Attorney General Merrick Garland is not going to save democracy. Nor is the attorney general of New York, Letitia James; the Manhattan district attorney, Alvin Bragg; nor the Fulton County district attorney, Fani Willis. As the apparent collapse of the New York district attorney’s investigation makes clear, criminal cases are hard to make. Donald Trump, despite his many seemingly criminal acts, is unlikely to ever spend a day in jail.
Observers of the Trump malignancy have an unfortunate habit of wish casting—believing that their most optimistic fantasies will become reality. They did this with the Mueller investigation—remember “It’s Mueller Time”?—and they did it with both of Trump’s impeachments. Their dream has always been that somehow, somewhere, someone would call Trump to account for his actions and, in doing so, save American democracy.
Today, many invest the ongoing criminal investigations of Trump in New York, Georgia, and Washington, D.C., with the same hopes. Even my good friend George Conway has speculated that this time things might be different.
I don’t see it happening. Please don’t misunderstand; I am as convinced as anyone of the criminality of Trump’s conduct, and nothing would please me more than to see him get his deserved comeuppance. He should, and very well may, be indicted in one or more of these jurisdictions. And the civil suits against him may have legs.
But years of experience prosecuting fraught political cases (and defending others) has taught me that the criminal law is a blunt tool for achieving justice and a poor means of resolving political issues. In my judgment, the chances that Trump will be convicted of any crime are slim to none. And though I am no political analyst, my guess is that a failure to convict will only embolden him and his followers.
Why won’t the criminal law suffice?
To begin with, one should not assume that the cases against Trump will be easy to prove substantively. There is a great deal of evidence of criminality on the public record, but having that evidence and presenting it convincingly in court are different things. We should not assume that the evidence is clear and that any vote for acquittal would be the unreasoning verdict of a Trumpist holdout. I don’t think that’s an accurate assessment. In truth, the putative cases against Trump are difficult ones that would be hard charges to bring even in the best of circumstances, much less before a highly politicized jury pool.
Consider the New York case. As far as one can tell from public reporting, the kernel of the case is that Trump lied in his financial statements. He allegedly inflated his net worth and the worth of his properties to secure loans while, at the same time, deflating their value to avoid taxes. The stark inconsistency is evidence of fraud. In addition, it appears that the Trump Organization compensated some of its employees through nonwage benefits such as apartments, cars, and tuition, and that those benefits were not declared by the employees or the company.
All of that is serious stuff, to be sure. But it is also, sadly, very common. Even if we accept that the case against Trump is strong enough to be charged (and the recent Mazars disclosures suggest that it is), the simple fact is that fraud of this sort is commonplace in the real-estate market, and failing to declare nonwage benefits is, likewise, a common occurrence. Trump’s criminality may well be greater in degree than that of others, but it is unlikely to be different in kind (at least not on the evidence we can currently see on the public record). To my mind (and also, apparently, to the mind of the Manhattan district attorney), that’s a pretty good defense, as is the lack of an insider who is willing to testify against Trump. If Trump can show that his practices are “common” in the real-estate community, that will make his “They’re picking on me” argument much stronger.
Likewise, a federal case against Trump related to the January 6 insurrection—if the Justice Department is indeed investigating one, though we do not know that it is—would be a challenging one. The main complaint, in reality, is both that Trump incited the riot and that he then did nothing to stop it once it got started. But as the former federal prosecutor Randall Eliason has said, making a criminal case out of inaction is difficult. To do that you have to prove a fairly close connection between a duty to act and a failure to do so. Though it is right and correct to say that Trump had a responsibility to act to forestall the riots, having at least partially contributed to their instigation, that’s as much a statement of moral culpability as it is a statement about a legal requirement.
And so Trump’s failure to act is less a crime in itself than proof of his intent—but his intent to do what? Obstruct Congress and prevent the counting of electoral ballots? Engage in a conspiracy to overthrow the government? Powerful suggestions to be sure (and the congressional committee investigating these events has argued that there is sufficient evidence to investigate these criminal allegations), but they run up against the evidentiary problem that all of the violent actions were taken by others. And, at least so far, we have no evidence linking Trump directly to planning the violence (as opposed to planning the rally). There are plenty of circumstantial reasons to think he welcomed the coup attempt, and from this one may infer his criminal intent. But inferences from circumstantial evidence tend to be weaker than direct proof. And here, at least so far, direct proof of Trump’s intent is lacking. As in the New York case, no witness has stepped forward to say that Trump commanded or ordered the insurrection.
This is not a new problem. It is how Mafia dons direct action without getting their hands dirty—by indirection and suggestion. But the dons do it that way for a reason: It makes potential criminal cases against them difficult to prove. Trump can say he was just joking and didn’t want any violence. Again, these cases can be won, and often are, based on inferences from provable facts. But unless a witness from inside the room steps forward and convincingly testifies that Trump wanted the violence and tried to start it, a juror might well find ambiguity and reasonable doubt aplenty.
