What do the latest installments reveal?
I’m Isaac Saul, and this is Tangle: an independent, nonpartisan, subscriber-supported politics newsletter that summarizes the best arguments from across the political spectrum on the news of the day — then “my take.”
Today's read: 13 minutes.
In our "Quick hits" section earlier this week, we alluded to Virginia Gov. Glenn Youngkin's victory in a "special election" in 2021. In fact, Gov. Youngkin did not win a special election. Virginia, like four other states, holds its statewide elections in years that do not coincide with presidential or congressional midterm elections.
This is our 74th correction in Tangle's 176-week history and our first correction since December 5th. I track corrections and place them at the top of the newsletter in an effort to maximize transparency with readers.
- The consumer price index (CPI) rose 7.1% in November compared to the same time last year, down from a June high of 9.1% and below expectations. It's the latest sign of inflation easing. (The latest)
- President Biden signed the Respect for Marriage Act into law, codifying federal recognition of same-sex and interracial marriage. (The law)
- The U.S. is preparing to send Ukraine the Patriot missile defense system, a form of long-range air defense that Ukraine has been requesting for months. (The system)
- Florida Gov. Ron DeSantis holds an early lead over Donald Trump among GOP primary voters, according to a new Wall Street Journal poll. (The poll)
- Congressional negotiators say they’ve reached an agreement on a framework to fund the government through 2023. (The deal)
Our 'Quick Hits' section is created in partnership with Ground News, a website and app that rates the bias of news coverage and news outlets.
The Twitter files. Today, we're going to be covering parts 2, 3, 4 and 5 of the "Twitter files," a series of leaks being posted to Twitter by the journalists Matt Taibbi, Bari Weiss and Michael Shellenberger about Twitter's moderation decisions. We covered part one of the files already, which included internal communications about the decision to suppress the Hunter Biden laptop story.
One update from that report: In our first edition covering this, we mentioned that Taibbi and the other reporters had agreed to “unnamed conditions” in their reporting. Since then, all three reporters have said that the condition was that they must post their reporting to Twitter first before publishing it anywhere else, and have all insisted they’ve been given unfettered access to Twitter’s internal files.
In Part 2 of the series, Bari Weiss explored the way Twitter uses blacklists to actively prevent certain accounts or tweets from trending, and reduces the visibility of those accounts. Included in her thread was evidence that Twitter had actively limited the reach of Stanford's Dr. Jay Bhattacharya, who argued Covid-19 lockdowns would harm children, as well as conservatives like Charlie Kirk, LibsOfTiktok, and Dan Bongino. She shared screenshots of how Twitter’s tools work, including “do not amplify” and “trends blacklist” lists.
In Part 3, Taibbi began sharing internal communications related to the decision to remove former President Donald Trump from the platform. He shared internal Slack messages showing that Yoel Roth, Twitter's former head of trust and safety, was meeting on a regular basis with officials from the Federal Bureau of Investigations (FBI), Department of Homeland Security (DHS), and office of the Director of National Intelligence (DNI) leading up to the 2020 election. Taibbi described an environment where "Twitter was a unique mix of automated, rules-based enforcement, and more subjective moderation by senior executives."
Taibbi also pointed to Twitter officials who were considering the "context" of Trump's tweets, not just the tweets themselves, to determine whether he was in violation of the platform's rules and worthy of a suspension. This was counter to their publicly stated rules, which said they could not consider the myriad of ways tweets could be interpreted while making content moderation decisions.
In Part 4, Michael Shellenberger writes that senior Twitter executives were creating justifications to ban Trump, sought a policy change for Trump alone, and expressed no concern for the implications of such a ban. Shellenberger adds the context that in 2018, 2020, and 2022, 96%, 98%, and 99% of Twitter staff's political donations went to Democrats and that Roth, head of trust and safety, said there were "literal nazis" in the White House in a 2017 tweet.
Shellenberger cites internal deliberations where Twitter officials concede its ban of Trump was based "specifically on how Trump's tweets are being received and interpreted," even though in 2019 it said it would "not attempt to determine all potential interpretation of content or its intent." Notably, Twitter executives decided to abandon their "public interest" policy that protected other politicians and public figures from being permanently banned on the site in order to kick Trump off of it.
In Part 5, Weiss explores the actual removal of Trump from the platform. In these tweets, she notes that Trump went into January 8, two days after the riots at the Capitol, with one remaining strike before a permanent suspension. He tweeted twice:
The first said: “The 75,000,000 great American Patriots who voted for me, AMERICA FIRST, and MAKE AMERICA GREAT AGAIN, will have a GIANT VOICE long into the future. They will not be disrespected or treated unfairly in any way, shape or form!!!”
