, be a 1A violation, and also suffer from the problem of it being undefinable what constitutes an "algorithm."
Define "COVID propaganda" please.
That was poorly written. When I said "promote its COVID propaganda" what I was referring to was actually when the administration pressured social media to censor people who opposed their messaging on COVID.
You should listen to the Ezra Klein podcast with Chris Hayes about Mamdani as an early example of a political campaign native to the vertical video format. They seemed to agree that as a medium TikTok/Reels etc don't reward outrage bait quite so much as Twitter/X do.
For what it's worth, they seemed to be relatively sanguine about Mamdani himself; and I think one of their points was roughly that his cautious defense of the "infatida" phrasing actually worked well in TikTok style media, even though it's obvious fodder for attack ads and negative memes in other media forms like TV and Twitter.
Good piece, and I very much agree with the diagnosis. I'd like to propose an alternative solution, though, one that doesn't rely on the goodwill and responsible moderation of actors like Elon or Zuck whose profits are built on "engagement" and have a very strong financial incentive to platform and spread controversial content that attracts eyeballs.
Simply put, we should repeal section 230 of the Communications Decency Act. Not reform it, completely repeal it. Many of the people who attack § 230 are interested in reforming it for partisan ends and think they can keep the current internet more or less the same, while simply advancing their worldview and suppressing their opponents'. I see it very differently: a full repeal of § 230 is a deathblow for social media and user-generated content on the internet, one which cannot easily be recovered from or worked around by making superficial tweaks to comply with the letter of laws and playing an endless game of whack-a-mole with regulators.
Platforms will not be able to function if they're held legally liable for whatever their users post; the litigious nature of the US legal system will crush them in short order. It means a return to a world where the ability to publish and widely disseminate information is gatekept by having skin in the game: you need financial resources to set up your own publishing and marketing infrastructure, or you need to persuade someone who does have those resources that you're trustworthy enough that they're comfortable taking responsibility for the consequences of your output. No more of the current world where any anonymous troll can sign up a free account and get access to a world-class broadcasting system.
The flaw I see with this proposal -- as with so many other proposals to reform our dysfunctional country -- is simple, though: the politicians we have elected like the division. They like the discord. They like the fear, the loathing, the utter hatred towards our fellow citizens, because it's so much easier to motivate people to support you by giving them villains to hate and boogeymen to fear than by articulating a positive vision someone might actually expect you to deliver on. The massive amounts of money that social media companies make and their ability to buy politicians is also a significant problem, perhaps never demonstrated more clearly than by Trump's abrupt about-face on TikTok. But I think you could take the money completely out of the system and the bloc voting paradigm would still be enough that Congress would never act on this.
This comment and your reply below are the best in this thread. I would really appreciate it Noah and others with big megaphones would take up whether the benefits of Section 230 even remotely approach the harms.
If section 230 was repealed, then Twitter and Instagram might block everything except cat videos to avoid getting sued, but all the political shit posting would just move to TikTok since the owners are in China and can’t be successfully sued for liable. You can say we could just outlaw it, but we already did and Trump is just ignoring the law.
TikTok is an American company that has very significant technical operations in America. They're absolutely vulnerable to being sued in US courts. Bytedance could try to address this by simply withdrawing all presence from the US, but that would render the service very slow, kludgy and difficult to use; modern video services need CDNs and globally distributed datacenters to offer a high-volume, performant service. US courts could also just order the service blocked by American ISPs, a remedy that's been used before in cases where a foreign site is hosting something in violation of US law. Yes, people can still use VPNs and the like to access banned services, but it's slow, clunky and most people won't bother. And once you start depriving a social network of its audience, it will collapse very quickly since network effects are what sustain them.
Plus Trump loves his Truth Social and wouldn’t sign any legislation rescinding section 230, so even if it’s possible it won’t be done until 2029.
The vast majority of speech on social media would be protected under the first amendment regardless of whether platforms are liable for unprotected speech on them.
Yes, it would. But that's not the point. The point is that defending that speech as protected is expensive. Very expensive. And with the sheer volume of stupid things people post on the internet, you're going to lose a lot of those cases, too, because it's not protected speech: it's easily-provable libel.
Many of these individual posters could be sued right now, successfully, but they aren't because the likelihood of recovering enough money from them to make the suit worthwhile is nearly zero. But if a multi-billion-dollar corporation is the one held legally responsible for the speech, the lawyers will come out like a swarm of piranhas, because winning a jury judgment against Facebook could set you up for life.
