ANDREW CABALLERO-REYNOLDS / AFP / Getty

This president doesn’t speak like other presidents, that much is clear. Since taking office in January 2017, President Donald Trump has used the bully pulpit in ways that break, often dramatically, from the rhetorical norms that preceded him. The president seemed to cross a new red line this week, as he took to Twitter to suggest—without legal foundation—postponing the November election. This latest rhetorical escalation has increased the urgency of a long-simmering question: Can anything be done to rein in the speech of a president unmoored from reality and unmoved by decency?

The answer is yes, and it hinges on understanding both the nature of presidential speech, and that speech’s dependence on a variety of mechanisms for actually reaching the public.

Of course, how a president speaks isn’t fixed; for more than a century, presidents have used evolving channels and styles of communication to speak to the nation, to shape public opinion, and to rally support for their positions. And however they choose to reach the public, presidents do not go it alone; they rely on intermediaries, who are crucial to the amplification of presidential messages.

Teddy Roosevelt, who coined the term bully pulpit (by which he meant, in the usage of the time, something like “great platform”), well understood this. He was a master of developing close and reciprocal relationships with members of the press, which was crucial to turning his vision of progressive reform into concrete policy. Decades later, a different Roosevelt—Franklin Delano—did this in his own way, guided by the technology of his time, making novel use of the radio as he attempted to steer the country through war and depression. And Ronald Reagan, drawing on his background in Hollywood, tapped the power of television as well as the print media to articulate and advance a very different vision of the role of government in American life.

As these examples underscore, presidential speech is not self-amplifying. People often talk about the bully pulpit as though the president simply possesses it by virtue of the officeas if it’s a megaphone handed to him along with the nuclear codes. But that formulation obscures the presence of a cast of intermediaries, and it overlooks their very real agency.

While members of the press and other intermediaries have more often than not served as willing participants in conveying presidential messages, there’s nothing inevitable about that. And in fact, when a president’s speech becomes sufficiently dangerous and destructive, intermediaries and other institutional players can choose not to assist in amplifying it—and can even engage in counter-speech of their own. Over the past few months, as President Trump’s rhetoric has appeared more and more to threaten discrete groups and basic rule-of-law values, intermediaries and other individuals have begun to respond in new ways. Twitter has started labeling presidential tweets it deems sufficiently false or inciting. News networks are choosing not to broadcast the president’s coronavirus briefings live. And the print media have begun to move beyond euphemism in describing the president’s lies.

These are important developments. Rhetoric is a key source of presidential power, but that power is not self-executing, and it may be subject to checks from unexpected quarters.

That President Trump would speak in a different register from previous presidents has been clear since at least as far back as his dark and dystopian inaugural address. He has addressed himself to a subset of the political community, not all of it; he has trafficked in fear and division, not uplift and unity; he has used his public platform less to offer and defend a policy agenda than to attack targets, which have included journalists, current and former executive-branch officials, private individuals, and federal judges. A recent New York Times analysis found that more than half of the president’s tweets from January 2017 to November 2019 were attacks. He has also been a font of lies and misinformation; at the time of this writing, The Washington Post’s running tally counts more than 20,000 presidential lies over the past three and a half years.

The president’s speech is, of course, not merely speech. Some of his words have had significant material consequences—entirely aside from the administration’s policy initiatives. A recent ABC News analysis identified 54 acts of violence in which the perpetrator explicitly invoked President Trump. The president’s bellicose foreign-policy rhetoric has worsened relations with adversaries like Iran and North Korea, and alienated allies across the globe. His spurious charges of absentee-voting fraud may already be causing states to underprepare for a heavily absentee-ballot-based November general election, will likely deter many voters from casting absentee ballots, and appear to many to be designed to sow doubts about the integrity of election results. His dangerous rhetoric around COVID-19, including his promotion of unproven or discredited treatment methods and his shifting messaging regarding basic public-health measures such as mask wearing, has no doubt impacted the behavior of private citizens and public officials, with the potential to result in tens of thousands of additional preventable deaths.  