The second said: “To all of those who have asked, I will not be going to the Inauguration on January 20th.”
Weiss shared chats from inside the company where staffers expressed confusion about how either tweet could be described as incitement, despite the fact they had prompted many employees to call for Trump's permanent ban. Just an hour later, however, Twitter’s Head of Legal, Policy, and Trust, Vijaya Gadde, suggested the tweets may be "coded incitement to further violence." Members of Twitter's scaled enforcement team then suggested they “view him [Trump] as the leader of a terrorist group responsible for violence/deaths comparable to Christchurch shooter or Hitler and on that basis and on the totality of his Tweets, he should be de-platformed.”
After a 30-minute all-hands meeting in which Gadde and former CEO Jack Dorsey tried to explain why they hadn't yet banned Trump, employees became increasingly upset. One hour later, Twitter announced a permanent suspension of Trump's account.
Together, Weiss, Shellenberger and Taibbi make the case that these files show how Twitter targeted conservatives and bent their own rules to punish users whose politics they did not agree with.
Critics of this narrative say Musk is selectively leaking internal documents to reporters who he knows will craft the narrative he wants, and that the files don't prove conservatives were treated any differently from liberals. Rather, many say, the internal communications show there were many difficult moderation decisions Twitter had to make and they took those decisions very seriously.
Below, we'll take a look at some commentary about the files from the right and left, then my take.
What the right is saying.
- Many on the right say the Twitter files have proven conservatives are not treated fairly on the platform.
- Others argue the revelations are evidence that Twitter needs more oversight.
- Some say the internal deliberations are reflective of the way the left views "truth" and how it aims to suppress dissent.
In The Washington Post, Hugh Hewitt said he was wrong to defend Twitter to other conservatives.
"Using Twitter’s own internal files, released with the blessing of new owner Elon Musk, Weiss demonstrates that Twitter was indeed censoring conservatives, despite vigorous and repeated denials from company brass over the years," Hewitt wrote. "Verified accounts of such prominent conservatives as activist Charlie Kirk (who like me hosts a radio show for Salem Media Group), radio host Dan Bongino and many others were flagged so that Twitter algorithms would not highlight their tweets. [Jack] Dorsey’s smoke screen masked other kinds of deception, too. Conservatives were led to believe that they had equal access to the Twitter audience.
"People and organizations on the right invested time, effort and sometimes money to craft messages in the belief that the results could be read on a level playing field. In truth, any message out of favor with Twitter management — or somehow offensive to lower-level content moderators — might find only a small fraction of its intended readership," Hewitt said. "Another apparent deception targeted Twitter users who counted on the platform for breaking news or bubbling debates. They relied on Dorsey’s promise of neutrality. When former president Donald Trump's account was canceled, and information about Hunter Biden was tightly rationed, at least the decisions were public, and Twitter users could factor them into their perceptions of the world. Not so with secret protocols."
In the American Conservative, Rod Dreher said it's time for the government to start regulating Big Tech.
"Weiss and her team have been given access to internal Twitter documents. What this new trove of documents released by Elon Musk shows is that the previous Twitter regime lied flagrantly to the public about its policies governing suppressing and censoring speech on the platform. The thread cites specific examples of how Twitter banned or shadowbanned conservative accounts while publicly denying that it was doing any such thing. Why did they do it? The internal reason was 'safety,'" Dreher said. "Twitter has immense power. You might not be on Twitter, but trust me, almost everybody who makes decisions in the media is.
"That's why so many media leftists have raised hell about Musk's takeover of Twitter: they understand correctly how much power the platform has over shaping public discourse," he wrote. "If Elon Musk hadn't spent $44 billion of his own money to buy Twitter, we would never know any of this, and Twitter would still be beavering away, punishing conservatives and others who challenge far-left ideas, while pretending publicly to be fair and responsible... I am not a lawyer, so I don't know what's possible within constitutional parameters -- but I know well that Republicans and free-speech advocates had better start thinking hard about this, and formulating a plan."
In The Wall Street Journal, Gerard Baker said the revelations were instructive but not surprising.
"It’s not that executives, editors, reporters and algorithm-writers at big media and tech companies consciously promote their ideological nostrums, mindful of and striving to overcome competing ideas. It’s much worse. If you’re an executive at Twitter with the Orwellian title of 'head of trust and safety' or a 'disinformation' and 'extremism' reporter at NBC News, or an executive at the New York Times charged with enforcing intellectual homogeneity, you’re not simply promoting a view of the world that you espouse," Baker wrote. "You are doing something much more important, which compels compliance and tolerates no alternatives: promulgating the One True Faith, a set of orthodoxies from which there is no legitimate dissent.
