Conservatives remain convinced that the tech industry is biased against them. They point to evidence that Facebook, Twitter, Google and YouTube are staffed disproportionately by liberals, a fact that nobody seriously denies.
In a fascinating essay for the “New Atlantis”, Adam White argues that there is a deep alignment between their corporate culture and the assumptions of modern liberalism. A nearly inevitable result, conservatives say, is that the companies are more likely to try to suppress conservative content as offensive: taking down tweets and posts from righties and directing traffic elsewhere.
The companies deny they have displayed bias, even as admit they have made occasional mistakes in content moderation. One school of conservatives argues that while the tech titans’ decisions may be wrong and worthy of condemnation, it is not government’s business to regulate their decisions because they are private companies.
According to this reasoning, people who are unhappy with Facebook’s policies should condemn or boycott it. They should cajole Twitter into adopting better policies. Or they should try to start a rival site.
What these conservatives don’t want is for government to force the companies to behave differently, beyond holding them to their own terms of service. Another group on the right thinks that the hands-off approach is dogma unsuited to the world we actually inhabit. The social-media companies are too large a part of our society, and host too much of our political debate, for the government to leave them alone.
A leading voice for this second group is Republican Senator Josh Hawley of Missouri. His criticisms of social-media companies are wide-ranging.
Political bias is among his main concerns. “Google and Facebook should not be a law unto themselves,” he has said. “They should not be able to discriminate against conservatives. They should not be able to tell conservatives to sit down and shut up.”
Soon after his election in 2018, Hawley argued that the big tech companies are benefiting from a government-granted privilege. He observed — on Twitter, naturally — that Section 230 of the Communications Decency Act gives these companies immunity from liability for content that appears on their sites.
They get this immunity, he suggested, on the condition that they serve as forums for “a true diversity of political discourse”. He called for congressional investigations to see if they were meeting that condition. Congress has subsequently held hearings on it.
Republican Senator Ted Cruz, questioning Facebook CEO Mark Zuckerberg, made the same point as Hawley: “The predicate for Section 230 immunity under the CDA is that you’re a neutral public forum.”
What they’re saying: Facebook can be a message board with no responsibility for the messages tacked onto it. Or it can take an active role in shaping content, in which case it assumes legal responsibility for that content.
Those who fear government regulation in this area have pushed back against Hawley and Cruz. They point out that the law also gives the companies immunity for actions they take against obscene, harassing or otherwise objectionable material.
We can split the difference on the question of what the law assumes. The CDA does not require companies to be neutral about the political content of messages they host in order to have immunity. It was, in fact, written to encourage them to police obscenity, which is why it’s called the Communications Decency Act.
The law was enacted in 1996, after a web service provider, Prodigy, had been found to incur liability for postings because it had tried to make itself family-friendly and deleted some material it found to be offensive. Legislators thought it was unfair and undesirable to punish a company for trying to do the right thing.
But it’s fair to say that back then there was a background assumption that the companies would not be tilting left or right. So what options does the government have to deal with the perceived problem?
Perhaps we should split the difference on the policy question, too. The companies are too important to ignore, but government regulation is full of dangers.
It does not seem like too much for Congress to ask that the immunity it has provided be conditioned on some more transparency from the companies about how they decide what content to promote, hide or suppress. One possibility would be to require the companies to disclose summary information about how they make these decisions; another, perhaps easier to administer, would be to require them to publish a database of content they have deleted.
Before we decide either to regulate the tech companies or leave them alone, let’s find out if the anecdotes of bias are just anecdotes, or something more.