The Old Regime and the Internet Revolution

On Friday, Twitter suspended the account of Robert Stacy McCain, a conservative blogger and dogged critic of feminism, apparently without warning or explanation. This has led, in true Twitter fashion, to protests under the hashtag #FreeStacy.

Only a few weeks earlier, Twitter had announced the creation of a "Trust and Safety Council," to which it appointed Anita Sarkeesian, a feminist known for denouncing "sexism" in video games, a prominent figure in the Gamergate controversy—and oh yes, a frequent target of criticism from McCain. So it sure looks like the moment Twitter gave Sarkeesian the power to do so, she started blackballing her critics.

Is this what has come from the Internet's promise of open and unfettered speech, liberated from the gatekeepers of the "legacy media"? Or did we make the old gatekeepers obsolete, only so we could impose new ones?

The whole thing reminds me of the book Alexis de Tocqueville wrote after Democracy in America. In The Old Regime and the French Revolution, he examined the failed promise of France's rebellion against monarchy. What concerned him was not just the Terror and the beheadings, but the fact that the French toppled all of their institutions and tried to remake their politics, only to see all the old institutions re-assert themselves. They ended up with the same system, just under new rulers. The main similarity between the new system and the Ancien Régime was its administrative centralization, the way everything was controlled out of Paris, sapping all power and initiative from local institutions.

This strikes me as a good analogy for what's been happening with the old regime and the Internet revolution. The impact of the Internet has been a radical decentralization, the breaking down of gateways and barriers to entry for the transmission of information. But the old imperatives for centralization and control are still there, and they're trying to reassert themselves.

Part of the revolutionary potential of social media is that it was supposed to allow users to regulate their own flow of information, expanding it, contracting it, filtering it, and policing it as they wished. No longer would a media gatekeeper—for example, the editors of a big newspaper—decide what news was fit to print. You would decide for yourself what you wanted to see.

This has had two opposite effects. On the one hand, it has freed up the flow of information and made it easier for those outside the mainstream to find an audience for their views. On the other hand, it has also made it easier for the consumer of information to insulate himself from opposing views, to live within a bubble of people who agree with him. It increases the Pauline Kael Effect, named after the New Yorker film critic who supposedly said that she didn't know how Richard Nixon could have won his 1972 landslide because she didn't know anyone who voted for him. (The real quote is both better and worse.) But there's a good argument to be made that a cocoon of one's own construction is better than a cocoon imposed on you by a small elite of newspaper editors and television network producers.

This revolution was not just technological but legal. Section 230 of the Communications Decency Act of 1996 exempted the proprietors of any "interactive computer service" from liability for what its users said. So if I libel you on Twitter or Facebook, you can sue me for defamation, but you can't sue Twitter or Facebook. You can see how this radically reduces the incentives for such lawsuits, because going after a private individual is much less likely to be worth the expense than filing suit against a billion-dollar corporation. So in addition to giving us the Internet's familiar background noise of random insults and verbal abuse, this helped reduce the threat of frivolous lawsuits and legal harassment intended to shut down dissenting voices.

Yet part of the original intent of Section 230 was to allow Internet services to engage in some filtering of their content. Before this law, under the Stratton Oakmont decision, editorial control was an all-or-nothing proposition. If an Internet service took any action to manage or restrict content, it was considered a "publisher" and was therefore responsible for anything said by anyone in that forum. If they wanted to restrict one thing, they had to restrict everything—and be legally liable for whatever they didn't restrict.

Section 230 changed that and created the free-wheeling Internet we all know and love, but part of its intent was to allow editorial control on a sliding scale. A forum host could restrict obscenities or pornography—hence the provision's inclusion in the Communications Decency Act—without making itself legally liable for everything said by its users. So this was not just a charter for unrestricted speech on the Internet; it was also a charter for private rules on user conduct.

As both a legal and philosophical issue, this is not a matter of freedom of speech vs. censorship. Censorship is what happens when a government prevents you from expressing your opinion. But Internet services like Twitter, Facebook, or Reddit are privately run networks and entitled to set their own rules. Think of it this way: if you decided to create an Internet discussion group only for people on the left, or only for people on the right, and you took active measures to eject any user who expressed views contrary to that political leaning, this would be fully within your rights. The same is true for services that encourage a much wider range of discussion.

The fact that sites like Twitter and Facebook have avoided narrow ideological restrictions is not because of some unusual dedication to tolerance. It serves a clear business purpose. To declare a specific ideological persuasion risks driving away at least half of their users who don't agree with that persuasion. Rather than being a forum for only one kind of discussion, these companies wanted to make themselves into a forum for all kinds of discussions.