"Here is the asymmetry: Most conservatives, or intellectually curious people, don’t think like this. They don’t think that someone with differing opinions on say, immigration restrictions, the right level of taxation, or the case for affirmative action is voicing a provably false and intrinsically illegitimate view that amounts to misinformation. They think their opponents’ beliefs are wrong and reflect flawed analysis or erroneous evidence," Baker said. "But they don’t think there is only one acceptable belief and that dissent from it is analytically impossible, intellectually dishonest and morally contemptible. But this is the left’s mindset. It is why they don’t need instructions from government officials or public censors to determine access to information. They are themselves the controlling authority."
What the left is saying.
- Many on the left say the Twitter files are being exaggerated.
- Some argue Weiss, Taibbi and Shellenberger are manufacturing a narrative without sharing all the relevant information.
- Others call on Musk to share the files more widely in order to give more accurate context.
In New York Magazine, Eric Levitz said the reporters on the Twitter files are misleading readers.
"The second installment of the Twitter Files had a bit more substance than the first. But like its predecessor, it affirmed conservative narratives of persecution by omitting key pieces of context, while also including one outright lie," Levitz said. "Twitter has made no secret of the fact that it punishes accounts by limiting their visibility. Since at least 2018, Twitter’s help page has said, 'When abuse or manipulation of our service is reported or detected, we may take action to limit the reach of a person’s tweets.' Twitter also listed 'Limiting tweet visibility' as an enforcement option under the company’s terms of service. Twitter’s current ownership has openly embraced this form of content moderation. Last month, Musk tweeted: 'New Twitter policy is freedom of speech, but not freedom of reach.'
"Nevertheless, after reporting that the conservative commentator Charlie Kirk had been put on a 'Do not amplify' list, Weiss bizarrely claimed that Twitter had long 'denied that it does such things,'" Levitz wrote. “Similarly, she suggests that the conservative personalities Dan Bongino and Charlie Kirk were placed on blacklists because of their political views. Yet both those commentators are provocateurs who quite plausibly might have violated the platform’s rules regarding abuse at one point or another... The Twitter Files provide limited evidence that the social-media platform’s former management sometimes enforced its terms of service in inconsistent and politically biased ways."
In Vox, Shirin Ghaffary wrote about what the Twitter files don't tell us.
"We don’t have a full explanation, for example, of why Twitter limited the reach of these accounts — i.e., whether they were violating the platform’s rules on hate speech, health misinformation, or violent content," she wrote. "Without this information, we don’t know whether these rules were applied fairly or not... And while Weiss has surfaced specific examples of Twitter limiting the reach of conservative accounts known for spreading hateful content about the LGTBQ+ community or sharing the ‘big lie’ about the US presidential elections, we don’t know if Twitter did the same for some far-left accounts that have also been known for pushing boundaries, such as some former Occupy movement leaders who have complained about Twitter’s content moderation in the past.
"Historically, most Twitter employees — like the rest of Big Tech — lean liberal. Twitter’s conservative critics argue that this presents an inherent bias in the company’s content moderation decisions," Ghaffary wrote. "Former Twitter employees Recode spoke with this week insisted that content moderation teams operate in good faith to execute on Twitter’s policy rules, regardless of personal politics. And research shows that Twitter’s recommendation algorithms actually have an inherent bias in favor of right-wing news. What’s been shared so far in the Twitter files doesn’t offer clear proof that anyone at Twitter made decisions about specific accounts or tweets because of their political affiliation."
In MSNBC, Steve Vladeck criticized Musk and others for claiming there was proof Twitter had violated anyone's First Amendment rights.
"The free speech clause of the First Amendment, like virtually every provision of the Constitution (except the Thirteenth Amendment, prohibiting slavery; and the Eighteenth Amendment, imposing prohibition), applies only to 'state action.' A private business no more violates the First Amendment by banning particular types of speech in its operations than I violate the First Amendment by not allowing particular types of speech in my home," he wrote. "This brings us to Musk’s insinuation that Twitter’s actions violated the First Amendment because it was 'acting under orders from the government.' There are at least two problems here, both of which would’ve been caught by any first-year law student (and plenty of undergraduates).
