NRPLUS MEMBER ARTICLE M ore than a few people have responded with bafflement or fury at my view, expressed here and here, that special counsel Jack Smith and the Biden Justice Department should not have indicted Donald Trump over his effort to overturn the 2020 election and the resulting riot at the Capitol on January 6, 2021. Noah has made the case here at NR for this indictment. Let’s walk through a few of the common lines of inquiry or criticism.
Are you saying January 6 wasn’t a crime? That there’s no crime in attempting a coup?
I’m not saying there is or isn’t a crime; I have no issue with charging Trump with a crime (as in the Mar-a-Lago boxes case), so long as he’s committed all the elements of something that was already clearly and unambiguously a crime before now. There are plenty of ways in which a crime could be committed in connection with these events, such as by perjury or the submission of false evidence in a court proceeding. People have actually been charged with real crimes arising from the Capitol riot. None of those have been charged here as crimes committed by Trump, and the efforts to connect Trump to the crimes of others are legally dubious at best.
As I noted in my prior columns, the only possible crime that is actually in the indictment is in paragraph 30, where Trump is charged with signing a false verification in a court case. It seems telling that Smith has not charged this as a stand-alone crime, which he presumably would have done if he thought he could prove it. I will change my mind if Smith is able to hinge his case upon material misrepresentations to a court — but that is not the case he has pleaded, which is premised upon criminalizing a much broader array of conduct.
If, instead of starting with the elements charged, you begin with the assumption that Trump must have committed a crime and try to argue your way to pinning one on him, you fall into the trap that Robert Jackson, when U.S. attorney general in 1940, famously called “the most dangerous power of the prosecutor: that he will pick people that he thinks he should get, rather than cases that need to be prosecuted.” In such a case, he says, “it is not a question of discovering the commission of a crime and then looking for the man who has committed it, it is a question of picking the man and then searching the law books, or putting investigators to work, to pin some offense on him.”
Are you saying Trump isn’t a liar and a criminal?
One of the major arguments made against prosecuting Trump for his challenges to the 2020 election is that Trump gives off every appearance that he actually believes he was robbed. I have little doubt that, at least on an emotional level, he truly has convinced himself that he rightfully should have won and was cheated. Jeff and Jim have addressed some of the issues this presents for prosecuting Trump, which are as much practical as strictly legal. By contrast, Smith’s defenders argue both that Trump must be guilty because he’s generally a known liar, and because he has a specific pattern of crying foul whenever he is losing, regardless of the facts. I dispute neither of those realities.
I may return another day to the legal questions of the standard of criminal intent required in this case, which I do see as a problem, but my issue with this indictment is not principally about intent, it’s about conduct.
Are you saying the indictment is legally defective?
Maybe, but it’s not essential to my argument. My argument is not that there is a slam-dunk legal defense to this indictment, as is the case for the Manhattan DA’s indictment. While the indictment undoubtedly contains some serious legal deficiencies, my argument is that sustaining its theory of criminality requires a significantly aggressive and creative extension of the law. That extension could take us to some very bad places.
That is especially worrisome because such an extension of the law must be erected atop an already existing series of aggressive and creative extensions of the statutes under which Trump was charged. (Andy McCarthy has made the case that these existing extensions have already taken those laws too far in ways that likely would fail review before our current Supreme Court — in some cases, even the liberal justices on the Court.)
The deficiencies are real, and they go to key portions of the indictment. The most egregious example is the effort to hold Trump legally responsible for the January 6 Capitol riot. That collides directly with the Supreme Court’s landmark First Amendment decision in Brandenburg v. Ohio (1969), under which only direct incitement of “imminent lawless action” can justify punishing public speech on the theory that it led to violence. Trump’s conduct plainly falls short of the Brandenburg standard.
Why is that so important? Because Trump is charged under 18 U.S.C. § 1512 with obstructing an official proceeding; with conspiring to do so; and with conspiring to defraud the government by means of thwarting “legitimate official action.” As I will explain further below, this indictment is painfully short on proof of any of those things. The one form of obstruction of an official proceeding that has been successfully charged in January 6 cases under 18 U.S.C. § 1512 is forcible interruption of the joint session of Congress by means of physically breaching the Capitol. Had Trump been among those who did so, or had he conspired with those who did, I’d have no issue with charging him. But he didn’t, he’s not charged with conspiring with anyone who did, and the people who did have not been charged with conspiring with Trump.
But wait! say Smith’s defenders. Under the law of conspiracy, you don’t have to commit the crime — you just have to agree to do it. So long as you wanted the rioters to obstruct the joint session, and did something in public that contributed to their decision to do so, you can be criminally punished for agreeing to incitement.
