Take Care is pleased to host a symposium on To End A Presidency: The Power of Impeachment—a new book by Larry Tribe & Joshua Matz.
In 1981, federal prosecutors charged Alcee Hastings, a federal judge from Florida, with conspiring to seek a $150,000 bribe. Two years later, after a federal jury found Hastings not guilty, Hastings boasted to a cheering crowd, “I am still a United States district judge . . . for life.” Hastings evidently spoke too soon, however, because a special investigation by the U.S. Court of Appeals for the Eleventh Circuit eventually revealed that Hastings won his trial only by committing perjury and tampering with evidence. With this new information, the House of Representatives voted 413–3 to impeach Hastings, charging him not only with conspiracy but also with perjury and falsifying documents. At the Senate trial that followed, Hastings insisted that the conspiracy charge violated the Fifth Amendment’s Double Jeopardy Clause. His objections were overruled. In 1989, the Senate found Hastings guilty of several “high Crimes and Misdemeanors” warranting removal from office, including one count of conspiracy and seven counts of perjury.
In an otherwise stellar book about the meaning and dangers of the Constitution’s Impeachment Clause, Laurence Tribe and Joshua Matz draw an unexpected lesson from Hastings’s story. The lesson is not what Hastings himself believed—that the Constitution’s Bill of Rights should protect impeached defendants from what could otherwise be a senatorial Star Chamber. Rather, to Tribe and Matz, the Hastings episode is evidence that “impeachment doesn’t require proof of a crime” (p. 44). Because Hastings was acquitted of conspiracy but then later impeached for it, Tribe and Matz write that “impeachment does not necessarily say anything at all about criminal liability” (pp. 44, 47).
This takeaway is surprising, not only because Hastings was impeached for criminal conduct, but also because the rest of Tribe and Matz’s book argues that people have been too cavalier, not too stingy, about what conduct should count as impeachable. The Impeachment Clause authorizes the removal of a judge or president for “high Crimes and Misdemeanors,” and Tribe and Matz spend most of To End a Presidency arguing that this vexing phrase should be interpreted narrowly. To the partisan mobs perennially raising pitchforks outside the White House’s Rose Garden, Tribe and Matz warn that not all inappropriate behavior should count as “high Crimes.”
But when it comes to people like me, who believe that “high Crimes and Misdemeanors” are restricted to actual crimes or misdemeanors, Tribe and Matz have less patience. They write that this belief “offends us as scholars and troubles us as citizens”; that it “staggers on like a vengeful zombie”; that it’s usually invoked “for partisan ends”; and that it “misunderstands the Constitution, US history, and the nature of criminal law in important ways” (pp. 44–45). Leading off with Judge Hastings, Tribe and Matz offer an eleven-page lineup of arguments that attempts to buckle the symmetry between impeachable offenses and criminal offenses (pp. 44–52). To Tribe and Matz, it is fine to impeach a president or judge like Hastings for legal conduct if the conduct was particularly bad.
These eleven pages are not only unnecessary to Tribe and Matz’s book, they’re also contradicted by it. Tribe and Matz’s own, earlier analysis of the Impeachment Clause explains why impeachment should be understood as a type of criminal prosecution—one that carries with it the Constitution’s protections for criminal defendants. These protections for impeached defendants should include the Bill of Rights and their requirement that a charged offense actually be against the law. In contrast with the eleven-page dam that currently blocks the middle of Tribe and Matz’s important scholarship, these protections would also flow naturally with their main argument: impeachment, like any criminal prosecution, is a political weapon that should be fired only after the most careful analysis.
The opening two chapters of Tribe and Matz’s book analyze the text and history of the Impeachment Clause to conclude that the phrase “high Crimes and Misdemeanors” refers to a specific type of “offense”: one “like treason and bribery,” one that “subverts core tenets of the US governmental system,” one that “require[s] proof of intentional, evil deeds,” and one that is “so plainly wrong by current standards that no reasonable official could honestly profess surprise at being impeached” (p. 42). By their own analysis, the offense they are looking for is a criminal offense—specifically, a particularly bad one.
The text of the Impeachment Clause is perhaps the strongest evidence that impeachment requires a crime. It reads: “The President, Vice President and all Civil Officers of the United States, shall be removed from Office on Impeachment for, and Conviction of, Treason, Bribery, or other high Crimes and Misdemeanors.” Tribe and Matz appropriately resist Gerald Ford’s realpolitik argument that the phrase “high Crimes and Misdemeanors” means “whatever a majority of the House of Representatives considers it to be at a given moment in history” (p. 25). They appear to agree that even if one concedes that the House and Senate are bound only by their own constitutional interpretations, the phrase “high Crimes and Misdemeanors” requires proof of criminal behavior. To understand how an ordinary member of the House would read this text, Tribe and Matz quote Rep. Brad Sherman. “There are 320 million people out there,” Sherman says. “When they hear the term ‘high crimes and misdemeanors,’ their reaction is, ‘Show me the crime’” (p. 28).
The context in which the phrase “high Crimes and Misdemeanors” appears supports Sherman’s crowdsourced interpretation. The Impeachment Clause begins by describing a type of trial: a punishment (“remov[al]”) that follows an indictment (“Impeachment”) and a finding of guilt (“Conviction”). The use of the term “Conviction,” as opposed to, say, “Judgment,” suggests that this trial is a criminal one. This suggestion is confirmed by the three things for which someone can be impeached and convicted: “Treason,” “Bribery,” or “other high Crimes and Misdemeanors.” Tribe and Matz recognize that treason and bribery are criminal offenses—ones that “drastically subvert the Constitution and involve an unforgivable abuse of the presidency” (p. 38). Invoking a Latin canon of interpretation, Tribe and Matz conclude that “high Crimes and Misdemeanors” must constitute similarly bad “offenses” (pp. 39–40). I agree. Just as treason and bribery are criminal offenses, high crimes and misdemeanors must be criminal offenses too.
For further explication of the Impeachment Clause, Tribe and Matz look to other clauses of the Constitution. These clauses further support Rep. Sherman’s interpretation that an impeachment requires proof of a crime. For example, Article I, section 3, declares that after the House of Representatives exercises its “Power of Impeachment,” the Senate has the power to “try” and “convict[]”—occasionally while the Chief Justice “preside[s].” This structure again implies that impeachment is the indictment that begins a criminal trial in which the Chief Justice is the judge and the Senate is the jury.
Tribe and Matz also discuss the two clauses in Article I, section 9, that prohibit Congress from passing a “Bill of Attainder” or an “ex post facto Law.” Tribe and Matz define the first clause as prohibiting any congressional attempt “to declare that someone’s prior acts were crimes—and to impose punishment for those ‘crimes’ without a trial or hearing” (p. 41). They define the second clause as prohibiting any congressional attempt to “ma[ke] illegal an act that was legal when committed” (p. 41). Tribe and Matz argue that these two clauses limit the types of offenses that qualify as “high Crimes and Misdemeanors” (p. 41). Specifically, they quote Charles L. Black Jr., who argued that the Impeachment Clause “[m]ust not be so interpreted as to make its operation in a given impeachment case equivalent to the operation of a bill of attainder or of an ex post facto law, or of both” (p. 41). Finding Black’s analysis “convincing,” Tribe and Matz conclude that any alleged high crime or misdemeanor must not “retroactively declare that particular presidential conduct was abusive or corrupt” (p. 41). Otherwise, the impeachment would violate the Ex Post Facto Clause or Bill of Attainder Clause.
I find this excerpt of Black’s analysis as convincing as Tribe and Matz do. And in my judgment, if Congress wants to avoid violating the Ex Post Facto Clause by “ma[king] illegal an act that was legal when committed,” Congress must avoid impeaching and convicting someone for conduct that was legal when he or she did it. Similarly, if Congress wants to avoid violating the Bill of Attainder Clause by “declar[ing] that someone’s prior acts were crimes—and to impose punishment for those ‘crimes’ without a trial or hearing”—Congress must only impeach and convict someone for what William Blackstone, in his own description of impeachment in English law, called “the already known and established law.” Satisfying both of these clauses is not difficult: all Congress must do is impeach and convict a defendant only for offenses that are already known and established as illegal—or in Tribe and Matz’s words, “so plainly wrong by current standards that no reasonable official could honestly profess surprise at being impeached” (p. 42). These offenses are located in the federal criminal code, the place that—ideally, if not in practice—provides a clear, nonvague definition of illegal conduct and makes its wrongness plain for all to see.