"First, the Supreme Court has made clear, for decades, that the government 'normally can be held responsible for a private decision only when it has exercised coercive power or has provided such significant encouragement, either overt or covert, that the choice must in law be deemed to be that of' the government actor," Vladeck said. 'It is understood that government requests to private entities don’t meet this test absent proof that the private entity did not believe it had any choice but to comply. Even the most conspiratorial reading of the 'Twitter Files' fails to uncover such evidence. Second, for those who are inclined to assume compulsion even without evidence, the orders have to come from the government. In October 2020, when the Hunter Biden story broke, neither Joe Biden nor the Biden campaign were 'the government.'"
Reminder: "My take" is a section where I give myself space to share my own personal opinion. It is meant to be one perspective amid many others. If you have feedback, criticism, or compliments, you can reply to this email and write in. If you're a subscriber, you can also leave a comment.
- These were more interesting and newsworthy files than installment #1.
- Most jarring to me were the weekly meeting with government entities like the FBI, and the very obvious invention of new standards to police Trump.
- We'd still benefit from seeing how liberal accounts were handled on the platform.
Well, these were a lot more interesting than the first installment.
When "Part 1" of the Twitter files was released, I wrote that "there was very little new information that hasn't already been reported by many outlets." It wasn't that the story wasn't important, or the details weren't damning, but there just wasn't any "bombshell" or "smoking gun" file showing Twitter had acted in an obviously unethical way to censor the Hunter Biden story. What it really showed was that there were internal deliberations about the decision, lots of people didn't think it was the right call, and the wrong decision was made. I said this at the time, as did many other commentators on the right (and, eventually, former CEO Jack Dorsey admitted they screwed up).
Part 2, 3, 4 and 5, taken together, are more interesting. These revelations included screenshots from Twitter administrative dashboards showing tools Twitter uses to suppress the reach of certain accounts. They also showed that the company, in simple terms, came up with new rules and potential violations in order to justify banning President Trump, and that several dissenting internal voices questioned the insular and ambiguous nature of how that decision was made. Additionally, Twitter regularly acted on the censorship requests from agencies like the FBI, DHS, DNI, and political campaigns, and Twitter's head of safety was meeting with those agencies on a weekly basis.
This is all big news. It is not a "nothingburger." And even though Twitter is a private company, there are very real free speech questions at play. How critical was the government's role in suggesting what content to suppress? Did these suggestions ever breach the level of coercion, or pressuring? How evenly were these standards applied across the political spectrum? Even if these instances ultimately fell short of violating the first amendment, liberals would do well to recognize "free speech issues" aren’t always about government suppression of speech. Private companies can and do suppress speech, and whether that is explicitly unconstitutional is separate from the question of whether it's a societal harm or not.
Twitter has also been slippery about its position on suppressing these accounts. Many on the left have pointed out that Twitter defines "shadowbanning" as “deliberately making someone’s content undiscoverable to everyone except the person who posted it, unbeknownst to the original poster.” But, as Ben Sturgis pointed out, this is also the most extreme version of this definition, and the conservatives who have said they were "shadow banned" over the years have obviously not meant that literally nobody could see their tweets. Rather, those posters meant that the engagement on their accounts was being artificially limited because of what they were posting, and they could tell.
Despite these claims getting mocked and derided by journalists and the left, these files show they were, at least in a few cases, actually right. And we have good reason to believe the active suppression of these accounts was probably more common than we know.
If you're on the left, think about it this way: Musk owns Twitter now. He clearly has issues with progressives, and has been hammering the "woke mind virus" he thinks is infecting the country. What if Musk, at the guidance of someone like Bari Weiss, were to label anti-Israel protests on Twitter as antisemitic hate speech and then instructed Twitter engineers to suppress tweets from Palestinian activists and/or suspend the accounts sharing those messages? What if, simultaneously, Twitter execs were meeting with the FBI for guidance on what counts as suspendable? Would liberals react with a shrug and insist that Twitter was a private company, so this isn't really a big deal?
I doubt it.
Which brings me to my final point: Twitter does suppress the voices of people on the left, too. The glaring hole in this entire ordeal is that Weiss, Taibbi and Shellenberger have failed to shed any new light on the ratio of this suppression, who Twitter most often targets, and why. They've shown us a handful of examples of prominent conservatives being throttled, that Twitter will create new standards on the fly in order to justify certain moderation actions, and did so in Trump's case. They've shown that, around the 2020 election, they were having regular contacts with government agencies (but again, it's worth remembering this was Trump's government, not Biden's).
But none of that tells us the full picture. I could, for instance, do a "bombshell" Twitter thread on all the big oil money being donated to Democrats, and frame it as proof that the Democratic party is owned by fossil fuels, and insist that if you want to do something about climate change you should vote for Republicans. But without also sharing that Republicans similarly take millions from big oil, this would be a misleading narrative to generate.
