Printer Friendly

Political trials in domestic and international law.

ABSTRACT

Due process protections and other constitutional restrictions normally ensure that citizens cannot be tried and punished for political dissent, but these same restrictions interfere with criminal convictions of terrorists and others who pose a nonimmediate but real threat to public safety. To counter these threats, governments may use various subterfuges to avoid constitutional protections--often with the complicity of judges--but when they do so, they risk losing the confidence of the public, which may believe that the government targets legitimate political opponents. This Article argues that the amount of process enjoyed by defendants in criminal trials reflects a balancing of two factors: their dangerousness, on the one hand, and the risk to legitimate political competition, on the other. Political trials are those in which the defendant's opposition to the existing government or the constitutional order is the main issue. The Article discusses various ways in which governments and judges adjust process protections, so that a public threat can be countered while the risks to political competition are minimized. International trials are also discussed within this framework.</p> <pre> TABLE OF CONTENTS Introduction I. Historical, Academic, and Legal Background

of Political Trials A. Historical Background B. Literature

C. Legal Background II. Theory A. Liberal Legalism B. An Instrumental Theory of Liberal Legalism C. Departures from Liberal Legalism: Political Trials D. Summary, Evidence, Implications III. Design Principles A. Laws against Political Opposition

B. Charges, Defense, and Evidence C. The Judge D. The Jury E. The Defense Lawyer F. Publicity G. Summary IV. Variations on the Political Trial A. Trials of Enemy Combatants

B. Transitional Trials C. Pedagogical Trials D. International Criminal Trials 1. General Considerations 2. The International Criminal Court Conclusion </pre> <p>INTRODUCTION

A political trial is a trial whose disposition--that is, usually, a finding of guilt or innocence, followed by punishment or acquittal, of an individual--depends on an evaluation of the defendant's political attitudes and activities. In the typical political trial, a person is tried for engaging in political opposition or violating a law against political dissent, or for violating a broad and generally applicable law that is not usually enforced, enforced strictly, or enforced with a strict punishment, except against political opponents of the state or the government.

Political trials are uncommon in liberal democracies but not unknown. In the United States, political trials were conducted in the late eighteenth century, when Jeffersonians were convicted of violating the Sedition Act. Many people believe that trials for sedition during the Civil War, World War I, World War II, and the Cold War were political trials. And, some have argued that military trials of suspected terrorists after September 11 will be political trials.

These political trials bear a family resemblance to trials of deposed leaders of enemy states, including the first Nuremberg trial of Nazi war criminals, the Tokyo trial of Japanese war criminals, the trials of Slobodan Milosevic and other officials of the successor states of Yugoslavia, and the trials of perpetrators of the Rwandan genocide. These trials have all been political trials, although burnished with legalisms: the defendants were charged with legal violations but prosecuted because they were political enemies of the states that operate the tribunals. The legal foundation of the trials is either explicitly retroactive or based on very general international laws or principles that are selectively applied against defeated or compliant states. Although recent efforts to establish legal foundations for the trials of war criminals and dictators have resulted in the creation of the International Criminal Court, American opposition to this court ensures that in the near future such trials will continue to be political, rather than legal, institutions. (1)

Another related group of trials occurs in transitional settings, in which a democratic system succeeds an authoritarian system. These trials are not international because they take place within a single state and involve only the nationals of that state, but they are not ordinary domestic trials either because they straddle constitutional regimes, raising special problems of retroactivity. Transitional trials occurred or were seriously considered in, among other places, Poland, Czechoslovakia, Hungary, and Germany in the 1990s; Argentina, Uruguay, and Chile in the 1980s; Greece and Portugal in the 1970s; and France and other occupied countries after World War II. These trials were political because the defendants were tried for their participation in a despised government, not for any legal violations for which they could have been convicted under the old regime. (2)

Political trials of all types are heavily criticized, but some are more heavily criticized than others. Political trials in authoritarian regimes are objectionable, of course, but no more objectionable than authoritarianism itself. Political trials in transitional settings are understandable, if sometimes regrettable. Because these trials are retroactive, they violate the rule of law, and to violate the rule of law during a transition to the rule of law seems unfortunate, even paradoxical. Still, some observers defend political trials in the transitional setting as a way of teaching the public a lesson about the evils of the old regime. Political trials in the international setting are open to the charge of victor's justice, but are often seen as expedients necessary to provide the foundation for an international criminal legal system. This was a common defense of the Nuremberg trial. (3)

