Thanh Do / The Atlantic

At 2:18 p.m. ET today, your smartphone probably buzzed and shrieked before displaying a notice that resembled a text message. This was the Federal Emergency Management Agency and Federal Communications Commission’s test of the Wireless Emergency Alerts system (WEA). A test of the Emergency Alert System (EAS), which sends emergency messages to radio and television, followed two minutes later. Both messages clearly indicated that the alert constituted a test and not a real emergency.

The wireless test was a “presidential alert,” the most serious kind of mobile bulletin U.S. emergency-notification infrastructure supports, and there’s no way for users to opt out. That means everyone with a compatible smartphone got a direct message from the office currently occupied by Donald Trump delivered to their palm, purse, or pocket.

That might thrill or terrify you, but it’s nothing new—the infrastructure to send presidential messages to smartphones has been around since 2012. And since the 1960s, the president has had the ability to directly address the nation live, via all its broadcast channels. But neither facility has been used, not in the past five years or the past 50.

What does the presidential alert mean in the Trump era, and beyond it? The answer is not encouraging. Disquiet about the test, including efforts to power down phones, cancel wireless plans, and otherwise agitate against the very idea of non-optional emergency notices, suggests that even general public safety is not a domain in which Americans can find common ground.

You might have received a WEA alert before, perhaps for an AMBER Alert about a child abduction or as a notice about a local weather emergency, such as a tornado warning. And you’ve probably seen or heard an alert on the television or radio before—especially if you were alive before cable, DVR, and streaming services made television viewable on your own schedule. It used to be called the Emergency Broadcast System (EBS), which was established in 1963 and which replaced CONELRAD, first brought online in 1951 to notify the public in the event of a Soviet attack during the Cold War. In 1997, the system gained the ability to reach broadcasting methods beyond over-the-air signals, such as cable, fiber, and satellite, and became known as EAS.

But by the 2000s, mobile phones were beginning to become universal. Faced with criticism for how the government handled notification and response during Hurricane Katrina, a new system was established, called the Integrated Public Alert and Warning System (IPAWS). IPAWS made multiple government warning systems interoperable. 2006’s Warning, Alert, and Response Network Act allowed carriers and manufacturers the ability to let their subscribers opt out of the alerts, “other than an alert issued by the President.” The fact that many do so only emphasizes how much today’s citizen chafes at unwanted disruption, and how much official, governmental communications have been downplayed in favor of personal or commercial ones.

Both EBS and EAS were expressly designed to allow the president of the United States to deliver a message to the American people in the event of a national crisis or catastrophe. But then the Cold War ended, and the idea of nuclear calamity dissipated, and the emergency-notification systems became nuisances, mostly. For decades, people interacted with the system almost exclusively through its mandated tests (This is a test of the Emergency Broadcast System. This is only a test.) and the occasional television or radio notice, usually about severe weather.

Today’s message was also a test—the first national test of the WEA system since its inception, as mandated by the IPAWS Modernization Act of 2015, which requires such a dry run to be conducted at least every three years.

There’s good reason to test the system. This January, a false alarm in Hawaii sent notice of a “BALLISTIC MISSILE THREAT INBOUND TO HAWAII” to smartphones. A similar but more detailed message was sent to broadcast media, thanks to the integrated IPAWS system. But there was no missile. The message was sent in error amidst a set of tests during a routine shift change. As a result, millions of Hawaiian residents and visitors were convinced that they had met their doom.

Other, less serious confusion about government notice in the age of the smartphone abounds. In addition to the national WEA system, many municipalities operate opt-in, text-message notification services for emergencies, traffic, and other civic inconveniences. On August 27, the system in Washington, D.C., called AlertDC, sent a notice to all its subscribers about a “Presidential Proclamation on the Death of Senator John Sidney McCain III.” The message was sent after the White House instructed federal offices to fly their flags at half-staff to honor the life of McCain, who had died two days earlier. It was lengthy, signed with the president’s name, and marked with the military acronym FYSA, “For Your Situational Awareness.”

The message confused many D.C. residents, who thought that the president was sending a message directly to them. Trump had faced criticism for failing to honor McCain, and the alert felt as if it might amount to an exasperated notice of acquiescence blasted to everyone as if to say, Back off.

But AlertDC is entirely separate from the WEA. Its messages come as ordinary texts, not as governmental alerts, and users who receive them have to sign up for the service. Even so, ordinary smartphone users couldn’t tell the difference. “Why is Trump sending me texts now?” one Beltway local tweeted.

Like the Hawaii alert, this one was caused by human error. A representative from the Homeland Security and Emergency Management Agency, which operates AlertDC, explained to me that the agency operates another distribution system for federal employees and buildings, and this message was meant for that list. It was accidentally sent to the general public.

Overall, people seem to object strongly to the idea that the president could send a message directly to their phones. Today’s test was delayed from September 20, in order to avoid confusion during Hurricane Florence, a real emergency. But in the run-up to that date, some people tried to threaten their wireless carriers or cancel their service. Others set up hashtag campaigns encouraging smartphone users to “go dark” in protest, turning off their phones completely until the window for the alert expired.