Pulling this all together, I think the constitutional structure of the Impeachment Clause strongly suggests that “high Crimes and Misdemeanors” are a restricted category of actual crimes and misdemeanors—a type that, as Tribe and Matz also conclude, “subverts core tenets of the US governmental system” (p. 42). “Jaywalking,” though a criminal offense whose definition is plain, is not subversive enough to be a high crime or misdemeanor. “Abuse of power,” though a subversive thing to do, is not a criminal offense whose definition is plain enough to be a high crime or misdemeanor, either. By the time I arrived at page 43 of Tribe and Matz’s book, I thought we were all on the same page about this. I thought wrong.
Over the next eleven pages of their book, from page 43 to 53, Tribe and Matz walk back their prior analysis and argue that an impeachable offense need not be a criminal offense. In Tribe and Matz’s judgment, Congress may impeach, try, and convict someone of conduct that is legal. Tribe and Matz argue that treating impeachment as a proceeding for trying criminal offenses “misunderstands the Constitution, US history, and the nature of criminal law in important ways” (pp. 44–45). I’ll address each of their arguments in turn, beginning with the arguments based in the Constitution’s structure.
Tribe and Matz’s first argument for why an impeachable offense is not necessarily a criminal offense comes from the Bill of Attainder Clause. “If impeachments were exclusively about proving that the president committed a specific crime,” Tribe and Matz write, “then the Impeachment Clause would be at war with the basis for [the Bill of Attainder Clause], since it authorizes a form of trial and punishment by the legislature” (p. 46). This argument is flawed for three reasons. First, the Impeachment Clause undoubtedly authorizes a form of trial and punishment by the legislature: its text specifically empowers the Senate to “try,” “convict[,]” and “remov[e]” civil officers.
Second, Tribe and Matz’s argument suggests that the basis for the Bill of Attainder Clause was to prohibit legislative punishments. But Tribe and Matz earlier write that a bill of attainder is not the same thing as punishment by the legislature; it is when the legislature “declare[s] that someone’s prior acts were crimes—and . . . impose[s] punishment for those ‘crimes’ without a trial or hearing” (p. 41) (emphasis added). Accepting this earlier definition for the sake of discussion, an impeachment followed by a Senate trial is not a legislative punishment imposed “without a trial.” It is precisely the opposite.
Third, Tribe and Matz’s definition of a bill of attainder is also not the only definition. Earlier, I referenced William Blackstone’s description of impeachment in English law, which he wrote two decades before the convention of 1787. Blackstone described impeachment in a chapter called, “Of Courts of a Criminal Jurisdiction,” and he stressed that impeachment proceedings were criminal proceedings. “[A]n impeachment before the lords by the commons of Great Britain, in parliament, is a prosecution of the already known and established law,” Blackstone wrote, “a presentment to the most high and supreme court of criminal jurisdiction by the most solemn grand inquest of the whole kingdom.” Blackstone specifically contrasted this sort of proceeding with bills of attainder. He defined those not as prosecutions of the “established law,” but as “acts of parliament to attaint particular persons . . . being to all intents and purposes new laws, made for the specific case, and by no means an execution of such as are already in being.”
Blackstone’s distinction between impeachment proceedings and bills of attainder contains a lot of jargon, but he elaborates. In Great Britain in 1765, Parliament was not only the highest legislative body, it was also the “supreme court in the kingdom.” This judicial court oversaw many appeals but only one kind of trial: “the trial of great and enormous offenders, whether lords or commoners, in the method of parliamentary impeachment.” Describing a practice that looks remarkably similar to the U.S. constitutional structure adopted twenty-three years later, Blackstone wrote that “articles of impeachment are a kind of bills of indictment, found by the house of commons, and afterwards tried by the lords.” Although a peer could be impeached for “any crime,” a commoner could be impeached “only for high misdemeanors.” Unlike bills of attainder, parliament imposed punishments for impeachments only after a criminal trial. During this trial, the House of Lords served as the jury of peers—“the peers of the whole nation.”
Blackstone’s description, which Tribe and Matz do not reference, offers a number of interesting insights about the Impeachment Clause: a discussion of “crimes” and “high misdemeanors”; a description of the House of Commons as a grand jury and the House of Lords as a jury of peers; and, most relevant here, an understanding of the legislature during an impeachment proceeding as a “supreme court of criminal jurisdiction” for a specific class of offenders. Just as Blackstone saw no overlap between a “prosecution of the already known and established law” and a bill of attainder “made for the specific case,” I see no overlap between the impeachment of a criminal offense and the prohibitions of the Bill of Attainder Clause. If anything, the impeachments Tribe and Matz would allow for noncriminal conduct would be “at war” with both the Bill of Attainder Clause and the Ex Post Facto Clause. Removing a president for outrageous but legal behavior would punish the president for an offense “made for the specific case”—or, to borrow Tribe and Matz’s definition of an ex post facto law, for “an act that was legal when committed” (p. 41).