I've been around long enough in the world of political media to see various attempts to create private networking lists or bulletin boards or discussion groups for people of a particular ideological persuasion, sometimes by invitation only. As a general rule, they haven't succeeded, precisely because of their restricted size. The kinds of things they were intended to do—to allow like-minded people to interact, draw one another's attention, influence each other with new ideas and arguments, and (let's be honest) coordinate how to spin the latest news—is now pretty much all done on Twitter, out in the open.

The best analogy I can think of is that political Twitter is like the DC cocktail party of myth and legend: it's a place where people go to be noticed and impress each other with their cleverness, partly in the hope that others will provide them with employment or promote their work. Except that this cocktail party is going on all the time, you don't need to be in DC any more, and you don't even need an invitation. This also means that it's a party with a very ideologically diverse crowd, where people with differing political views can take public jabs at one another.

All of that gives Twitter a unique and exciting vibrancy, and its importance to the media profession in particular explains why the fate of Twitter receives so much coverage, despite having a much smaller user base than Facebook. (Twitter has 320 million active users; Facebook more than 1.5 billion.)

At the same time, the rough jostling of an uncontrolled crowd can sometimes make for an unpleasant experience. The traditional ethos of the Internet has been to encourage everyone to develop a thick skin and to shrug off hostility—witness the ritual of celebrities reading mean tweets about them on television—and there's something very healthy about this attitude. Still, being insulted online by random strangers is not everyone's idea of a good time. But that's the problem the Internet was supposed to solve by allowing us to block or mute obnoxious users or filter certain kinds of content out of our feeds. It's supposed to give us the tools we need to control and limit or own experience, without the need for an online nanny to do it for us.

Ah, but there's the rub, because there are always those who want to do it for us, those who miss the old centralized ways of filtering the transmission of information, because it could give a lot of power to small but committed pressure groups.

There have long been whispers about this. It is an open secret that Twitter is highly capricious in awarding the little blue check marks for "verified" accounts, which gives those accounts an advantage in visibility and gaining new followers. Obscure leftist writers will get their blue check marks while prominent figures on the right will not—or, as in the case of British firebrand Milo Yiannopoulos, have it taken away as punishment for inflammatory views. Yiannopoulos has followed up with reports accusing Twitter of "shadowbanning": surreptitiously restricting the distribution of tweets from users whose ideas some Twitter employee don't like.

All of that is the background for the latest Twitter controversy: the company's decision to appoint a "Trust and Safety Council" that will be "a new and foundational part of our strategy to ensure that people feel safe expressing themselves on Twitter." Among a list of organizations described as members of the council are the Dangerous Speech Project, which has produced a set of guidelines for recognizing such "dangerous speech"—a set of criteria so vague and broad that they could encompass anything from the incitement that led the Rwandan genocide to an average afternoon on talk radio. And maybe that's the point.

Then there is Feminist Frequency, the outlet used by Anita Sarkeesian as a platform to criticize video games for being sexist. Sarkeesian's critics accuse her of using claims of online harassment to discredit her critics and get them kicked off of services like—you guessed it—Twitter. So when Sarkeesian and the "dangerous speech" types got appointed as Twitter's speech police, there was reason to think they're going to turn Twitter into a university-style "safe space" where "safety" is assured by removing advocates of opposing ideas.

No, really, that's what a "safe space" is. As one student explained to a New York Times reporter, she needed a "safe space" after briefly hearing a conservative speak on campus because "I was feeling bombarded by a lot of viewpoints that really go against my dearly and closely held beliefs." Safe spaces are not about freedom from harassment or physical threats. They're about freedom from intellectual opposition.

You get the same impression when Twitter CEO Jack Dorsey proclaims that "Twitter stands for freedom of expression, speaking truth to power, and empowering dialogue. That starts with safety."

Twitter stands for freedom of expression, speaking truth to power, and empowering dialogue. That starts with safety. https://t.co/PbEoM1lCyb— Jack (@jack) February 9, 2016

No, actually, "freedom of expression" starts with "freedom." It's right there at the beginning of the phrase. But by recasting it as an issue of "safety," Twitter de-emphasizes the "freedom" part and sets up a rationale to limit or ban its users in the name of protecting others.

And so we see, less than two weeks later, a particularly sharp-elbowed and combative critic of feminism getting his account suspended.