If accepted, this argument would blow a hole in Brandenburg so big any prosecutor could drive a tanker truck through it. The Klansman defendant who gave an incendiary speech in Brandenburg must have wanted violence, right? Surely, a jury could be persuaded to say so. A different jury could be persuaded that Bernie Sanders wanted James Hodgkinson to shoot Republicans on a baseball field, or that the Southern Poverty Law Center wanted Floyd Corkins to shoot up the Family Research Council. It’s always easy to convince yourself that the other guy, in his heart, is a Robespierre or a Himmler. Brandenburg is supposed to draw a bright line in the law against that temptation in order to protect a vibrant enough free-speech climate that it protects even Nazis, communists, Klansmen, tiki-torch carriers, and all manner of other bottom-feeders up to the point at which they actually incite imminent, lawless action.
This is precisely why traditional criminal law has always required proof of both a criminal act and criminal intent — actus reus and mens rea, in the Latin argot of lawyers. Without intent, we risk criminalizing innocent acts; without criminal acts, we risk simply branding our enemies criminals on the theory that they mean ill. Every attempt to collapse the law of conspiracy, racketeering, or other complex hybrid offenses into elements that can be proven simply by attaching criminal intent to legal or constitutionally protected acts presents a menace to a free society.
But isn’t this a normal application of the conspiracy statute?
The first and central count in the indictment is a conspiracy to defraud the federal government under 18 U.S.C. § 371. I have previously explained why it is troublesome that this conspiracy count is based entirely upon a series of acts that were (1) legal, (2) constitutionally protected speech and political activity, and/or (3) violations only of civil law that is customarily resolved in politics or civil litigation. More fundamentally, they lack the crucial element of deception that has long been the centerpiece of a section-371 charge. “Dishonest conduct is at the heart of the crime of defrauding the government,” United States v. Barker Steel, Inc. (First Cir. 1993), which is why such cases almost invariably involve either outright deception of the government, the concealment of information necessary for the government to enforce the law, or the corruption of government officials by traditionally corrupt means such as bribery.
A major theme of Smith’s defenders has been to point to the breadth of the law of conspiracy in general (in which the overt acts in furtherance of the conspiracy need not themselves be crimes so long as they evidence and advance the illicit purpose) as well as the breadth of section 371 in particular. But taken together, these arguments produce a most curious “crime” that requires no crime and no fraud. That is dangerous. It is also a long way from how the Supreme Court, at least, has read this statute — a task that has long been hung with red-flag warnings by the courts.
As the Court cautioned in Dennis v. United States (1966), “indictments under the broad language of the general conspiracy statute must be scrutinized carefully as to each of the charged defendants because of the possibility, inherent in a criminal conspiracy charge, that its wide net may ensnare the innocent as well as the culpable.” Jackson, when sitting as a justice, wrote (in language later embraced by a majority of the Court) that “the modern crime of conspiracy is so vague that it almost defies definition” and is so “elastic, sprawling and pervasive” that, if not given judicially defined limits, it would “constitute a serious threat to fairness in our administration of justice.” Krulewitch v. United States (1949) (Jackson, J., concurring) (Alterations omitted). Because “Congress has left ambiguous” the conspiracy statute “since it was first adopted in 1867” to prosecute tax evasion, courts “must proceed with special care in interpreting” it. United States v. Minarik (Sixth Cir. 1989).
In Tanner v. United States (1987), its most recent case directly construing the scope of section 371, the Court found that the statute could be violated by lying to a power company to cause it to make false reports to federal regulators about conflict-of-interest rules in federal contracting, but that prosecutors could not simply treat fraud on the power company as fraud on “the government” just because it received regulated federal funds. Once again, the Court warned that the latter theory “has not even an arguable basis in the plain language of § 371” and that “a criminal statute, after if not before it is judicially construed, should have a discernible meaning.”
In Haas v. Henkel (1910), the Court launched the modern expansion of section 371 by declaring that “the statute is broad enough in its terms to include any conspiracy for the purpose of impairing, obstructing or defeating the lawful function of any department of Government.” Haas, however, was a financial-fraud case: The defendants conspired to get inside information on government reports about the cotton crop and have one of their confederates in the Bureau of Statistics issue a false report — all so they could make a killing on the commodities markets. Deception and secrecy was central to the scam.
