Sen. Kamala D. Harris, like many Democratic lawmakers, is calling for an impeachment inquiry of President Trump. But her Twitter account and her presidential campaign have issued a second call to action in recent days: Suspend Trump’s @realDonaldTrump account from Twitter.

In a letter sent to Twitter on Tuesday, Harris’s presidential campaign asked Twitter chief executive Jack Dorsey to suspend Trump for violating the site’s rules against “targeted harassment.” The Harris camp says that multiple recent tweets from the president were meant to “target, harass, and attempt to out the whistleblower” who has alleged that the president pressed Ukraine’s president to investigate former vice president Joe Biden and his son Hunter.

The California senator and her campaign have singled out a few tweets that they believe break Twitter’s rules. Here they are:

Harris’s letter also pointed to Trump tweets promising “Big Consequences” to anyone who provided information to the whistleblower and questioning whether House Intelligence Committee Chairman Adam B. Schiff (D-Calif.), who is leading the investigation against Trump, should be arrested for “treason.” And Trump’s tweet quoting the Rev. Robert Jeffress that invoked the idea of a “civil war” if he’s removed from office.

If @realDonaldTrump were a normal Twitter account run by a random person, Twitter would evaluate these tweets for potential violations. If the company then determined that the tweets were breaking their rules, it would either temporarily lock Trump’s account (and ask him to delete the offending tweets) or suspend the account.

But @realDonaldTrump is not a normal account, and over the years, Twitter has established a few loopholes that protect him from having to play by the rules. Below is a guide to those loopholes, how they’ve applied in the past, and how they might apply to Trump’s recent tweets.

A rule loophole for “military or government entities”

Twitter announced in late 2017 that it would begin enforcing new rules designed to stem hateful and abusive content. Twitter bans “specific threats of violence or wishing for serious physical harm, death, or disease to an individual or group of people.”

The Harris campaign suggested that Trump’s civil war and “Arrest for Treason?” tweets were “blatant threats.” In the past, some have also argued that Trump violated that policy when he tweeted about his “Nuclear Button,” or how North Korea “won’t be around much longer” or ... this

But in the section of Twitter’s rules that bans “accounts that affiliate with organizations that use or promote violence against civilians,” the company has created an exemption for “military or government entities.”

The president of the United States would certainly qualify.

In January 2018, a few days after Trump tweeted about the size of his nuclear button, Twitter outlined how it approaches enforcing its rules against world leaders.

“Twitter is here to serve and help advance the global, public conversation,” the company wrote in a blog post. “Elected world leaders play a critical role in that conversation because of their outsized impact on our society. Blocking a world leader from Twitter or removing their controversial Tweets would hide important information people should be able to see and debate. It would also not silence that leader, but it would certainly hamper necessary discussion around their words and actions.”

Twitter has since tweaked how it handles some rule-breaking tweets in this category. The company announced that it would begin flagging tweets by national political figures with more than 100,000 followers that violate its policies.

As my colleagues Tony Romm and Elizabeth Dwoskin wrote in June, flagged tweets would require users to click on a screen that carries a warning message: “The Twitter Rules about abusive behavior apply to this Tweet. However, Twitter has determined that it may be in the public’s interest for the Tweet to remain available.” Flagged tweets would also circulate to fewer people in Twitter’s algorithms and search.

Twitter confirmed that, to date, it has yet to apply its public interest notice to a rule-breaking tweet.

“Newsworthiness”

Trump tweeted two years ago that North Korea “won’t be around much longer” if it doesn’t change its tune toward the United States, prompting calls to ban Trump for violating Twitter’s rules against physical threats.

Twitter defended its decision to keep the tweet and Trump’s account online, saying that the platform considered “newsworthiness” as a factor while examining potential rule violations.

“We hold all accounts to the same Rules, and consider a number of factors when assessing whether Tweets violate our Rules,” Twitter’s public policy account wrote in a statement. “Among the considerations is ‘newsworthiness’ and whether a Tweet is of public interest. This has long been internal policy and we’ll soon update our public-facing rules to reflect it. We need to do better on this.”

Other platforms have used “newsworthiness” as a factor that limits its enforcement of the rules against major politicians. Last week, Nick Clegg, Facebook’s vice president of global affairs, cited Facebook’s newsworthiness exception in outlining why the company will keep most rule-breaking content from politicians on the site.

“It is not our role to intervene when politicians speak,” Clegg said. “We do not submit speech by politicians to our independent fact-checkers, and we generally allow it on the platform even when it would otherwise breach our normal content rules.”

Twitter’s “newsworthiness” exemption doesn’t necessarily protect every piece of content the president touches. In July, when the president retweeted a conspiracy-theory-laden account claiming that “Democrats are the true enemies of America,” Twitter suspended that account, causing the tweet to disappear from Trump’s otherwise newsworthy timeline.

The rules themselves

Twitter has done a lot of work over the past couple of years to clarify its rules and how they are enforced, but the company still leaves itself room to interpret the rules as it wants in any given situation. And that gives Trump a third loophole — one that doesn’t just apply to the president. Twitter could determine that something he tweeted doesn’t violate its rules.

This is a kind of flexibility social-media companies tend to afford themselves. Reddit chief executive Steve Huffman once alluded to the “specifically vague” approach to writing rules for social platforms. The argument is that sites should always have the leeway to interpret their own rules as needed, because if the rules are too specific, bad actors will be able to get off on a technicalities.

Twitter doesn’t believe that Trump’s tweet about the size and effectiveness of the president’s “Nuclear Button” broke its rules. In a statement at the time to Business Insider, Twitter said the tweet was not a “specific threat” and therefore wasn’t banned under the company’s rule against “specific threats of violence or wishing for serious physical harm, death, or disease to an individual or group of people.”

Twitter declined to comment to The Washington Post on whether the tweets flagged by Harris violate the platform’s rules. But the company acknowledged having received her campaign’s letter and said it intends to respond.

More reading: