As we approach the one-year anniversary of the National Guard’s occupation of Washington, D.C., it’s worth pausing to learn what D.C. resistance has to teach us — especially heading into the midterms.

Many of us are carrying real fear about how far Trump might go to interfere with this election. Our fear isn’t unwarranted. But it doesn’t have to be ungrounded, either. Yes, he will say outrageous things. Yes, the heads of the Department of Justice and FBI are trying to keep power for their boss and intimidate voter rights groups. Yes, if Trump could stop elections he would.

But an authoritarian who has to sabotage an election is doing it from a place of weakness. Trump is not Narendra Modi, who steadily gained popularity (until a grassroots “cockroach” movement turned the tables). Trump is in political freefall with cratering poll numbers and is lashing out; his only path to retain power is undermining the will of voters.

Much of his weakness now has to do with how our movements have harnessed backfire. Trump’s use of ICE is limited by the fierce national resistance to his paramilitary. Backlash against the voter suppression of Callais brought Black voters in Alabama and other states to the polls in force. Trump’s power of controlling narrative isn’t as strong as it used to be. Even Fox News avoided re-airing his recent rant about the 2020 election. Admittedly we have more to do, but he is overreaching — his seizure of ballots in Fulton County in January to boost his Big Lie only woke up more people in the election protection space to the threat he posed.

So rather than let our fears run wild, we can ground them in what’s already happened and how people responded. That’s the idea behind my “Election Defense Scenarios” book. It’s an attempt to lay out the scenarios of election sabotage — from National Guard at the polls to disinformation, Department of Justice threats and last-minute rule changes. In many cases, we don’t have to imagine what might happen — we can look to very recent history.

D.C. is one of the clearest examples we have. The city faced a wide array of threats and prepared for the June 16 primary without letting fear set the terms. Here’s what they did.

D.C.’s challenges

Some of the challenges were the ordinary kind that come with any election reform. D.C.’s primary was the first election in the city’s history to use ranked-choice voting, after voters approved Initiative 83 in 2024. That meant there had to be a real voter education lift — organizers needed to teach people how ranking works. And there were the usual worries about longer lines as poll workers and voters adjusted to a new system.

But bigger issues loomed. By election time, D.C. had been living with a federalized National Guard presence for nearly a year, including a month-long federal takeover of the Metropolitan Police Department the previous summer. Ahead of the July 4 “America 250” celebrations, federal officials announced a “summer surge” that added roughly 2,500 more troops, bringing the total in the city to over 5,000.

The threat of the National Guard loomed large. At an estimated cost of over $1 billion per year for Trump’s domestic deployments, the National Guard has harassed citizens and decreased trust in the military. During the 2020 election, organizers say, National Guard troops had been deployed following Black Lives Matter protests and reportedly told people in line they had to get out of line to enforce the city’s curfew — an incident that went undocumented because no one had thought to film it in the moment.

Surprisingly, National Guards were deployed from Democratic-led states, including Michigan (and immediately after the election, Minnesota and Hawaii). At least Minnesota Gov. Tim Walz has since removed its troops.

Using National Guards to interfere with elections was repeatedly threatened. As the Brennan Center reported, “It is illegal to deploy federal troops or armed federal law enforcement to any polling place.” Yet, Trump said, “We’re going to have sheriffs, and we’re going to have law enforcement” at polling locations.

Alongside the troops came fears of political interference — a worry sharpened by D.C.’s unusual constitutional position. Congress controls D.C.’s budget and has repeatedly attempted to gut the city’s local governance. When President Trump was asked what he’d do if Ward 4 Councilmember Janeese Lewis George — a self-described democratic socialist whose campaign was frequently compared to Zohran Mamdani’s — won the mayoral primary, he answered, “I wouldn’t like it — and maybe we take back Washington, run it on the federal basis. We won’t put up with it.” Organizers couldn’t dismiss the threat as bluster. It had legal teeth: Congress and the president hold unique levers over the city.

Each location has different conditions and scenarios. But no matter where you are, we can still learn a lot from what D.C. did to make their election a success.