At the moment, this doesn't happen because § 230 makes these lawsuits utterly futile. The people disseminating the speech who have actual money to take are completely and unconditionally immunized from responsibility for it, so lawyers know better than to bother, and the only lawsuits you see are the purely performative ones where some rich person wants to make an example of an individual and doesn't care about the cost. Take that shield away, though, and suddenly these lawsuits become very viable. Even if the vast majority fail because they're going after protected speech, it doesn't matter: companies can't afford to defend against all of them, and they really can't afford to lose many of them. It's about risk calculus on the balance sheet: you can't afford to take responsibility for the speech of every random person with an email address.
Edit: the follow-up to note here is that what I'm proposing is literally how it works in every medium _except_ the internet. If you start publishing every random thing someone sends you in a newspaper, no matter how crazy or false, you'll get sued over it. This is not a First Amendment issue because people do not have an unconditional right to say anything they want while being totally immunized from consequences. They never have in the history of the country. This is about a special legal carve-out immunizing a specific class of communications platforms from legal liability in a way that we don't give to any other form of communication.
"you can't afford to take responsibility for the speech of every random person with an email address." Even Musk and Zuckerberg don't have pockets that deep!
Occasionally I try to imagine a world where § 230 did not pass. Every time it looks like a better world.
The intentionally addictive algorithms hurt us. The lack of accountability hurts us. The obscenely rich and arrogant social media oligarchs hurt us. We are ruining our children, we advantage the worst politicians, we may yet ruin the nation. And for what?
Love you Noah but this plan sounds more like a wishful thinking and less like a ‘plan’, sorry.
I’m a retired physical scientist so the depth of my historical knowledge isn’t what many of my Liberal Arts college classmates is. But my guess is each change in communication technologies was disruptive to existing social structures. Most of these changes likely significantly reduced the power/impact of those who were at the heads of aging structures of information dissemination and there were disruptions to society. Even what some may consider trite examples today likely had outsized impacts at the time. Some trite examples include: The only tool we know that Jesus had was speaking without amplification which naturally limited the spread of what he had to say. Diffusion of information, in the information theory sense, was limited by multiple factors for much of human existence, travel was limited to foot, then horse in some places but not others, and water in some places but not others. At various points in time and space writing and reading were developed but typically limited to a tiny subset of humanity. We all learned about the invention and impact of the printing press in elementary school and no doubt it was a highly disruptive technology, even when a majority of people couldn’t read. Catholic mass was in Latin until ca. 1965 yet only educated people in most of the world couldn’t read or write Latin. Radio was a disruptive technology. Would Hitler or FDR achieved power in the ways they did without radio? Then TV and then the internet. Today some are studying and discussing how AI is measurably changing neural pathways in the brains of people who use the most of it (e.g. recent David Brooks article in NYT, but I’ve read others). But are these changes greater or more important than the transition from entirely verbal communication and history transmission? I don’t read as fast as Brad DeLong but reasonably fast with some material, e.g. 150-200 pages/hour for light mystery type reading. But today when I read papers with lots of math my speed has decreased significantly from when I was in my 20’s to 40’s. My ability to conceptualize and visualize something like a Greens function is highly diminished. But I know how to get what I need on the internet (if it isn’t behind paywalls which wasn’t an issue until I retired). My internet search skills were obviously zero before the internet, when you spent hours a week in the library just reading Chemical Abstracts, which now would even be foreign to me. For instance I don’t understand why people watch videos or listen to podcasts for non-visual content. The transmission rate of information is a fraction of reading. I’ve tested myself with video podcasts where transcripts are available (most are on NYT, not all are on Substack or many other sources) and I’m at least 3x faster reading even when the speakers are good and don’t spend a lot of time in conversational habits with little information content. My kids have shown me parts of Tik Tock videos but I’ve never watched one on my own, nor have I ever used X/Twitter. So man has gone through a lot of disruptions in information transmission and a lot of it has always been junk. The current disruption is occurring faster than before, but I suspect every technological change in information transmission has been faster than earlier ones, in part because each bootstraps using the earlier technology. My parents didn’t let us watch “junk” TV growing up but I would see some at friend’s houses. They thought the 3-stooges disruptive to development, and I suppose I still read more and faster than most of my childhood contemporaries who had a steady diet of TV, but there is no evidence today that it shaped their views or response to the real world, or even at the time. Taking phones away from kids in school makes sense, but figuring out how to change internet behavior of adults is a difficult problem. There have always been people prone to conspiracy theories or similar things. Put bluntly there have always been, and will always be a lot of stupid people. Objectively the greatest, most powerful, and disruptive conspiracy theories of all time, organized religion, are still with us. The country is currently in the grips of the pure evil of Calvinist racism, but that has been with us since the Puritans, and certainly in the US South it had, and still does have, a huge impact, but is the impact greater today than the long history of US slavery for which most of the information to ordinary people was transmitted in person verbally at church. There are still people who believe in Noah’s Ark and other ludicrous fables. Has the internet changed that?
