Democracy & Elections

Libertarian Panic, Unlawful Action, and the Trump Presidency

Jack Goldsmith
Tuesday, November 22, 2016, 2:28 PM

Since Donald Trump’s election two weeks ago, we have witnessed—on the pages of Lawfare, and throughout mainstream and social media—what my colleague Adrian Vermeule once described as “libertarian panic.” Libertarian panic is the flip side of the more familiar “security panic,” which is essentially a sharp reaction and possibly overreaction to a security event (like a terrorist attack) that leads to actions later deemed excessive or oppressive.

Published by The Lawfare Institute
in Cooperation With

Since Donald Trump’s election two weeks ago, we have witnessed—on the pages of Lawfare, and throughout mainstream and social media—what my colleague Adrian Vermeule once described as “libertarian panic.” Libertarian panic is the flip side of the more familiar “security panic,” which is essentially a sharp reaction and possibly overreaction to a security event (like a terrorist attack) that leads to actions later deemed excessive or oppressive. By contrast, a “libertarian panic” (I adjust Vermeule’s formulation to fit the Trump context) is a sharp reaction and possibly overreaction to perceived or anticipated security measures that are believed to represent unjustified attempts to violate the law or curtail civil liberties.

One of Vermeule’s central points is that the same social mechanisms—“cognitive heuristics and biases, various forms of cascading and herding, conformity and preference falsification, and so on”—underlie both forms of panic, with implications that he worked out. My aim here is different. Drawing on some points I made in Power and Constraint, I want to explain why I am much less panicked than my colleagues about Trump ’s pledges for illegal action, and why my colleagues’ panic is one reason why I am not too panicked.

The day after the election, Ben gave this catalogue of concerns about Trump.

  1. his often bizarre, erratic, and egomaniacal behavior that raises serious questions about his management of foreign and military affairs, particularly in a crisis or a situation in which he is insulted;
  2. his oft-expressed illiberal attitudes about religious and ethnic minorities and foreigners;
  3. his promised abuses of power with respect to free speech and the press and in the context of overseas conflict;
  4. the strange affinity he has shown for Russian strongman Vladimir Putin, combined with his campaign’s numerous and unexplained entanglements with Russia and Russia’s intervention in the presidential campaign to his apparent benefit; and
  5. his rejection of mainstream foreign and defense policy thinking on matters as basic as the American commitment to traditional allies.

I agree with Ben that (1) is indeed very worrisome in a Commander in Chief—indeed, it is at the top of the list of things I worry about with President Trump. Stability, prudence, self-control, self-awareness, foresight, constancy, reliability, informed judgment, a thick skin—these are among the essential qualities for a successful Commander in Chief that Trump appears to lack. Any number of things can go very wrong as a result of these deficits. I also worry about (4) and, to a lesser degree, and depending on the issue, (5).

If you are finding Lawfare useful in these times, please consider making a contribution to support what we do.

But my focus in this post is on (2) and (3), about which I have much less worry. The basic concerns here are that Trump will get away with unlawful or abusive action. Specifically, based on Trump’s campaign statements and early appointments, my Lawfare colleagues worry that Trump will (among other things) return to or “normalize” torture, order “indiscriminate attacks” on civilians, abuse the intelligence agencies to engage in political and other forms of unlawful surveillance, “throw his opponent in jail,” create a “Muslim registry,” “censor the internet,” and in general engage in “counterterrorism policies that would violate domestic and international law.”

I have no idea if Trump will attempt to follow through on any of his campaign pledges or suggestions for unlawful action, but if he does I seriously doubt he will succeed.


“He'll sit here," Truman said of Eisenhower, "and he'll say, 'Do this! Do that!' And nothing will happen. Poor Ike — it won't be a bit like the Army. He’ll find it very frustrating.” Richard Neustadt famously built on this story to illustrate that hard power does not by itself translate into effective power. It is a lesson Trump is about to learn.