Political trials in liberal democracies, however, have virtually no defenders; they are reviled as a corruption of the judicial process and a betrayal of liberal principles. The standard view in the legal literature holds that (1) governments have strong incentives to limit process and attack their opponents; (2) judges are the guardians of due process rights, as well as of political rights such as free speech; (3) during times of emergency, governments exploit public fears to crack down on dissent; (4) judges should and do stand in the government's way; and (5) when they do not, it is regrettable. (4) The defect in this theory is the absence of a plausible account of the motivations of the actors. Why would judges try to restrain the government, and why would a government bent on political domination allow itself to be restrained by judges? History shows that judges often enthusiastically facilitate political prosecutions and that governments are perfectly capable of ignoring judges who do not. (5)

The thesis of this Article is that political trials in liberal democracies reflect an unsurprising balancing process at work. In an ordinary criminal trial, established procedures reflect a balance between liberty and security that is suitable for normal times. Due process protections force governments to prove that the defendant is dangerous to the public, and not simply a political opponent or someone whose conviction would be politically convenient for the government. Governments acquiesce in generous due process protections in the hope of showing that they can survive legitimate political competition: if they do not need to suppress critics, then they must have confidence in their policies.

But during emergencies and times of heightened tension, the balance changes. The public demands security, and the government can supply it only by detaining and even convicting people without strong evidence that they have committed serious crimes. (6) These people are targeted because of the harm they threaten, not because of harm they have done, so ordinary due process standards would prevent their conviction. To address these threats, governments press courts to tolerate reductions in process, and courts generally (but not always) comply. Although at the same time the reduction in process creates the danger that the government will target people who are effective critics or political opponents rather than threats to public safety, this risk to political competition is tolerated by the public when the security threat is sufficiently high.

Thus, what is distinctive about a political trial is that the liberty-security trade-off that is only implicit in the due process standards of ordinary trials is brought out into the open. The government cries "security!" The defendant cries "liberty!" The judge and the public make a political judgment about how this trade-off should be adjusted, if at all, in light of the real or purported emergency.

This argument has normative implications, which can again be contrasted to those of the standard view. The normative implication of the conventional wisdom is that judges should enforce political and process rights during emergencies more vigorously than they ordinarily do. Political trials are simply never justified, and judges should do all they can to prevent them from occurring. (7)

The normative implication of my thesis is that political trials are unavoidable and must be tolerated, although they can be better designed and managed than they often are. (8) Due process in a political trial balances the credibility of the prosecuting authorities with national security. In a nonpolitical trial, process has the main purpose of minimizing error and administrative cost. In a political trial, process has the additional purpose of maintaining the credibility of the government or (what is the same thing) of allowing the conviction of people who are public threats while preventing the conviction of people who are mere partisan political opponents. Because the credibility of the government is an issue, certain steps should be taken to enhance the government's credibility, and in the process to reduce the risk that the trial is being used for partisan purposes. These steps include involving judges and jurors from opposition parties and allowing defendants to mount political defenses.

On this view, there is no normative objection to a political trial per se. Such objections arise only to political trials that target partisan opponents rather than public threats, and to political trials that target public threats but that also erode public confidence in government to such an extent as to interfere with normal governance.

I will not focus on current controversies, including the use of military commissions to try suspected al Qaeda and Taliban members for war crimes, the use of administrative procedures to identify enemy combatants prior to their indefinite detention, the enhanced use of deportation procedures against aliens suspected of ties with al Qaeda, and the criminal trials of people suspected of terrorist activity, such as Zacarias Moussaoui. But these practices may become easier to understand in light of historical practices, and I will briefly discuss a few of them at the end of this Article.

Part I provides the historical, legal, and academic background to the political trial. Part II explains why rational governments in liberal democratic systems will sometimes conduct political trials. Part III discusses ways in which these trials can be designed and managed so as not to undermine the principle of political competition at the heart of liberal democracy. Part IV extends this discussion to related trials, including trials of enemy combatants, transitional trials, and international criminal trials.

I. HISTORICAL, ACADEMIC, AND LEGAL BACKGROUND OF POLITICAL TRIALS

A. Historical Background

In the classic domestic political trial, the defendant is tried for opposing the government or ruling class. The political trial need not be based on retroactive lawmaking, though it often is. There may be an existing law that prohibits opposition to the government; if so, the defendant may be tried under that law and convicted. Such a trial is political, even though all process rules may be observed, because the defendant's guilt or innocence depends on the defendant's political beliefs or activities. The reason that political trials are often retroactive, either in form or in fact, is that generally applicable laws forbidding political opposition are highly unpopular and frequently unworkable. A general prohibition of criticism of the government is draconian in all but the most authoritarian states. Governments depend on criticism and, even when they do not, they are usually too weak to outlaw it. In democracies, legislators shy from laws prohibiting political opposition because the legislators know that these laws can be used against them if their party loses power. Even when the majority favors such laws, opposition parties may have enough strength in the legislature or other institutions, such as the judiciary, to prevent the majority party from passing laws that have force.