> The devastation of the America I grew up in is reversible; the spread of social media technology is not. The key is to use technological solutions to solve the technological problem that we’ve inadvertently created — to yoke the shoggoths and make them work for us once more, instead of against us.
Overly online, antisocial miscreants have existed since before the dawn of the world wide web, let alone modern social media. Eg, USNET and Richard Stallman-adjacent open source activism, not to mention various Libertarian and Communist communities. Moreover, we didn’t even need the internet to create insular, long distance communities as demonstrated in both right wing and left wing organizations throughout American, European, and global history; commonly centered around physical publications, exchange of letters, and occasional in person meetups. Eg, Vladimir Lenin and his network of nihilists, anarchists, and other socially destructive revolutionaries. See also the more recent networks of domestic terrorists (eg, Timothy McVeigh) and international ones (eg, Al Qaeda and Bin Laden).
To me this all comes down to the age-old questions of meaning and the challenges many face in accepting the banality of modernity as well described by Nietzsche's “The Last Man”. The internet and associated platforms are just our modern instantiating mechanisms and it is foolish to be overly focused on the tech aspect, let alone to believe that humanity can find the philosophical, dare I say spiritual, solution to the real metaphysical challenges of “man’s quest for meaning” and the multitude of pathetic, even at times destructive, behaviors that entails in modern world.
Instead, I believe we should be excited for the ever more fragmenting national and global social reality. The real dangers came from people coordinating on increasingly asinine right and left wing populist thought, facilitated by both the internet and broadcast media. Yet that world peaked in the mid 2010’s and we’ve been dealing with ever more fragmented media spaces. Eg, the demise of Twitter and the failure of BlueSky, Threads, Mastadeon, etc to recapture that global community aspect. Ideally our modern antisocial individuals can increasingly find small, niche communities for diversion. Many may even captivate themselves with artificial companions. Over time that should weaken the strength of antisocial political movements like MAGA and The DSA. Hence, reality can return to those actually living in it.
I think your proposed solutions here are simple, effective, and impossible.
> parents need to be encouraged to severely limit their kids’ use of phones at home.
Good luck with that. I don't see a good way around "but you're on your phone all day!" And even if you could keep everyone off social media until they're 18 you haven't solved this. Boomers were off social media until they were 50 and yet still lots of them get sucked in.
> the people who run social media platforms — Elon Musk, Mark Zuckerberg, etc. — can implement algorithmic tweaks that will direct people toward more moderate, reasonable content and away from extremist, incendiary content
We see how that's going already. Social media platforms face a powerful financial incentive not to do this. That's been enough so far to prevent Zuckerberg from taking any action. And that's when the CEO himself hasn't drunk the political cool aid...
I am highly sceptical that social media companies will ever want to filter content and downrank inflammatory content. The reason? It makes them more money. As Noah has pointed out in his article, studies show clearly that negativity draws more attention and retains it longer. Well, this is what the business model of these companies is built on. Why would they ever give up profits?
The biggest tragedy is that what we witness nowadays, and not only in the United States, mind, is that social media companies did not cause deliberately. They just sought a strategy that will make them the most money, and then made sure they will not be sued for it, it's that simple. Even if LLMs make it easier to filter negative content, I don't expect anyone will use them that way. All that investment in advanced algorithms needs to be recouped in some way, doesn't it?
So unfortunately, it is up to us to deal with the issue. Restricting children from being overly exposed to social media is a good step. Yet, it will initially lead to social stratification, as well-to-do parents are more likely to pay attention to their children's time spent on social media than those poor folks who need to work double shifts to make ends meet. It is the same old story as with education. Parents who pay attention to how their children do at school are more likely not to have serious financial problems than those who neglect education, and it reflects on their children's adult lives.