The Georgia election case is perhaps stronger (after all, Trump is on tape asking the Georgia secretary of state to “find” him some votes). The problem here is likely also one of persuasion. We all know it is Trump on the tape, but someone has to authenticate the tape, and someone else has to explain what the request means and why it was significant. Sadly, public reports suggest, again, that none of the witnesses wants to testify. To be sure, they can be compelled. But here, too, the reluctant-witness syndrome may save Trump. (And yes, if you have begun to see a theme, you are right. Loyalty protects. The only consolation is that once the dam breaks, the trickle of information becomes a flood. The problem is that, at least so far, the dam is holding.)
In short, substantively, none of these cases is a slam dunk. Are they provable in court? For sure. And were Trump an ordinary defendant, these quanta of proof might suffice.
But Trump is no ordinary defendant, and that makes all the difference. Criminal cases are hard to prove generally. High-profile criminal convictions are even more difficult to pull off, and especially so when a case has political overtones. Many political defendants have resources that exceed those of the prosecution. They also have a platform from which they can trumpet their innocence, controlling the media coverage and shaping public opinion in ways that the prosecution cannot.
But perhaps most significant, criminal defendants with public profiles typically enjoy a more robust presumption of innocence than everyday criminals do. In theory, of course, everyone who is charged with a crime has the same presumption. But the grim reality is that, for many quotidian crimes, the prosecutor has the substantial advantage of knowing that this isn’t true. For almost every case I ever brought, I could be sure that the jury began the trial thinking, There has to be something here or they wouldn’t be wasting my time.
Not so in political cases. In those trials, some jury members are likely to start out with the belief that the prosecution is an ill-founded witch hunt, motivated by politics and not the pursuit of justice. And remember: Criminal convictions require unanimity. All that Trump needs is for one single juror to have an unchangeable pro-Trump bias and Trump will walk, no matter what the evidence is.
Now, it isn’t an impossible task to find an unbiased jury, but jurors sometimes lie to get on juries if they think doing so is important. And though the urban venues in question—New York, Atlanta, and D.C.—probably do strongly lean anti-Trump, they are awfully big cities. Each has hundreds of thousands of pro-Trump citizens. The risk of a hung jury is, in my judgment, very, very real. To be sure, an outright acquittal seems unlikely, but an ambiguous result is as good as a win for Trump. After all, a hung jury is as good as an acquittal in the short term.
Consider as well the evidentiary restrictions of a criminal trial. These are the rules that limit the facts that can be put before the jury. These rules have a good reason and a solid purpose: The right to confront the evidence against you in court is grounded in the constitutional guarantee of a fair trial.
But that often means that useful evidence is excluded from the courtroom. Courts don’t accept hearsay. They don’t let witnesses testify to what someone else told them was true. They don’t allow anonymous witnesses.
Imagining how these limitations will play out in the Trump trials is easy. Some of what we know about what happened, we know only indirectly. To cite but one example, consider the infamous meeting among Trump, Vice President Mike Pence, and one of Trump’s advisers, John Eastman, just before January 6. Almost all of what we know about that meeting is from external sources—such as the legal memo Eastman handed out that is now public. And those who were there will be (at best) unwilling witnesses. It seems almost certain, for example, that Pence will do everything he can to avoid being called as a witness, and that if he fails he will be a reluctant one. The story of what happened that day is well known, but without the cooperation of eyewitnesses it will be difficult to prove.
In describing the criminal process so far, I’ve deliberately left out the complexities of these cases in particular, but they are very real. Some of the possible charges, such as seditious conspiracy, have almost never been brought. Others, such as obstruction of an official proceeding, are rare and invoke criminal provisions that are potentially ambiguous in meaning. Though the government has won some notable legal victories so far in the district courts, those victories are, at best, contingent. The appeals courts (and especially the Supreme Court) have historically been far more protective of politicians than the lower courts have been. Even if the jury in a lower court someday convicts Trump (already implausible, as I have explained), that verdict may not survive when appellate judges review the law.
And don’t get me started on the appeals process itself. In criminal matters it is long and cumbersome. Convictions are not final until the last appeal is determined. So, again, to cite only one possibility, one can readily imagine a second Trump administration dropping the appeals of Trump’s conviction and confessing error. His last administration did it for his cronies; there is little doubt he would do it for himself if given the opportunity.
Due process is the hallmark of the rule of law. It is justly held up as a bulwark of liberty and a protection against governmental overreach. As Felix Frankfurter put it: “The history of liberty has largely been the history of observance of procedural safeguards.” But just as due process is a shield for the innocent, it is also, in the hands of a malicious actor, a sword to defeat punishment. There are (appropriately) many barriers to criminal conviction, and Trump will take advantage of all of them.
Which leaves society with a challenge. What to do? We count on criminal law to serve as a deterrent to malicious behavior. But Trump seems likely to evade it. Just as the protection of impeachment against mendacity has proven inadequate, criminal law will probably prove the same. And that means that the only real protection against Trump’s malignancy is the ballot box. Don’t invest too much hope in Merrick Garland. Even with the best of will (which I do not doubt he has), he can’t save democracy. Only we can.