For these reasons, governments that seek to harass or eliminate political opponents through the judicial process usually resort to generally applicable laws against subversion, conspiracy, disorderly conduct, incitement to violate the laws, and so forth. However, they enforce them only against people who pose a genuine threat to their power or to public security, not against mainstream political rivals or ordinary citizens blowing off steam.

The domestic political trial can be located on a spectrum that extends from the summary execution or detention at one end to the procedurally correct trial at the other end. At one extreme, the government identifies and then captures or kills its opponents without informing them of the charges, giving them a chance to defend themselves, or involving independent agents such as judges. As one moves along the spectrum, one adds procedural protections: general, public, prospective substantive rules; the right to a trial; lawyers; judges; rules of evidence; rights to cross-examine; jurors; and so forth. Military trials and deportation hearings fall at the midpoint of the spectrum; the ordinary criminal trial at the other end. As process increases, the government loses its power to disable its political opponents, but it gains something as well: the ability to claim credibly that its prosecutions serve the public interest rather than (solely) the government's interest in its own survival. Part II describes the logic of this theory in more detail.

As noted in the Introduction, not all political trials are domestic. Many are international, and others are transitional. Table 1 provides some historical examples of each type; the regular domestic trials are drawn from American history. Note that the classification of many of these cases as "political trials" is controversial, and I include them only because they recur in the literature. (9)

There have been few international political trials of leaders or major officials. (12) For an international criminal trial to occur, a state must deliver up leaders for prosecution by another state or else suffer a decisive military defeat. But no state willingly yields its leaders, and throughout most of history victorious states saw no value in conducting trials of the leaders of vanquished states. Before the modern era, leaders were executed, imprisoned, exiled, or welcomed as guest-hostages. Tamerlane displayed Bayezid I in a cage after defeating him in the battle of Ankara in 1402. The Thirty Years War ended in 1648 with a settlement, and no leaders were tried or punished. Napoleon ended his days on the island of St. Helena. The War of 1812, the Crimean War, the Franco-Prussian War, and the Russo-Japanese War all ended in political settlements. Germany was decisively defeated in World War I, but the desultory efforts to prosecute Kaiser Wilhelm collapsed when Holland refused to extradite him. Britain abandoned its efforts to prosecute Ottoman war criminals because it could not maintain control over Turkish territory. (13)

The pattern continued after World War II. Most wars ended in a political settlement or suspension of hostilities that failed to provide for trials; these wars included the conflicts between Pakistan and India and between Israel and Arab states, the Korean War, the Vietnam War, the war between Britain and Argentina over the Falkland Islands, the Soviet-Afghan War, and the first Gulf War. The U.S. prosecuted Manuel Noriega for drug crimes after it ousted him from the helm of Panama, but this was a far cry from Nuremberg. The U.S. has no interest in prosecuting Saddam Hussein, preferring to leave him to the Iraqis, albeit with substantial U.S. assistance and influence. (14) It is ironic that Nuremberg brought respectability to the political trial, as it has had virtually no value as a precedent for trying leaders of a state that has started wars, or for holding international trials of war criminals. (15) Only the Yugoslavia conflict, fifty years later, resulted in major war crimes trials before an international tribunal. (16)

Trials in transitional regimes, by contrast, have been common; Panel B lists just a few of the dozens that have occurred. The transitional or successor trial is intended to punish members of the old regime--to do substantial justice--and to persuade the public that the old regime was evil so that the new regime will be seen as a legitimate replacement.

The most common type of political trial is that which occurs within a regime. Panel C is confined to the American experience, but it is important to remember that political trials are more common in authoritarian regimes than in liberal democracies because opposition to the government is usually illegal. Britain, France, and the other established democracies also have a long history of political trials. American political trials (or trials that are arguably political even if not obviously so) have occurred most often during times of upheaval: during the founding era, the Civil War, World War I, World War II, the height of the Cold War in the late 1940s and early 1950s, and the Vietnam War.