The Court quickly began reining in the broadest readings of Haas. In United States v. Gradwell (1917), a unanimous Court held that the statute did not criminalize bribing voters to win congressional elections, finding it a “strained and unreasonable construction to apply to such elections this . . . originally a law for the protection of revenue and [long] confined to ‘Offenses against the Operations of the Government’ as distinguished from the processes by which men are selected to conduct such operations. . . . The section was never intended to apply to elections.” The Court drew on the fact that federal courts have no power to create common-law crimes, so “before a man can be punished as a criminal under the federal law his case must be ‘plainly and unmistakably’ within the provisions of some statute.”
Gradwell has never been overruled. As recently as Arizona v. Inter Tribal Council of Arizona, Inc. (2013), the Court noted that under Gradwell, section 371 “did not reach election fraud.” Smith will have to thread a needle in arguing that the process of counting presidential electors is meaningfully different from voting and elections.
The Court retrenched further under Hammerschmidt v. United States (1924), which involved the prosecution of leafleters who encouraged men to refuse to register for the draft, on the theory that they impaired the function of the draft. The government cited Haas, but Chief Justice William Howard Taft warned against giving “a wide meaning” to the language of Haas because the Court in Haas was “not considering whether deceit or trickery was essential to satisfy the defrauding required under the statute.” He added: “The facts in the case were such that that question was not presented. The deceit of the public, the trickery in the advance publication secured by bribery of an official, and the falsification of the reports, made the fraud and deceit so clear as the gist of the offenses actually charged that their presence was not in dispute.”
Taft, writing for a unanimous Court, framed section 371 in words that prosecutors have ever since read broadly: “To conspire to defraud the United States means primarily to cheat the Government out of property or money, but it also means to interfere with or obstruct one of its lawful governmental functions by deceit, craft or trickery, or at least by means that are dishonest.” But the Court threw out the convictions of the leafleters, noting that they may have impaired a government function, but not by deceiving the government: “A mere open defiance of the governmental purpose to enforce a law by urging persons subject to it to disobey it” does not violate the statute. That is because the statute aims at “trick, deceit, chicane or overreaching . . . by dishonest methods or schemes. One would not class robbery or burglary among frauds.”
The crucial role of defrauding the government (such as by hiding income from taxes) has been apparent in the Court’s section 371 cases ever since. In Glasser v. United States (1942), the Court found “dishonest means” where two Chicago federal prosecutors took bribes to drop charges in liquor cases. In Dennis, the Court upheld the conviction of union officials who filed false affidavits with the NLRB claiming that the union wasn’t run by communists. The so-called Klein conspiracy often used in tax cases, derived from United States v. Klein (Second Cir. 1957), is typically based upon concealing income from the tax authorities.
While the Trump indictment is full of his lies, what it is missing is any actual deception of the government, either directly or through others. It is a bedrock principle of the law of fraud — which the Court has long found to be incorporated in the federal fraud statutes — that deception requires hiding from others something they’re not in a position to know and that would have the capacity to affect their decisions by tricking them. In a criminal-scheme case, the target doesn’t need to actually act in reliance upon the lie, but the scheme needs to conceal the truth — not just tell the targets something they already know to be false or could easily determine to be false from the public record.
Deception can occur simply because a falsehood has legal significance. In Dennis, for example, the Court held that the union officials could be convicted for lying to the NLRB even though the NLRB wasn’t even empowered to determine whether the affidavits were true or to impose any consequences if they were false. But because the law required the affidavits to be true, and because the mere fact of their filing had legal significance in automatically triggering NLRB actions, the filing of a false affidavit amounted to a fraud.
Smith puts a lot of weight on the idea that Trump deceived the Trump electors, the National Archives, and Congress through the filing of bogus electoral certificates. But at no point did those certificates have any legal significance on their own; they mattered only if political actors were persuaded to throw out the proper certifications. The National Archive never had any power to take any meaningful actions. Legally, under the Electoral Count Act, the bogus certificates would change the process only if they were certified by a state executive — and they never were and never purported to be.
Moreover, the electoral certificates were done out in the open, and absolutely every decision-maker in the process knew what they were. Nevada’s Trump electors, for example, cast their ballots in a live stream on YouTube. Press coverage of the whole thing was pervasive. This was “open defiance” within the meaning of Hammerschmidt, not deception of the government. That is precisely why it did nothing to persuade the vice president, the Senate, the House, the federal or state courts, or the governor, secretary of state, or state senate or state assembly of any state. Nothing was impeded in the way that the Capitol riot impeded the vote-counting process. Even if the scheme had succeeded, it would have been because politicians who knew better went along with Trump. Nobody in government was fooled. Where’s the defrauding?
The same applies to Trump’s effort to talk Mike Pence into violating his constitutional role under the Twelfth Amendment with legal advice Pence knew was bogus, or to browbeat Brad Raffensperger with claims that Raffensperger repeatedly debunked. They all knew Trump was full of it; anybody who cared to know that did.