Tribe and Matz also cite the Pardon Clause as evidence for why an impeachable offense need not necessarily be a criminal offense. That clause gives the president power to grant “Pardons for Offenses against the United States, except in Cases of Impeachment.” As Tribe and Matz note, under the Pardon Clause, a president can save someone from criminal punishment “but not from impeachment and removal” (p. 47). Tribe and Matz conclude from this clause that impeachment is therefore unconnected to criminal punishment.
But the text of the Pardon Clause just as easily supports the opposite conclusion. The clause considers all “Offences against the United States”—i.e., federal crimes—and makes an exception “in Cases of Impeachment.” In other words, cases of impeachment are a subset of criminal offenses to which the president’s pardon power does not extend.
Imagine if the Pardon Clause contained no exception, and the president could grant “Pardons for Offenses against the United States,” full stop. Under Tribe and Matz’s view, nothing would change, because they argue that any power regarding criminal offenses would inherently not extend to impeachment. But I think the existing exception does real work: it says that pardons for criminal offenses might help someone from being prosecuted before a judge, but they don’t help someone from being prosecuted before the Senate. Absent the exception, pardons would keep criminal offenders from being impeached.
Tribe and Matz’s final structural counterargument is perhaps the most significant. Like Judge Alcee Hastings, they point to tension between understanding impeachment as a criminal proceeding and the Bill of Rights, particularly the Double Jeopardy Clause. “The Impeachment Clause expressly contemplates that an official, once removed, can still face ‘Indictment, Trial, Judgment and Punishment, according to Law,” Tribe and Matz write (p. 47). “It would be inconsistent with the spirit of the Double Jeopardy Clause to envision criminal trial and punishment for the president after removal, but to insist that he can be removed only if Congress first finds that he did commit a crime” (p. 47). Because of this inconsistency, Tribe and Matz conclude that “[t]he better view is that impeachment does not necessarily say anything at all about criminal liability” (p. 47).
But the inconsistency Tribe and Matz identify from afar disappears on closer examination. The Double Jeopardy Clause says that no person shall be “subject for the same offence to be twice put in jeopardy of life or limb” (emphasis added). If the Impeachment Clause gave Congress the power to put someone in jeopardy of life or limb, there would indeed be inconsistency between an impeachment and subsequent punishment by the judiciary. But as Tribe and Matz write earlier in their book, the Impeachment Clause expressly withholds this power from Congress (p. 11). Instead, the clause says that “Judgement in Cases of Impeachment shall not extend further than to removal from Office, and disqualification to hold and enjoy any Office of honor, Trust or Profit under the United States.” Tribe and Matz interpret this language as intentionally restrictive: “unlike in England and France, where legislatures could impose capital punishment in cases of impeachment, Congress can do nothing more. Only the criminal justice system can impose fines, imprisonment, or a death sentence as punishment for misdeeds committed while in office” (pp. 11–12). In other words, assuming the Fifth Amendment applies to impeachments, the Double Jeopardy Clause cannot be violated when Congress impeaches someone because Congress cannot put the defendant in “jeopardy of life or limb.”
When impeachment is understood as a criminal proceeding, other clauses of the Bill of Rights fall into place like a jigsaw puzzle. The Grand Jury Clause, for example, prohibits anyone from being “held to answer for a capital, or otherwise infamous crime, unless on a presentment or indictment of a Grand Jury.” In the early years of the republic, a presentment was a formal written accusation returned by a grand jury on its own initiative—equivalent to the articles of impeachment returned by the House of Representatives. The Vicinage Clause, for its part, requires that anyone accused of a crime be tried by an impartial jury of a district “previously ascertained by law.” As Blackstone observed, the “district” for a legislative impeachment is the only court with original jurisdiction: the upper house of the legislature. The “jury” is that house’s membership.