The GOTV machine kept running

Groups with get-out-the-vote chops, like grassroots movement Free DC, ran the playbook for turning out voters. They did things like matching voter files to canvassing tools, texting unregistered residents and working precinct by precinct to make the case for coming out to vote. Free DC in particular leaned into its base-building structure. It gave volunteers the tools to text out surveys asking people when they planned to vote, then follow up with 20 people closest to them whom they’d personally commit to reaching.

It’s classic deep-organizing infrastructure, repurposed for a high-fear election environment. Because the threats loomed so large, many organizers wondered if turnout would sink. In the end, good organizing plus frustration over being occupied for nearly a year drove the highest primary turnout since 1994.

But turnout operations don’t answer the question: What do you do about troops near the polls?

Build a coalition where none existed

D.C. needed more robust election protection infrastructure than it had before. The DC Election Protection Coalition came together under the Lawyers’ Committee for Civil Rights Under Law and the Advancement Project, alongside Free DC, Harriet’s Wildest Dreams, the Washington Interfaith Network and more than 100 other partner organizations.

They trained hundreds of nonpartisan volunteers, who were divided into two tracks: election protection field attorneys (for people with legal backgrounds) and grassroots poll monitors (for everyone else). Both were stationed outside voting locations across all eight wards to answer questions and help resolve problems as they came up.

Long before Election Day, National Guard troops became one of those problems. The official election protection hotline fielded a report that the National Guard was outside a drop off box in the Foggy Bottom neighborhood. Keya Chatterjee, co-founder of Free DC, decided to go check out the situation. When she arrived, she immediately whipped out her phone, capturing a video of armed Florida National Guards members right outside one of D.C.’s 450-pound, six-foot, double-locked ballot boxes. The Guard members said they had not been told to stay away and showed no knowledge they were apparently infringing on the law.

Keya’s documentation forced a response. After footage was shared with the D.C. Attorney General and Board of Elections, they opened a direct line to the Joint Task Force overseeing Guard operations in the city — and got a written commitment that troops would keep a 50-foot distance from polling locations on Election Day (an affirmation, at least, of what the law says).

This move did several things. Yes, it smoothed the way for a cleaner election. But the video confrontation went viral and helped model handling fear in productive ways. Seeing Keya talk directly to the National Guard showed residents how to stick up for their rights — a behavior a few folks would end up having to do on Election Day.

Train monitors

Everyone in the coalition wanted to reduce people’s fear. They didn’t always agree on how.

The traditional election protection approach for communicating with the public tends to minimize threats. The instinct is to keep messaging that elections are being run freely and fairly. But D.C. residents are already living the violence of Trump’s occupation.

Federal marshals shot and killed Julian Bailey in northeast D.C. in February. Car pursuits by the U.S. Park Police resulted in at least two deaths, including bystander Nolberto Meza, after the agency loosened restrictions on high-speed chases. And a damning report from Brookings Institution shows that nearly 7 in 1,000 children in D.C. have had a parent snatched away by ICE or other federal agents. In such a context, speaking to people’s lived experience is more reassuring than disguising the ugly truth.

The advice to name the threats maps onto recent messaging research. Following extensive qualitative and quantitative testing, ASO Communications and others found, “speaking about the anti-voter agenda increases self-reported desire among pro-voter Americans to feel concern over this agenda (+5-10 [percent] over the course of the survey), to engage in politics (70 percent are more motivated), and to motivate friends to vote (+5 [percent] over survey). We can frame voting as an act of resistance.” They noted that concrete descriptors about how the regime is keeping people from voting created the greatest engagement.

The DC Election Protection Coalition took up that approach as they trained their monitors.

The two approaches weren’t always perfectly in sync. Some election protection trainers taught volunteers to document, document, document if they saw Guard troops acting illegally — take horizontal videos for the widest capturing of information and file a formal incident report. By contrast (even sometimes in the same training!) grassroots groups’ guidance put the emphasis on making sure the voter in front of you got to vote right now — heckle, press, advocate directly for the Guard to leave, rather than only documenting for later.