Yeah, but isn’t this is just about Trump trying a coup?
“Coup” isn’t anywhere in the law; the charge is fraud and obstruction. This is my big worry here: Even if you can argue your way around the legal problems with this indictment, what you’ve done is justify an open-ended, loosey-goosey theory of the law where you stitch together a series of noncriminal political acts, declare that their end is to accomplish something bad and at odds with our constitutional design or system, and — voila! — you have a federal criminal conspiracy to defraud the government. Consider what this might enable:
I could go on with different and less prominent examples (including some of the efforts to induce Trump electors to violate state bans on faithless electors), but you get the idea: There are all manner of situations in which a political act is taken on the statewide or national stage that seizes some power, alters the outcome of an election contest, and/or thwarts the proper constitutional function of some branch of government. It is frequently the case that, in the view of the opponents of that act, the legal arguments for doing so are risible, and the consequences grave.
The counter-argument is that of course, none of those cases would be brought, because none involved facts as extreme as an effort to overthrow the process by which we elect presidents. There are two glaring problems with this argument. The first is that there’s no such distinction in the statute. The rule of law is, as Justice Antonin Scalia said, a law of rules. Those rules should be made by Congress. Distinctions drawn by prosecutorial discretion rather than in rules set down by the courts are not laws, and distinctions made by courts require some sort of limiting principle. It’s always a very bad idea to try to rely on holding the line in limiting a law’s reach based on an unwritten norm that it will only be applied in cases where there is broad social disapproval of the defendant’s conduct.
Second, simply to say that these theories will be used only against Trump and not his political foes is to damn the whole argument. It may also be a bad bet. The political graveyards of Washington are full of people who did unto others what they thought would never be done back unto them.
Are you sure you’re reading the right statute?
Critics of our editorial, and of Andy’s defense of it, argue that a string of recent and nearly unanimous Supreme Court cases trimming back the mail- and wire-fraud statutes are totally irrelevant to section 371. Those cases aren’t really my argument, but the claim that they are wholly irrelevant to this indictment is misleading.
It is true that the objects of a section-371 conspiracy may be different than the objects of some other fraud statutes — broader in some ways, narrower in others. But section 371 describes the actual offense by using the term “conspire . . . to defraud . . . in any manner for any purpose.” If you compare the mail- and wire-fraud statutes (“having devised or intending to devise any scheme or artifice to defraud”), or the bank-fraud, health-care-fraud, and criminal-securities and commodities-fraud statutes (“executes, or attempts to execute, a scheme or artifice to defraud”), it is clear that the same sort of conduct is covered by each of these statutes, and one would expect courts to treat them the same way — at least in terms of the crucial question of whether a scheme to defraud requires fraud. It is likely that the current Court would start with the premise that the scope of prohibited conduct under section 371 should be read in a way that is not inconsistent with Congress’s use of parallel statutory language in numerous other federal fraud statutes. That means fraud.
Are you saying it’s okay for Trump to do what he did and go unpunished?
No, I’m not. I have said quite the contrary for two and a half years. I’m saying that we shouldn’t charge people with crimes just because we think they did something deserving of punishment where there’s no actual crime. Our society has a broad array of weapons with which to defend our elections against subversion.
Some of those weapons were actually deployed at the time: The courts and the political branches refused to give Trump the cooperation he required in order to obtain control of the United States government without the use of force, and the strong arm of the law (with the charge led by the Capitol Police) put down the riot. Some tools should have been used to impose consequences, and weren’t: The Senate should have convicted Trump, removed him from office, and barred him from seeking it again. Some are up to the voters, who punished “stop the steal” candidates in 2022 and will likely punish Trump again in the general election if he’s nominated for president in 2024.
Dissatisfaction with the punishments applied thus far should not be a justification for abusing the criminal law. One some level, this prosecution is worse than if we had simply applied lex talionis and mounted Trump’s head on a spike in front of the Capitol on the morning of January 7 as a warning to future generations. That, at least, would have dispensed with the pretense that we were doing law in a way that ought to be followed as precedent.
Prospectively, should we treat Trump’s conduct as a loophole worth closing, in order to prevent its repetition? There’s a case for doing so, but it, too, requires drafting laws with care and specificity. One step Congress already took in 2022 (albeit stuffed into a dreadful omnibus spending bill) was to clarify and fortify the Electoral Count Act. Having done so, Congress could proceed to create specific penalties in order to punish specific types of pressure on the system — that would be good, if done with the appropriate precision. In any event, arguing for additional legislation is a separate conversation; it should not be a demand that must be met in order to refrain from bringing abusive prosecutions.