Most importantly, the Notice of Accusation Clause requires that a criminal defendant be “informed of the nature and cause of the accusation,” and the Due Process Clause requires that any offense charged be defined in advance with enough clarity that an ordinary person could understand what conduct was prohibited. Tribe and Matz recognize the significance of these requirements: they write earlier that an impeachable offense must be “so plainly wrong by current standards that no reasonable official could honestly profess surprise at being impeached” (p. 42). But just as these clauses would prohibit a federal prosecutor from charging a judge with “abuse of power” without pointing to some statute defining that term with sufficient clarity, these clauses should also prohibit the House of Representatives from charging that same judge with the same vague offense. An impeachment trial is not some constitution-free zone; it is a place where the Constitution’s protections for defendants should be at their height.
Tribe and Matz’s second set of counterarguments maintain that it “misunderstands . . . US history” to treat impeachment proceedings as criminal proceedings (p. 44). They offer two types of historical arguments to explain why. The first, and less persuasive, is the immediate public exposition of the Impeachment Clause by Alexander Hamilton and other participants in the convention of 1787. The second, and more persuasive, is how the Impeachment Clause historically has been applied without regard to the criminality of the accusations.
Tribe and Matz are familiar with the problems of using quotes from convention delegates to explain whether “the Framers definitely meant a particular offense to be impeachable (or not)” (p. 37). Nevertheless, they’re not the first scholars to look for clarity in Alexander Hamilton’s Federalist No. 65 (p. 46). In the opening paragraphs of his essay, Hamilton wrote that that the Senate’s impeachment jurisdiction included “those offenses which proceed from the misconduct of public men, or, in other words, from the abuse or violation of some public trust.” Impeachable offenses, Hamilton continued, are “of a nature which may with peculiar propriety be denominated POLITICAL, as they relate chiefly to injuries done immediately to the society itself.” Tribe and Matz read this sort of language as evidence that the authors of the Constitution understood impeachable offenses broadly—not through a lens of criminality, but one of malfeasance and corruption. (pp. 45–46).
There are at least three reasons to be skeptical of Tribe and Matz’s conclusion that the delegates to the 1787 convention weren’t thinking about criminality when they wrote the Impeachment Clause. First, and most narrowly, Hamilton’s Federalist No. 65 was not an essay about what kinds of offenses qualify as “high Crimes and Misdemeanors.” The essay was about “the judicial character of the Senate” and whether the Senate—as opposed to, say, the Supreme Court—was the best “court for the trial of impeachments.” In answering this question yes, Hamilton explicitly cited the British model of legislative impeachment, the same one described by Blackstone, in which the lower house brought criminal charges and the upper house tried them.
Second, because Hamilton’s focus wasn’t on what qualifies as an impeachable offense, his language could easily have been limiting the range of impeachable offenses rather than expanding the range beyond criminality. Just as Tribe and Matz earlier conclude that an impeachable offense must be as subversive as “Treason” or “Bribery” (p. 38), Hamilton also could have been saying that an impeachable offense is not just any criminal offense, but a criminal offense relating to the violation of some public trust.
Third, and most significantly, there actually is quite a bit of evidence that delegates in the 1787 convention understood impeachment proceedings as criminal. The first draft of the constitution, reported by the committee of detail, declared that “The trial of all criminal offences (except in cases of impeachments) shall be in the State where they shall be committed; and shall be by Jury.” This exception—which was likely introduced because that same draft gave the Supreme Court original jurisdiction over “cases of impeachment”—makes no sense unless the authors believed that an impeachment proceeding was a “trial of . . . criminal offences.” George Mason, who Tribe and Matz quote at length in their book, supported presidential impeachment precisely because “[w]hen great crimes were committed he was for punishing the principal,” i.e., the president, “as well as [his] Coadjutors,” i.e., his co-conspirators. Dozens of participants in the debate over presidential impeachment used the word punishment, as in, “Guilt wherever found ought to be punished.” In short, some delegates may not have understood “high Crimes” as criminal offenses, but I doubt they would have been surprised if most others did.
Tribe and Matz’s second and more persuasive historical argument describes the many examples since 1789 of attempted and successful impeachments for noncriminal offenses. In addition to Judge Alcee Hastings, Tribe and Matz also briefly discuss the successful removals of Judge Robert Archbald in 1913 and Judge Halsted Ritter in 1936 (pp. 45–46). They also might as well have included the impeachments of President Andrew Johnson in 1868 and President Bill Clinton in 1998. One of the articles against President Johnson, for example, alleged that he fired his Secretary of War without legal authority (pp. 53–58).