I appreciate your analysis, but I think you are too pessimistic when you conclude only private action can counter the ills of social media. Yes, social media companies will not police themselves, but we, the public, can police them. There are political solutions as well as private acts.
Section 230 enabled this cancer, Section 230 can be repealed. If you doubt repeal would much matter read Dan's comments in this thread, he makes the case so well there is no point repeating here.
This piece doesn't really address the underlying problem: companies found that due to the way people like to engage with technology they're more likely to engage with negative content than positive, so they engineer algorithms to maximize outrage and thus engagement.
Who tiktok's owners are will not change that basic fact. If anything as a private company it is less beholden to the growth mindset of Meta.
Not sure that there is a solution other than breaking up with social media. The techno optimists were wrong about it.
This article is excellent. The thesis is consistent with E.O. Wilson's observation that modern humanity is a mixture of "paleolithic emotions, medieval institutions, and god-like technology." Human nature remains vulnerable to manipulation through stoking the emotions of fear and anger and our norms and institutions have not yet adjusted to prevent bad actors from misusing the latest information technologies. Language is humanity's superpower and we need to protect our ability to engage in social thinking and reasoning; otherwise what makes our species unique and special (our ability to cooperate) will continue to erode.
Section 230 and the DMCA carve out a new, unique protection for publishers: they can publish and profit off of slander/libel/copyright infringement/lies, and they are fully protected. It's like eliminating slander/libel laws, and then being shocked that tabloids are destroying people's lives. The solution is to add a few extra words to qualify the protection: if the publisher is making money through attention (usually through ads), and it is choosing what to show users, then DMCA and/or Section 230 should not apply. Websites which just show content in chronological order, or stuff that's upvoted, or any other transparent way, would be completely unaffected. Basically, companies can't profit by shoving lies and outrage at us. It would have 0 effect on most websites, but would force all the major social media sites to change for the better.
I agree excluding algorithmic content promotion from Section 230 protection would do a lot of good, but I'm afraid it is too abstract to catch on politically. Repeal is much easier to explain, it does not require voters to understand Meta's business model. It only requires them to agree unaccountable social media is not their friend and the richest people in America do not deserve protection we do not give their local newspaper.
"The same rules for all publishers" is easy to understand, fair, and the best way to counter inevitable 1st amendment propaganda. "We had free speech before Section 230, we'll have free speech after it" cuts nicely through the BS.
I understand your concern about non-algorithmic bulletin boards, a legitimate but second-order problem. In the absence on Section 230 courts would decide if internet bulletin boards or forums are publications like newspapers or more similar to directories such as physical bulletin boards or the phone book. Criteria you mention "Websites which just show content in chronological order, or stuff that's upvoted, or any other transparent way" would later be relevant in court but first: Repeal the law.
Dean Baker has worked out the implications of Section 230 reform https://cepr.net/publications/reformed-section-230/
Dean Baker proposes a new Defamation Takedown protocol like the DMCA. Basically, if someone reports a post as defamatory, the hosting site would need to take it down, just like it does with copyright infringement notices today. That might be more acceptable to the giants (since it creates yet another barrier to entry for small players, like the DMCA does), but it is far less effective than my proposal, which is simpler and harder to game. It also requires the person being defamed to monitor the entire internet and request takedowns, which can be days or weeks later. Just like Youtube makes reporters jump through hoops to prove they own the copyright (unless you are another giant corporation, then they just trust you), I think this wouldn't really be very effective unless crafter very carefully. The DMCA works since giants on both sides of the problem worked out a system they felt was fair--there's no giant representing the people being defamed, so the law will be gutted as it is being created. And my approach halts most disinformation and misinformation as well (it can't get amplified) even if it's not defamatory.
Good link, thanks.
I view algorithmic curation as the problem, not the solution. When Facebook was just things your friends posted shown in chronological order, it was fine. Twitter becoming the "global commons" and putting everyone in the same "room" is what caused the current escalation and polarization. Humans can't handle interacting with so many people/information sources; algorithms give us the illusion of being able to surmount this limitation, while actually taking advantage of it by selecting for "engagement". Algorithmic curation should be regulated into the ground like cigarettes. Turn the internet back into a bunch of separate "rooms", and things will go back to normal.
Agree! My concern is political: "algorithms" is abstract. Meta's business model is not widely understood.
I wonder if it is more feasible to go after special Section 230 protections big platforms enjoy. "Elon Musk should be just as accountable as your local paper" seems an easier political sell and would have similar effect.