These last two points are key. With the country still in the grip of a once-in-a-century pandemic,and an election ahead in which polls show the president down nationwide against Joe Biden, the president’s always-aberrant rhetorical style is regularly traversing more dangerous, destructive, and demagogic terrain. And as the potentially catastrophic consequences of the president’s rhetoric come into sharper focus, so too do some of the ways individuals and institutions might respond.

Twitter—the company, not its millions of users—has recently emerged as an important player in this sphere. From early in his administration, the president has proudly claimed he uses Twitter to take his message “directly” to the American people. But Twitter is not a purely passive conveyor of information, and it has recently begun, in small but significant ways, to do to the president’s Twitter feed some version of what it has long done to others’: moderate. Twitter’s first intervention took the form of a warning flag it appended to two presidential tweets that contained misinformation about mail-in ballots: One claimed that mail-in ballots would lead to a “rigged election”; the other falsely claimed that California would send ballots to all individuals living in the state, when only registered voters will receive such ballots. The president responded to this warning with a dashed-off attempt to use the apparatus of government to bend Twitter to his will, issuing an executive order calling on federal agencies to revise their interpretations of Section 230 of the Communications Decency Act, which has long provided a liability shield to internet platforms, including Twitter. The order, though of dubious constitutionality and in any event largely hortatory, was likely intended, at least in part, to dissuade Twitter from any further salvos. But the company doesn’t appear to have been deterred; since the issuance of the EO, the platform has only ratcheted up its responses to the president, including placing warning labels atop presidential tweets that violate the site’s rules about threatening harm against identifiable groups.  The labels require users to click in order to access the content.

Given Twitter’s centrality to Trump’s presidential ethos, this was a significant development. And although Facebook has so far conspicuously declined to label any of the president’s postings there, it is facing significant pressure, including from within, that could well lead it to change course.

Traditional media, too, have begun to function as something other than passive amplifiers of presidential utterances. Consider the decision by a number of networks to stop broadcasting the president’s coronavirus briefings live. The briefings “rated,” in the industry’s term, but by mid-April, a number of networks had stopped carrying them, perhaps in response to the president downplaying the threat of the virus, his promotion of unproven treatment methods such as hydroxychloroquine, and his outlandish suggestions that disinfectants or ultraviolet light be applied directly to (or within) the body.

The press has also, after initially wrestling with how to treat a president so prone to untruths, begun to call out his lies directly. Trump began his presidency with a flurry of lies about the 2016 election, insisting that his massive popular-vote loss was attributable to millions of fraudulently cast votes. Even at the time, some outlets, including The New York Times, referred to those claims as lies. But the stakes of those backward-looking lies were, in national terms, relatively low: the president’s vanity, perhaps aspects of his legacy. As the president has begun deploying similar rhetoric in the context of the approaching November election, the stakes are far higher: A president who lies about an upcoming election in ways he perceives as advantageous to his electoral prospects might move from rhetoric to action, including the use of federal resources, to shore up his electoral prospects. So it is all the more pressing that the president’s false claims about voting and this election not be permitted to stand unchallenged.

Beyond online platforms and traditional media, we have seen the emergence of an important kind of counter-speech from senior members of the president’s own administration, who have publicly resisted some of the president’s rhetoric around both the protests against police violence still roiling the country, and COVID-19. After the president, accompanied by senior military leadership, forcibly cleared Lafayette Square of protesters for a photo op, Chairman of the Joint Chiefs of Staff Mark Milley issued an extraordinary public apology for his presence at the event, one that strongly, if implicitly, criticized the president’s decisions. Defense Secretary Mark Esper also broke publicly with the president, expressing his opposition to use of the Insurrection Act, which the president had threatened to invoke to quell protests. Soon after that, former Defense Secretary Jim Mattis issued a blistering statement describing Trump as an existential threat to the Constitution.

Members of the president’s own administration have also resisted some of his messaging around the coronavirus: Anthony Fauci has criticized the president’s language about the virus, as well as his endorsement of unproven treatment methods. Looking beyond the pandemic, last month saw the extraordinary testimony of two sitting Department of Justice officials, who described a culture of cronyism and political favoritism inside the DOJ. The former Roger Stone prosecutor Aaron Zelinsky detailed the sequence of events surrounding the department’s decision to cast aside ordinary DOJ protocols and principles—partly in response to a 3 a.m. presidential tweet—in order to provide Stone with preferential treatment. And the career lawyer John Elias described a baseless antitrust probe into an auto-emissions agreement, opened solely in response to presidential rage tweets.