One hurdle to a president executing his wishes is the powerful permanent bureaucracy in the intelligence and defense communities that transcend administrations. A President’s commands are not self-executing. He has to convince the bureaucracy to go along, and he often has political control only over the tippy-top of the agencies in question. The CIA has only three political appointees—the Director, the Deputy Director, and the General Counsel. I believe that at NSA, only the Director is a political appointee. The DOD has more, but especially in the context of its size, not that many more. President Trump can surely place loyalists in “non-political” spots in these agencies, but the vast majority of people running the Intelligence Community and DOD will be the same people that ran it under Obama. These permanent officials have deep expertise, trans-administration interests and values, and deft infighting skills that enable them to check and narrow the options for even the most aggressive presidents.

The “persistence in the interests and outlook of the national security leadership and especially of the national security bureaucracy,” is one reason why Obama continued so many Bush-era counterterrorism policies, I argued in Power and Constraint. I predict the national security bureaucracy will also be a force for continuity and restraint in the other direction in a Trump administration. Michael Glennon wrote a good book that agreed with my descriptive point about the continuity-preserving power of the bureaucracy and argued that this power was largely illegitimate. Glennon seems to Leon Neyfakh (and to me) to back away from his descriptive thesis in suggesting that Trump but not Obama could steamroll this bureaucracy. I think this is much harder to do than Glennon suggests. Massive bureaucratic firings are politically hugely costly, bureaucratic expertise narrows presidential options in ways the White House cannot control, and the bureaucracy contains many veto points and accountability checks.

One set of such points and checks consists of the hundreds of intelligence community lawyers. I know many are skeptical of lawyers’ checks because of (among other things) some early Bush administration practices, but I believe those practices were ultimately a result of excessive internal executive branch secrecy that I do not think can be replicated today. For reasons I explained at length in Power and Constraint and lack space to reiterate here, IC lawyers today operate in a significantly different legal and bureaucratic and cultural environment than they did in 2001, in large part as a reaction to early Bush-era initiatives. The law governing their operations is clearer and (mostly) more restrictive than in 2001; there are many forms of institutional transparency and redundant accountability built in to executive branch legal deliberations that were not present in 2001; and everyone in government is aware of the painful recriminations suffered by the early Bush lawyers and other officials as a result of their repudiated legal interpretations.

Beyond the lawyers, career officials in the intelligence and defense communities are deeply committed to lawful action to a degree that outsiders do not appreciate. I do not believe that the armed services under Trump will carry out orders to target civilians or to unwind the interrogation constraints in the Army Field Manual, because doing so would be clearly unlawful. I do not believe that the NSA career officials will—as the NSA did in the 1960s and 1970s—engage in domestic political surveillance, because doing so would be clearly unlawful. I do not believe—for reasons I explain more fully elsewhere—that CIA will return to waterboarding. Not only are all of these practices that people worry about unlawful, the culture and interests of these agencies would not permit these practices to occur, I predict.

A final and very important internal check comes from the semi-independent inspector general in each intelligence agency that sees himself or herself as beholden to Congress as the Executive, and has authority to conduct basically any investigation and report those findings to Congress. The CIA Inspector General effectively ended waterboarding long before the practice became public.

External Constraints

An array of actors outside the Executive branch that will also make it hard for Trump to engage in illegal national security action. During the Bush administration, the Press did an extraordinary job of surfacing secret information about controversial or illegal programs that led to reform. I expect this practice to work just as well and probably more robustly to check a Trump presidency. I do not think the Obama-era leak prosecutions will chill such accountability journalism. Journalists can discover a lot without leaks, and leaks persist at an unprecedented rate despite Obama’s actions. Similarly, the civil society groups were hugely consequential during the Bush administration in using FOIA and novel lawsuits and activist campaigns to bring extraordinary change to Bush administration practices on detention and interrogation, among others.

