“First, do no harm,” a doctrine typically associated with the practice of medicine, is the right ethic when it comes to decisions surrounding Silicon Valley’s paid promotion technologies and their effects on elections and democracy. A desire to avoid harm — in particular, the spread of misinformation — is part of what persuaded Twitter’s chief executive, Jack Dorsey, to announce that his company will no longer run political ads. And Twitter is not alone: LinkedIn, Pinterest, Microsoft and Twitch also refuse political ads, while Google accepts them in some states but not others.

Facebook is now the outlier, and it is increasingly hard to understand why it is insisting on accepting not only political advertising, but even deliberate and malicious lies if they are in the form of paid advertisements. Given how much can go wrong — and has gone wrong — the question everyone is asking is: Why does Facebook think it needs to be in this game? Naïveté is at this point the most flattering explanation.

It isn’t, as some think, just about making money, for as a revenue source, the money at stake is minor. But the money does matter, in a different way. Paying for promotion is how, on social media, some speakers gain priority over others. This creates an advantage unrelated to actual popularity. Paired with the freedom to lie, the effect is to give political lies and paid misinformation campaigns a twisted advantage over other forms of election speech (like “the news.”) Even as Facebook’s “integrity” teams try to stamp out other forms of deception, paid promotions gain access to the full power of Facebook’s tools of microtargeting, its machine learning and its unrivaled collection of private information, all to maximize the influence of blatant falsehoods. What could possibly go wrong?

If the idea of prioritizing lies over truth doesn’t sound very appealing, Facebook’s defenses of its policy are almost their own misinformation campaign. Nick Clegg, Facebook’s vice president for global affairs and communications, has suggested that Facebook sees itself as providing the “tennis court” where politicians play the game of politics. But tennis actually has strict rules; Facebook has embraced, instead, the norms of a fighting cage. More important, Mr. Clegg is hiding the more fundamental question: Who ever said Facebook needed be the tennis court in the first place?

Facebook, which for years has declared it is not a media company, is now asserting that it is a medium necessary to a fair electoral process. That’s the implication of the tennis analogy, and also of Mr. Zuckerberg’s favourite defence of his policy, that Facebook must run political advertisements, even blatant lies, to help political challengers take on incumbents. Having personally run on a ticket against an incumbent, I am sympathetic to measures that might level the playing field. But not only is there no evidence to back Facebook’s assertions, there are many other, more obvious and less dangerous ways to fight the advantages of incumbency, like the public matching of campaign donations. It is ludicrous to suggest that allowing paid political lies online is what’s really necessary to help out the little guys.

So strange is the policy, so confusing even to Facebook’s own employees, perhaps it really is nothing more than an effort to placate Facebook’s conservative critics — giving Mr. Zuckerberg space to loudly declare that he maintains a fair and balanced policy when it comes to political speech. Leave aside that this is unlikely to mollify mainstream conservatives, whose real concern is that Facebook’s filtering of hate speech will enshrine “political correctness.” It is the effort to mollify itself that should raise red flags.