Given the political leanings of Twitter employees, executives, and the users of the site itself, I think it is a very safe bet that Twitter's moderation policies are unevenly applied and targeted against conservatives. But, again, Taibbi, Weiss, and Shellenberger haven't yet shown that, and Musk has hurt his own case by refusing to share the files with journalists outside of these three — who were already on his team before this whole controversy started.
As I said after Part 1, Musk should release the files more widely, and the journalists involved should flesh out differences in how moderation decisions were made between liberals and conservatives. I want to see some of the internal deliberations on moderating liberal activists and voices, and then compare. On top of the very important information we've already seen, that would add a great deal of context and strength to the narrative they say they have.
Your questions, answered.
Q: Hi Isaac! Has there been any recent developments concerning the Nord Stream 1 pipeline that went kaboom a while back? It hasn't been getting a lot of press lately -- and I don't recall hearing anything definite as to who might've been responsible. Is this getting memory-holed?
— Dave from Healdsburg, California
Tangle: In short: No. The closest thing we got to an update was a rather juicy story about the discovery of two ships who had turned off their trackers and passed within several miles of the pipeline in the days immediately before the leaks. The ships, between 95 and 130 meters long, were spotted by a firm called SpaceKnow, which used machine learning and satellite data to account for the "dark ships."
It's possible the automatic identification system (AIS) transponders on the two ships failed, but that is rare. Turning them off is rare, too. Usually it's a sign of a ship involved in illegal fishing, shipping contraband, or a clandestine military operation. SpaceKnow says they turned the data over to NATO, who continues to investigate the leaks, and has maintained it was a deliberate act of sabotage. That position has been supported by evidence of explosives at the site of the pipeline leaks.
For now, I'm not sure we are any closer to understanding who did this than we were in September. But investigations are ongoing. In the meantime, Russia recently resumed delivering gas to Europe on the Nord Stream 1 pipeline.
Want to have a question answered in the newsletter? You can reply to this email (it goes straight to my inbox) or fill out this form.
Under the radar.
Some of the latest data from the 2022 midterms suggests younger voters' support for Democrats may actually be waning, not increasing. Voters under the age of 30 broke 53% for Democratic House candidates compared to just 41% for Republicans in 2022. But that level of support was down significantly from 2020, when 61% of under 30 voters supported Joe Biden compared to just 36% for Donald Trump. In the 2018 midterms, young voters broke 64% for Democrats compared to 34% for the GOP. Political scientist Michael McDonald cautioned against making a trend out of what could be an anomaly, but said its possible young voters, who have the weakest partisan attachments, were hit hardest by inflation and voted accordingly. The Associated Press has more.
- 6 out of 7. Of the United Kingdom, United States, Canada, France, Germany, Spain and Japan, the number of countries where tweets from right wing politicians received more amplification from Twitter's algorithm than those on the left.
- $909,431. The amount of money Twitter employees collectively donated to Democrats during the 2020 election cycle.
- $14,137. The amount of money Twitter employees collectively donated to Republicans during the 2020 election cycle.
- 98.7%. The total percentage of money donated by Twitter employees in the 2020 cycle that went to Democrats.
- ~300. The number of Twitter employees who signed an open letter in January of 2020 calling on Jack Dorsey to permanently ban Trump from the platform.
Have a nice day.
NASA scientists say they have just recorded the first ever sounds of dust devils on Mars. When the rover Perseverance landed on Mars, it was equipped with the first working microphone ever put on the planet's surface. Now, scientists have a recording of a whirlwind, which will help them better understand Mars' atmosphere and prepare for future trips. “We can learn a lot more using sound than we can with some of the other tools,” Roger Wiens, professor of earth, atmospheric and planetary sciences at Purdue University said. “They take readings at regular intervals. The microphone lets us sample, not quite at the speed of sound, but nearly 100,000 times a second. It helps us get a stronger sense of what Mars is like.” The team had observed about 100 dust devils on Mars since landing, but the microphone only records for about three minutes every few days – so this was the first time it caught one. Purdue.edu has the story.
Here are some ways to help...
💵 If you like our newsletter, drop some love in our tip jar.
🎉 Want to reach 50,000 people? Fill out this form to advertise with us.
😄 Share https://readtangle.com/give and every time someone signs up at that URL, we'll donate $1 to charity.
📫 Forward this to a friend and tell them to subscribe (hint: it's here).
🎧 Rather listen? Check out our podcast here.
🛍 Love clothes, stickers and mugs? Go to our merch store!