Congress and the courts, spurred on by journalists and civil society, will be consequential as well. Normally we would expect a President whose party controlled Congress to get a free ride from Congress. But that is not what happened with Bush, when a Congress controlled by his party enacted the Detainee Treatment Act of 2005 to raise the bar on interrogation, and did not give Bush some important elements of presidential discretion that he asked for when it enacted the Military Commissions Act of 2006. We can expect much less obeisance from Congress toward Trump, who has a fraught relationship with Republican Party politicians, to put it mildly. One sees this dynamic already in McCain’s red line on waterboarding and Rand Paul’s threats on surveillance.

Finally, there are the courts. These courts stood up the President Bush’s assertions of military prerogative in several famous cases, and can be expected to stand up at least as much to President Trump should he engage in excessive presidential action.

The Distrust Dynamic

Every one of these factors—the permanent bureaucracy, including inspectors general and government lawyers; the press; civil society; Congress; and courts—will operate in much more robust fashion to check President Trump than they did to check President Obama. Presidential actions do not take place in a vacuum, but rather in a context where they are interpreted based on perceptions about the President’s intentions and trustworthiness.

The early Bush administration loudly and proudly proclaimed that it wanted to expand presidential power, and appeared to act on this proclamation in many legal opinions and contexts that scared all of the checking institutions above and led them to push back very hard against Bush in an unprecedented fashion. President Obama came to office with a reputation for a commitment to the rule of law and civil liberties. He continued to talk that talk, even when, with anguish and earnestness, he didn’t exactly walk the walk; and his attorneys made robust Executive branch power arguments couched in narrower and less inflammatory terms than the early Bush lawyers.

As a result of these factors, Obama generally received more solicitous treatment from the press, and civil society, and courts, than did his predecessor, even though Obama too expanded presidential power. Indeed, by the time Bush left office, almost all of his aggressive presidential actions (e.g. interrogation, detention, surveillance, military commissions) were checked, altered, and eventually put on a firm Article I footing and mostly blessed (in one way or another) by courts. By contrast, Obama’s most aggressive presidential actions—the unilateral Article II War in Libya, the giant loophole for hostilities in the War Powers Act, the extension of the AUMF to ISIL—were for the most part not redressed by Congress or courts, and remain as precedents for unilateral presidential action. (This is a nutshell is what Robert Goodin meant, in his famous article on the Nixon-to-China dynamic, when he said that “sometimes its seems that to get left-wing policies, [voters] have to pull the lever marked for the right-wing candidate, and vice versa.”)

The point is that presidential actions, and even presidential precedents, are not judged in a vacuum. They are judged against a background of beliefs about (among other things) the president’s trustworthiness and commitment to the rule of law. In its first term the Bush administration was (as I argued in The Terror Presidency) unprecedentedly indifferent to these factors, and the checking institutions bit back unprecedentedly hard. Trump is already in a much worse position than was Bush on the trustworthiness score. Ben captures this perspective well when he stated why he would treat Trump much differently than his three predecessors:

Trump’s election will fundamentally change my work on this site over the next few years, and probably off the site too. Because at least for me, Trump does not enter office with a presumption of regularity in his work. He does not enter office with a presumption that as President he will pursue a vision of what national security means that is remotely related to my own or that he will do so in a rational fashion—or even that he and I share a common idea of what aspects of this nation we are trying to secure. I take what he has said, over a long period of time now, too seriously for that.

So in a way I never did with George W. Bush or Barack Obama or Bill Clinton—the other presidents I have covered as a journalist or analyst—I will write about the actions of the Trump presidency with the working assumption that our nation must be protected both by and from the president. I will support him and dispassionately analyze policy and law related to his functioning as the lawful executive responsible for the nation’s security. But I will always also keep the sharpest of eyes out for the areas where he himself is the threat and dispassionately analyze policy and law related to the threat he poses.

