Dissent Under Pressure

The views expressed are those of the author and do not necessarily reflect the views of ASPA as an organization.

By Parisa Vinzant
July 17, 2026

For the first time, in May 2026, the federal government proposed a blanket, government-wide non-disclosure agreement (NDA) for current and future public servants. This move doesn’t come out of the blue. It’s the latest chapter in an eighteen-month pattern of career public servants pushing back against orders they see as harmful, a pattern that helps explain much of what has unfolded across federal agencies since January 2025.

Rosemary O’Leary’s framework from The Ethics of Dissent: Managing Guerrilla Government offers a way to trace that pattern. O’Leary defines “government guerrillas” as career public servants who resist policies they believe cause real harm, motivated by outrage at organizational wrongdoing rather than self-interest. Her 30 tactics, “Guidelines from Guerrillas to Guerrillas,” range from talking to a supervisor to leaking information to resigning, spanning a spectrum from low-risk moves to high-risk, career-ending ones.

In 2025, public servants went public with formal objections, addressed to agency leadership and to the public alike. At the Environmental Protection Agency (EPA), hundreds of career scientists issued a formal Declaration of Dissent accusing agency leadership of “politicizing the agency.” Nearly 500 National Institutes of Health (NIH) staff signed the Bethesda Declaration. NASA employees followed with the Voyager Declaration in July 2025, signed by 287 current and former staff, including astronauts, warning that budget cuts and changes to safety oversight would cost lives. More than 700 State Department diplomats signed a dissent cable to Secretary Rubio challenging the dismantling of the U.S. Agency for International Development (USAID). Each of these acts fits O’Leary’s category of collective, public resistance and connects with the first tenet of the American Society for Public Administration’s Code of Ethics to “put service to the public above service to oneself.”

The administration’s response was swift and broad in scope. The federal workforce shrank by more than 12 percent by early 2026. The EPA scientists who signed the Declaration of Dissent were suspended or fired. Polygraph tests were administered to identify employees who were talking to the press. In May, the administration proposed government-wide NDAs for federal employees, citing leaks about immigration enforcement as justification. The reclassification of thousands of career civil servants as at-will employees, paired with the NDA proposal, marks a shift narrowing which dissent tactics are still viable.

Public dissent has become riskier, so public servants are likely relying on other O’Leary tactics, including slowing implementation and quietly declining orders deemed unlawful. In June 2026, a 25-year Social Security Administration veteran blew the whistle on Department of Government Efficiency (DOGE) personnel who had pressured the agency to mark 2.7 million living people, mostly immigrants, as deceased in the Death Master File so that “self-deportation” would follow.

As opportunities for internal dissent have narrowed, accountability-seeking has migrated elsewhere, including state legislatures, courts and immigration-focused nonprofits. Eric Schickler’s recent paper cautions against reading this as evidence that civil society has become a stable check on executive power, since this administration has directly targeted universities, foundations, law firms and nonprofits. Immigration nonprofits in particular have faced direct pressure on their institutional standing even as they carry out this accountability work.

It is in the immigration detention space that this shift shows up most clearly. New Mexico’s Immigrant Safety Act, signed in February 2026 and co-sponsored by the American Civil Liberties Union (ACLU), prohibits the state from facilitating federal immigration detention. Colorado’s HB26-1276 and a Washington State law both require greater transparency at detention facilities, including unannounced inspections and public reporting of deaths and emergency calls. In New York, the Dignity Not Detention coalition pushed in a budget provision giving local jails 180 days to end U.S. Immigration and Customs Enforcement (ICE) contracts. In California, the First Amendment Coalition is co-sponsoring SB 423 to require the release of 911 call records from private facilities where a CalMatters investigation found unaddressed reports of sexual violence. The numbers behind these efforts are not theoretical. More than 73,400 people were in Immigration and Customs Enforcement (ICE) detention on a single day by mid-January 2026, a record. Thirty-nine people died in ICE custody during the administration’s first year.

These efforts by state legislatures and nonprofits are coordinated but remain fragmented and underfunded. The Deportation Data Project uses Freedom of Information Act litigation to bring ICE enforcement data into public view through tools like the Vera Institute’s ICE Detention Trends dashboard. Nonprofit legal organizations are representing disciplined federal employees before the Merit Systems Protection Board. This is work the federal government has largely stopped doing, carried out by groups that lack the resources or legal authority federal institutions once held, with uncertain durability. What is clear, though, is that scientists, state and nonprofit leaders and career public servants are filling an accountability gap created as federal oversight capacity has diminished, often at significant personal and institutional cost.

Yet what remains unresolved is whether this decentralized model can adequately fill the centralized accountability gap, much less for how long it could. Public administrators inside federal agencies still have O’Leary’s tactics available, though the room to use them keeps narrowing. Outside those agencies, a multi-state, nonprofit-legislative coalition has taken shape, and its gains, most visible in checking federal detention practices, remain real but fragile. Whether that model can scale and affect change beyond immigration detention is the open question.

(No Ratings Yet)
Loading...

Follow Us!