To me, the historical precedent since 1789 is the strongest of all possible arguments for why an impeachable offense need not be criminal. Like Tribe and Matz, I accept that words that meant one thing in 1789 can evolve new meanings by 2018. It is possible that successive generations of constitutional interpretation have domesticated the term “high Crimes and Misdemeanors” to mean something very different than what the words connoted in the wild.
But it is difficult to say that the phrase “high Crimes and Misdemeanors” really does mean something different now than it did two centuries ago. As discussed above, the text and structure of the Constitution strongly suggests that an impeachment is a criminal proceeding. Even the most partisan impeachments, such as the impeachment of Justice Samuel Chase or the impeachment of President Andrew Johnson, adopted the nomenclature of criminal trials, with “charges,” “prosecutors,” a “defendant,” a “verdict,” and an attempted “conviction.” And as Rep. Brad Sherman confirms, the words “high Crimes and Misdemeanors” still connote criminal behavior to lawyers and nonlawyers alike. People still say: “Show me the crime.”
Admittedly, there are several cases in which the House has impeached people for noncriminal conduct; the House may have even assumed that the Bill of Rights and other constitutional protections don’t apply during cases of impeachment. Looking at these historical precedents from the standpoint of today makes me think of what it must have been like to look at state prosecutions before the incorporation of the Bill of Rights. For decades after the ratification of the Fourteenth Amendment in 1868, states prosecuted criminal defendants without guaranteeing them the Fourth Amendment protection from unreasonable searches, the Fifth Amendment protection from self-incrimination, or the Sixth Amendment right to counsel. Despite these decades of precedent, the Supreme Court eventually reasoned—based on the text, structure, and history of the Fourteenth Amendment—that the amendment required states to abide by these basic constitutional protections. These same considerations strongly support the application of the Bill of Rights, the Bill of Attainder Clause, and the Ex Post Facto Clause to impeachment proceedings. Path dependence alone is hardly a persuasive reason to exclude their protections.
Tribe and Matz conclude their argument against treating impeachment as a criminal proceeding with a parade of horribles (pp. 48–50). They note that the federal criminal code is “haphazard” and does not necessarily protect against the type of presidential or official behavior that could subvert US governmental institutions (p. 48). They fear that a president might spend a year in Saudi Arabia or collude with a foreign power without any congressional remedy. They write that the Impeachment Clause was meant to provide an “adaptable, flexible standard” against presidential misbehavior that a list of statutes could never capture.
These horribles, however, are not so bad. If Congress truly fears that a future president will conspire with a foreign power to undermine the Constitution, Congress can outlaw that behavior so that future presidents know exactly what not to do to remain on the right side of the law. Indeed, Congress is bound only by the Due Process Clause in prescribing the “adaptable, flexible standard” that Tribe and Matz wish were located in the Impeachment Clause itself. Tribe and Matz even offer standards for a draft statute on page 42, something like, “Any officer of the United States who intentionally subverts core tenents of the US governmental system shall be punished by removal from office.” This standard still might be too vague, but at least it would give officers more notice than currently exists of what they must avoid if they wish to keep their jobs.
Requiring such a statute might seem like a formality. But the formality is not trivial—it confirms that the same constitutional standards of due process and criminal procedure protect presidents as well as paupers; that no one, not even Congress, is above the law. Absent such a formality, Tribe and Matz do not supply a satisfactory response to someone like Judge Alcee Hastings, who argued that the Bill of Rights applied to his trial like any other. If Hastings or some future impeached defendant were to raise an objection based on the Ex Post Facto Clause, or the Fifth Amendment, or the Sixth Amendment, how should the Senate respond? That these constitutional provisions categorically do not apply to impeachment trials? That any process the House and Senate provide is all that is due?
I think the weight of Tribe and Matz’s important book points to protecting defendants in impeachment trials for the same reason we protect defendants in judicial trials. As much as a majority of Americans might want to see an undocumented immigrant get deported on the first bus, or a confessed murderer get the electric chair, or an abusive president get kicked out of office, there are fixed stars of civil and criminal procedure that the Constitution does not subject to a vote. One of these polestars is that defendants receive notice of what is against the law so that they can keep on the right side of it without being punished. Surely this principle is consistent with Tribe and Matz’s conclusion that most misbehavior is not grounds for impeachment. Surely it is as fundamental to our Constitution as any principle enshrined in