That is their prerogative. Twitter is a private company, and it's free to decide that it's more important to make themselves into a "safe space" rather than a free space. But they have to face the consequences. In the process of trying to make a "safe" experience for certain kinds of users, they make it an unsafe experience for those who fall afoul of Twitter's new guardians. Those people will leave for other forums that don't restrict them.

That's a loss for Twitter. In trying to lure new users with promises of "safety," the company risks destroying the sense of free-wheeling openness that attracted many of its existing users. Twitter has had a lot of trouble monetizing its user base, which is why the company's stock is in trouble, but before it can make money from it users, it has to have users, and lots of them. Ask your friends on MySpace what happens when a social network stops being the place where everyone wants to hang out.

And something wider will be lost. One way to restrict the public debate is not to impose an outright ban on ideas you disapprove of, but to segregate them. Relegate them to special unsafe spaces everyone is warned not to enter, or drive them into a partisan zone where people who already agree with those ideas can talk to each other, but which is quarantined from respectable, "mainstream" discussion. Fox News Channel, for example, was formed as a right-leaning alternative to left-leaning network news shows. But instead of driving the ideas of the right into the mainstream, it tends to segregate them, so that people on the left can dismiss anything that appears on Fox as right-wing propaganda. (People on the right do the same thing with MSNBC.)

One of the things that makes Twitter vibrant and interesting is that it is not like this. There are users who like to block everyone they disagree with, but in doing so, they are segregating themselves out of the mainstream discussion. Placing the filtering of content into the hands of individuals not only liberates them from centralized control; it also creates an incentive toward engagement with opposing views rather than isolation from them. That's very healthy for our public debate, and for the forums that host it.

But most of all, a re-establishment of centralized gatekeepers is a counter-revolution against the whole technological promise of the Internet, which was supposed to put more power into the hands of individuals.

If Twitter wants to re-establish itself as a free space for the discussion of political ideas, and keep all of its right-of-center users, it has to do more than just re-instate capriciously banned users like McCain. Here is a little five-point manifesto for the Twitter users' rebellion.

1. Create objective rules of conduct.

In place of its current vague and elastic rules, Twitter should list clear, specific, ideologically neutral rules of conduct, so that anyone can predict what will qualify as abuse, and it can't simply be interpreted to mean "any speech a Twitter employee doesn't like."

2. Give more tools to users.

If Twitter is concerned that some users might have a rougher experience than they would like, give them the tools to manage their own accounts and let them cocoon themselves off from the mainstream discussion if that's what they want. I happen to think that the current "block" and "mute" functions work just fine, but maybe give users the option to filter out messages with certain kind of keywords. Put the power in the hands of the users, not a centralized Internet nanny.

3. Dismantle the kangaroo court.

Twitter needs to disband the "Trust and Safety Couneil," and kick off the "abuse" abusers, including Sarkeesian and anyone else with a record as an unprincipled partisan. If they still feel the need for a decision-making process for eliminating truly abusive accounts, put those decisions in the hands of a new council with transparent proceedings, and populate it with an ideologically diverse group of people from both the left and the right. This will discourage outright political bias, while giving credibility to Twitter's decisions when it does become necessary to ban someone.

4. Fire CEO Jack Dorsey.

The "Trust and Safety Council" was Dorsey's brainchild, and he's the one who chose to give it over to political hacks with an axe to grind. In other words, while Twitter's user base has been leveling off and its share price has been going down in flames, he's been busy hatching a scheme to drive away even more of its users. If Dorsey were sole owner of Twitter, he would have a right to run it into the ground however he likes. But it's a publicly traded company, and he has shareholders, co-owners of the firm who have a right to complain that he's crashing their share prices for the sake of his own personal political crusade. The company's remaining shareholders need to rebel before he sets fire to more of their money.

5. "Liberals" need to join the rebellion.

If there's anybody on the left who still takes the "liberal" label seriously and cares about open discussion and a frank exchange of ideas, then this is their opportunity to stand up and be counted.

Twitter needs to take all of those measures seriously, or be on notice that they have a technological model that is really, really easy to duplicate. If they decide to restrict themselves to a lefty debating club, they should expect a more open-minded rival to come along and eat their lunch.

Twitter could deal with concerns about harassment by creating an honest, transparent process and by giving users more tools to filter content and control their own experience. But putting an Orwellian and ideologically biased "Trust and Safety Council" in charge of filtering our content for us is a return to the Old Regime, and it squanders the xpromise of the Internet revolution.

Follow Robert on Twitter, until they kick him off.