This attitude will pervade the checking institutions in our constitutional system. The vast majority of the permanent corps in the intelligence and defense bureaucracy will be on edge to ensure that Trump does not violate the law or their values (and, ultimately, their institutional self-interest), and it will leak at the slightest hint of illegal action. (I expect national security leaking, already widespread, to increase a lot under Trump.) Civil Society will raise more money than ever off of Trump anxiety and will use these unprecedented resources to pull out the stops in monitoring and constraining the new President. The Press will not be cowed by Trump’s threats, but to the contrary will be emboldened to report on his national security actions even more aggressively and skeptically then they did under Bush. The Republican Congress will not, in most national security contexts at least, treat Trump with the same presumption of deference that it would normally give a Republican president. (Remember, a Republican Congress did not cow to Bush on interrogation or military commissions.)

And courts especially will be on guard against Trump’s excesses. This is potentially bad news, I think, for military detention and military commissions. Detention law seemed settled in the D.C. Circuit under Obama, but if Trump starts bringing new detainees to GTMO, we can expect more searching legal review on habeas (especially about whether the 2001 AUMF extends to ISIL, but also on factual questions), and a greater likelihood of Supreme Court review. Military Commissions have won a number of important legal victories in the D.C. Circuit this year and seem finally to be turning the corner toward trial of the 9/11 plotters. But federal courts will be more searching and skeptical about those commissions under Trump than under Obama. Trump will lose many legal battles in court that a rhetorically more moderate president would win.

In short, Trump’s seeming indifference to the rule of law and his pledges to act unlawfully will cause the checking institutions to judge all of his actions with much greater scrutiny and skepticism. Chest-thumping about expanding presidential power was self-defeating in the Bush administration and will be more so in the Trump administration.

The Virtues of Libertarian Panic

These are some of the reasons why, even assuming the worst about Trump, I believe that libertarian panic about brazen law violation is largely unjustified. As Dan Drezner suggests in a related context, one trouble with libertarian panic (my term, not his) is that it is hard to know if it is exaggerated or not until it is too late. Drezner suggests that one danger of seemingly unjustified panic may be a perverse wolf-crying effect before some audiences. While Drezner does not go there, it is easy to imagine that a constantly loudly beeping panic alarm causes many to tune out, which in turn allows more bad actions to occur than would have been the case with a more targeted and discriminatory reaction. Moreover, I have heard many people claim that overheated reactions to Trump’s campaign rhetoric, especially among elites, helped Trump quite a lot during the campaign.

I have no confidence in my ability to assess the validity of these mechanisms, so I won't try. But I am confident that there is an important upside to libertarian panic, even if it is unjustified: Libertarian panic makes the checks outlined above work robustly. (The relative paucity of such panic under Obama, compared to Bush, explains a bit why Obama was able to push the margins of executive power more successfully than Bush). As I wrote in Power and Constraint, after explaining why I did not think waterboarding would ever return:

There is a paradox in the human rights community’s belief that waterboarding is still on the table, despite the unprecedented accountability and pushback of the past decade. That accountability and pushback resulted because human rights organizations, like many others throughout the presidential synopticon, believed that there was inadequate accountability and pushback and fought hard to achieve more. Many in the presidential synopticon, especially in the human rights community, remain alarmed by what they see as endless and undefined war, excessive presidential secrecy, insufficient judicial review of the President’s actions, too much surveillance, inadequate congressional involvement, and many other evils of the post-9/ 11 presidency. They continue to push hard against the government with lawsuits, FOIA requests, accountability campaigns, and strident charges against public officials. This is all very healthy for the presidency and for national security. The continued efficacy of the presidential synopticon depends on just this type of skeptical attitude about the synopticon’s efficacy.

Or as Tim Edgar put it more succinctly when I made a similar point on twitter: “.@jacklgoldsmith says Trump won't get away with the worst abuses, but only if we don't believe @jacklgoldsmith is right about this.”

Libertarian panic is ingrained in our national psyche, and has been one of the secret mechanisms of constraint on Executive power since the Founding. And perhaps even before. Tim pointed out in response to a draft of this piece that “the United States of America is quite literally founded on libertarian panic. The complaints against the King in the Declaration of Independence are not a fair indictment of British policy towards the colonies; its list of ‘repeated injuries and usurpations’ is highly polemical and unfair, designed to put the British government in the worst possible light.”