The aversion seems partly tied to a distaste for President Trump specifically. For some, that’s because they dispute Trump’s competence or his politics; those who detest Barack Obama might have had the same reaction. For those who dislike him personally, presidential alerts might feel like violations: Donald Trump sliding into the nation’s collective DMs. Unlike televisions and radios, even ones in the home, smartphones are intensely personal devices. They host the moment-to-moment details of their owners’ private life. An unwanted text or phone call is bad enough, but an unwanted presidential invasion might seem beyond the pale.

But that’s not what’s really happening, either. Even though it is labeled a “presidential alert,” these notices are not sent by the president, not directly. Within the emergency infrastructure, the transmission that was tested today is called an Emergency Action Notification, or EAN. The president or a designee would indicate the need for an EAN—but as a senior FEMA official confirmed yesterday, the president does not directly trigger the alert, and the criteria for such an alert are limited by law to national emergencies. The scenarios that would likely result in an EAN are mostly established already through federal interagency operations, pre-scripted in anticipation of possible scenarios, and perhaps subject to adjustment for specific situations. But the president himself would probably not even be involved.

The content of the alert would be authored for specific formats required for IPAWS transmission (WEA messages, for example, are 90 characters or fewer). FEMA officials explained that the systems that transmit it are proprietary, developed specifically for the agency to transmit emergency messages. Put more simply, the process by which that takes place is not as simple as texting or tweeting. There is not an app, like Twitter or Messages or WhatsApp, that the president loads on an iPhone and uses to fire away alert after alert.

On top of that, remember that a real presidential alert has never been issued. Not just in the six years that the WEA system has been able to send text-like messages, but not ever, in the 55 years that the EBS and its successors have existed.

Even so, some worry that if any president would abuse the system, it would be this one. Trump loves Twitter. He sometimes uses it unwisely, and the short WEAs look and feel a lot like tweets. Is it possible? Even hypothetically, Trump would need FEMA’s aid to misuse the service, but in theory, a loyalist could help facilitate it. FEMA Director William Brock Long is under investigation for possible misuse of government funds for travel, a minor matter given the other fires burning in the Trump administration, but also enough to fan the flames of suspicion among those who fear that collusion is possible. Hacking of the system has also been a concern, but external vulnerabilities are different from internal complicity.

Whether Trump, or any future president, could or would abuse the system is perhaps less interesting, and concerning, than the fact that citizens appear to be so easily convinced that a complex, long-standing piece of national infrastructure—one created in the hope that it will not have to be used, rather than that it might be employed regularly—is presumed to be untrustworthy. When the false alarm occurred in Hawaii, people were angry and confused: If an emergency-alert system can notify a whole region about an inbound ballistic missile, it better work correctly. But now that FEMA is carrying out its duty to test that very same system nationally, some lament its very existence.

It’s possible to peg that malaise on Trump and on Trumpism. But as Andrew Facini observed in The Boston Globe, people are sneering at emergency-notification systems partly because presidential credibility was eroding long before Trump, through decades of polarization, partisanship, and scandal. Fears of terrorism have enabled an endless, victor-less war. Rekindled nuclear threats from North Korea and elsewhere have opened the door to proliferation anew. FEMA has leaned into these political circumstances, promoting preparedness while also introducing confusion and stoking fear. Ipaws, WEA, EAS—the general public can’t be expected to keep any of this straight, yet FEMA is foregrounding them for citizen understanding. The agency also might be stoking unnecessary fear: A recently re-aired television public-service announcement from FEMA, ready.gov, and the Ad Council about WEAs showed a family who received one loading themselves up into a Cold War–style outdoor bunker.

But political polarization doesn’t tell the whole story. Technology also plays a role. Smartphones and social media have dulled the public’s expectations for one another, and for public personalities of all kinds, from celebrities to politicians. Services like Twitter and Facebook have proven themselves effective at spreading deceit and misinformation, not to mention fanning the flames of strife even among earnest interlocutors. Trump can’t be blamed for all these ills. In fact, President Obama’s “cool-dad presidency” embraced these and other tools, implicitly endorsing the social infrastructure that helped produce his successor, rather than reining in and controlling the companies that undermine the very idea of a government that seeks to protect its citizens through the technologies and media channels its business leaders have popularized.

Some critics of today’s test encouraged smartphone owners to yoke a protest of the presidential alert with a lamentation for digital obsession. Powering down your phone before 2:18 p.m. ET today would have not only averted the eyes from Trump’s presidential alert, but would also have offered an opportunity for a welcome digital detox from smartphones and the services people access on them.

But that gesture amounts to burying one’s head in the sand. Instead, I hope that today you kept your phone on and nearby. When the system delivered the alert test, binding the White House and FEMA to the carriers and handset manufacturers and software makers, I hope you let its piercing klaxon course sting your ears. There it was, in your hand or on your desk, the glass rectangle that you now trust more than your government and covet more than your safety, screaming at you to realize the obvious: This device that exemplifies suspicion about the government, during today’s test, was also instrumental in causing that distrust in the first place.

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