Not every corner of government or civil society has risen to the occasion. Congress has to date shown little interest in engaging explicitly with the president’s rhetoric. In late 2017, Representative Al Green introduced articles of impeachment charging President Trump with supporting “white supremacy, bigotry, racism, anti-Semitism, white nationalism [and] neo-Nazism,” and accused him of “inciting hate and hostility” by “sowing discord among the people of the United States, on the basis of race, national origin, religion, gender, [and] sexual orientation.”

But those articles were quietly voted down in the summer of 2019. Although the House did formally impeach the president last fall, the impeachment charges that were actually approved by vote made no reference to the president’s public speech; such speech was similarly sidelined in the House proceedings and Senate trial. This was a missed opportunity: Presidential speech has played a significant role in every other major presidential impeachment effort in our history, most notably in the case of Andrew Johnson, whose dangerous and demagogic rhetoric actually formed the basis of the tenth article of impeachment approved by the House of Representatives against him. With rare exceptions, members of the president’s political party in Congress have proved remarkably unwilling to use their own speech platforms to condemn or even criticize the president’s rhetoric.

And courts have so far had a mixed record. The Supreme Court has been all too willing to set aside presidential speech as immaterial—in the 2018 travel-ban case, for example, and in this term’s Deferred Action for Childhood Arrivals decision, in which the majority, although setting aside the DACA rescission as inconsistent with the Administrative Procedure Act, also rejected the argument that the rescission may have been driven by discrimination. But a number of lower courts have examined presidential speech in evaluating the motives of various presidential actions. And back in 2018, Chief Justice John Roberts issued a pointed statement condemning presidential rhetoric: After President Trump criticized a district-court ruling as coming from “an Obama judge,” Roberts responded in an unusual rebuke, in which he insisted, “We do not have Obama judges or Trump judges, Bush judges or Clinton judges. What we have is an extraordinary group of dedicated judges doing their level best to do equal right to those appearing before them.”

Trump’s suggestion that he might seek to postpone the November election has drawn criticism from new quarters, including harsh condemnation in an op-ed by the prominent conservative and Federalist Society co-founder Steve Calabresi, who described the tweet as “fascistic” and suggested it was “grounds for the president’s immediate impeachment again by the House of Representatives and his removal from office by the Senate.” Whether or not it will take another run at presidential impeachment predicated in part on the president’s words, Congress does have other tools at its disposal, and it could use those to respond to the president’s dangerous and destructive speech. Just a few defecting members of the president’s party in the Senate could place holds on nominees, or block all legislation, if the president again uses threats of force or actual force against peaceful protesters. Congress could agree to provide additional funding to states seeking to facilitate safe voting during the pandemic, ensure adequate funding to the post office (and engage in active oversight of that agency, in light of grave concerns about its political leadership), and lawmakers could use their individual speech platforms to encourage participation in the election, either by absentee or in-person voting, depending on individual circumstances. Courts, too, can heed Justice Sonia Sotomayor’s concurrence in the DACA case, in which she criticized the majority’s “blinkered approach” of disregarding presidential speech.

What all this amounts to is that although the president has a uniquely powerful speech platform, that platform’s power relies on a number of relationships whose terms can be revised and renegotiated. Internet forums do not need to uncritically convey every presidential utterance, no matter how destructive or antidemocratic. News networks do not need to cede the airwaves to press conferences that obscure and confuse more than they illuminate or educate. Print media do not need to deploy euphemisms and indirection when describing the speech of the most powerful person in the country. Neither Congress nor the courts need to avoid confronting head-on the legal consequences of the president’s words.

The bully pulpit is understood to give the chief executive a significant advantage over any institutional or political rival in the constitutional firmament. And under normal circumstances, it does. But these are not normal circumstances, and other institutions already have the tools to respond to this president’s aberrant rhetoric. They just need to use them.

This story is part of the project “The Battle for the Constitution,” in partnership with the National Constitution Center.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.