The thoughts above should not be construed as a prediction that President Trump won’t do very bad things on the national security and intelligence fronts. I expect and predict that he will do very bad things, primarily due to personal qualities that I think make him uniquely poorly suited to be Commander in Chief.

Also, a president acting within his lawful discretion can take action (including action related to war—the Obama precedents are robust), issue statements, make findings, and do many other things than can seriously harm our nation and other nations. Some of the checking functions I describe can temper lawful but irresponsible presidential action, but my confidence in these checks in the context of lawful discretion is much reduced. We depend on a President’s good faith and judgment in exercising lawful discretion to an extent that most of us have not previously appreciated.

More generally, the checking functions I described assume that the President wants to change the status quo and that various institutions resist. But Presidents can also do harmful things by not initiating action—such as, for example, when he refuses to initiate action needed to prevent a crisis, or when he fails to condemn (and thus gives aid to) certain forms of abusive private actions.

In addition, and obviously, Trump’s mere pledges to violate the law—to waterboard, or kill civilians, or discriminate on the basis of religion or nationality, etc.—can have a devastating impact on U.S. interests, at home and abroad, even if he is unwilling or unable to carry them out.

There will also be a great deal of dispute in the coming years, I realize, about what is lawful and what is not lawful. In some ways waterboarding and targeting civilians are easy cases because they are so clearly unlawful. But many things that president Trump might try to do—for example, certain types of intelligence investigations of Muslim organizations with alleged ties to Islamist terrorism, or certain types of tax investigations of disfavored groups; there are a million other hypothetical examples—will be legally contested, depending on how it is carried out and where one sits. I have no doubt that President Trump—like all Presidents—will in some contexts test the line of legality and see how far he can move it to serve his values and interests. And I expect the checks outlined above, infused with libertarian panic, to push back hard. I don’t expect the equilibrium that emerges to be uncontroversial—far from it. But I have more confidence in these checks than my many of my Lawfare colleagues, and I believe that Trump cannot succeed in his more controversial plans at home over the medium term without a chop by and the approval of the other institutions of government.

Finally, I do not rule out unlawful action in the face of an attack amounting to an emergency. History teaches that no set of U.S. government institutions can prevent security panics in the face of truly devastating attack, no matter who the President is. I do believe, however, that even in that event we likely won’t see a return to repudiated practices like waterboarding. As I wrote in Power and Constraint:

A persistent theme in American history is that when Presidents act aggressively to curtail civil liberties at the dawn of a war or an emergency in a way that is later regretted, those regrets are remembered in the next war or emergency and are not repeated. No President has replicated Lincoln’s indiscriminate suspension of the writ of habeas corpus or his military trials for civilians in the United States. We have never again seen loyalty prosecutions as in World War I, or anything like the World War II exclusion of Japanese-Americans. The public criticism, regret, and ostracism that followed these practices stigmatized them and prevented their repetition. The same is likely true for officially sanctioned waterboarding.

So while I don’t want to deny that many awful things might happen under the command of President Trump, I nonetheless believe he won’t be able to execute the unlawful commands that so many are worried about except in the circumstances so extreme that any President would act aggressively in ways later deemed unlawful.

Short of that situation—and that has not been the context for most Trump anxiety to date—I think the widespread panic about Trump and law compliance in the intelligence and military communities is misplaced. And yet that panic can be useful, indeed vital, even if misplaced.

Jack Goldsmith is the Learned Hand Professor at Harvard Law School, co-founder of Lawfare, and a Non-Resident Senior Fellow at the American Enterprise Institute. Before coming to Harvard, Professor Goldsmith served as Assistant Attorney General, Office of Legal Counsel from 2003-2004, and Special Counsel to the Department of Defense from 2002-2003.

Subscribe to Lawfare