Expand the circle

Election protection efforts are too often narrowly built around technical expertise in election work. By building the coalition with grassroots mobilizers at the table, it brought in the expertise of those who knew how to rapidly mobilize. To combat the often fractured disconnect between grassroots election work and traditional election protection, Freedom Trainers and All Voting is Local, as one example, have teamed up to host cross-silo sessions between these groups to strengthen ties.

One of the D.C. coalition’s sharper moves was pulling in local rapid response and mutual aid networks so they were feeding into the same control room as the election protection operation. That meant in addition to information like “which polling place has a printer problem,” groups like Migrant Solidarity Mutual Aid fed into a citywide picture, including neighborhood locations of ICE activity and checkpoints. That meant volunteers had better information about what was going on across the neighborhood that day.

The day of the election

June 16 brought the ordinary hiccups of any election — printers breaking, standard supply issues — but D.C.’s neighborhood-level organizing infrastructure meant problems got solved fast. In the summer heat, one neighborhood network located a leftover cooler of water and got it to a polling site to help overheated voters and volunteers. In Columbia Heights and the other seven sites approved to stay open past the normal closing time because of long lines, volunteers swarmed with snacks and drinks to help people stay in line as long as it took.

There were also multiple National Guard interactions that went unreported. According to Keya Chatterjee, West End Library did have National Guards come into the poll site. Monitors were already at the site and immediately began telling them to leave and filming the confrontation with armed troops. Monitors repeatedly insisted the military had no legal basis to be there, and after forceful haranguing, the troops left without further incident.

A decision then had to be made: Do we release this or not? I can’t speak to the exact decision-making, but the reports of the National Guard at polling sites appeared sporadic (not a coordinated threat) and had limited impact because monitors moved fast. It seems voters still voted.

All this affirms the mantra of “record, record, record.” Even if it ends up not being released to the public, getting the facts on record is critical.

The plans that didn’t happen

Because ranked-choice tabulation can take days if no candidate wins a first-choice majority, organizers built a “patience campaign” in advance — a plan to spend up to a week reminding people that a delayed result is not a stolen one, and that counting ballots correctly takes time.

They also built a second, sharper contingency in case results were seriously contested or tampered with: a template borrowed from their experience organizing on May Day 2026. People were told if things went seriously wrong, to recreate the May Day activity — morning intersection blockades, a midday march to the Capitol, evening gatherings — all under a “no work, no school, no shopping” banner.

It did not have to be activated. But having a real plan for nonviolent noncooperation ready was itself part of what let residents and organizers approach Election Day with confidence instead of dread.

The threat isn’t over — and neither is the lesson

Trump’s “take back Washington” threat hasn’t gone away just because the primary is over. Bills to weaken D.C. home rule continue through Congress, and the National Guard’s presence has been extended through the end of Trump’s current term — to the next presidential inauguration in January 2029.

D.C. has to take the takeover threat seriously, knowing not all threats will materialize.

As D.C. is marking one year of occupation, Free DC is inviting organizations and elected officials to sign on in opposition to the troops’ presence. The ask uses simple messaging like, “Publicly oppose any effort to prevent the seating of duly elected officials representing D.C. or any other jurisdiction.” And from individuals, Free DC is seeking support asking governors to get their troops out of D.C..

But D.C.’s story matters well beyond D.C. What happens in D.C. affects us all. And heading into the midterms, what happened during the primary is some of the clearest evidence we have for how to prepare for what’s coming.

Folks working on election protection need to get more comfortable with confrontation. A recent guide on responding to National Guard at polling places somehow left out D.C.’s lessons entirely. They said, yes, document — but there was no mention of confronting the National Guard as D.C. monitors did, telling troops they aren’t legally allowed to be here and insisting that they leave.

On the other hand, those of us coming out of organizing need to learn and respect the rules that already govern election work, and recognize we’re not the first people to think seriously about defending an election. There’s a deep well of experience to draw on, from movements across the U.S. and around the world — which is a lot of why I wrote the “Election Defense Scenarios” booklet with stories from each scenario.

What’s clear throughout is that we can defend our elections. Don’t get bogged down by the noise and while you can take every threat seriously, it doesn’t mean they’re all going to happen — and they may be calculated just to slow you down. So keep on the work, building with people across silos and teaching people to stand up for their rights.