Many political trials are clearly identifiable as such because they involve laws that target people for their political views or the peaceful expression of those views. But because "political trial" is an epithet, it has been applied to a broad range of questionable judicial practices and controversies that have little in common, and, as a result, the term is hard to define. However, the core meaning is relatively clear. (17) On this definition, a political trial occurs when the government uses the judicial process against its opponents (including foreign enemies and internal dissidents) who have not violated formal, generally enforced laws or who have violated only formal laws against political dissent. This can happen when a formal law prohibits opposition to the government or the constitutional order that the government protects, in which case the normal judicial process rules can be respected, or when charges are trumped up, and the defendant is convicted of violating laws that have been not been violated or that are very general and not enforced against people unless they are critics of the government. In the latter case, the conviction must occur with the complicity of the judge or the jury, or both; normal process protections are relaxed. (18) Political opponents, as used here, include leaders and civilians of hostile foreign states as well as domestic partisans. (19)

B. Literature

Most Americans prize their country's tradition of tolerance for political dissent, and the political trial would appear to have no place in such a tradition. Political trials are associated with various unjust or dubious events: trials of draft resisters and government critics during the Vietnam War, of harmless eccentrics or well-meaning dissenters during the Cold War, of labor organizers and peace activists during World War I. To say that a trial is political is always to condemn it. The mainstream literature on these events accepts this popular view. (20)

However, there has always been a counterpoint in the literature, stimulated by a signal event in the history of the political trial: the Nuremberg trial of major war criminals after World War II. Many commentators have been unable to allow their reservations about "normal" political trials to apply to Nuremberg, when a political trial seemed preferable to the alternatives--the release of the Nazi leaders on the ground that they violated no international law, or their summary execution on the ground that they were evil men and had lost the war. Thus, there is a tension: if the political trial at Nuremberg was desirable, then one must abandon the popular view that all political trials are objectionable. But if not all political trials are objectionable, what are the grounds for condemning the trials of dissenters in the United States?

This tension permeates Professor Judith Shklar's prominent discussion of the Nuremberg trial. (21) Shklar defends the trial on the ground that it promoted the rule of law by applying the forms of legality to a novel set of circumstances. It was, in essence, a theatrical act of legislation. The trial also served valuable political ends by helping to discredit Nazism in Germany and to awaken Germany's dormant legal traditions. (22) These arguments are in the service of Shklar's main thesis, a critique of "legalism"--the ideology under which the rule of law becomes an end in itself rather than a means to the accomplishment of liberal values--and the legalistic view that all political trials are objectionable because they violate legal norms. (23)

However, Professor Shklar flatly denies that domestic political trials can have a valuable role in a liberal constitutional order. (24) Consider her analysis of Judge Hand's opinion for the Second Circuit Court of Appeals in United States v. Dennis. (25) Rejecting the defendants' First Amendment challenge to their convictions under the Smith Act, which prohibited advocacy of destruction of the American government by force, Judge Hand argued that Justice Holmes's "clear and present danger" test of the First Amendment was too narrow; Congress ought to have the power to address future dangers that are probable but not too remote. (26) At the height of the Cold War, Judge Hand believed that international (that is, Soviet-directed) Communism posed just such a danger, and for that reason the conviction of the Dennis defendants was justified. (27)

Professor Shklar disagrees. Citing Justice Jackson's opinion on appeal, she argues that the political trial corrupts the judiciary:</p> <pre>

The judicial process, Justice Jackson observed, is not designed to deal with radical political movements. It deals with individual

offenders against law, not with the elimination of political groups. To attempt such tasks is to injure the judicial process, because the principle of legality cannot survive them. (28) </pre> <p>This is an argument by definition: because the judicial process is designed to enforce the law, it cannot be used to eliminate political enemies. But why not? Why can't it do both? Indeed, Justice Jackson concurred with the majority's affirmance of Judge Hand's decision, apparently because he believed that the challenge posed by Communism could be cabined as an international threat: judicial persecution of Communists would not necessarily result in judicial persecution of legitimate, indigenous opposition groups. (29)

The quotation above makes it appear that Professor Shklar thinks that courts can convict people only for past acts and not on the basis of future threats, but she backs off from that position, which would be in tension with her critique of the legalist mentality, and instead argues that Judge Hand was wrong about the facts. The threat posed by American Communists was remote, not probable; therefore, it was wrong to imprison them. (30)</p> <pre> If the mobilization of the judiciary were a necessary muting of law in time of war one would not complain, especially if this war, like those in the past, had a foreseeable end. However, the Cold War is not like that, nor does it require the abandonment of the principle of legality. This abandonment is a necessity conjured up by an abandonment of pragmatic liberalism and a paranoia created by an interminable, frustrating, and exhausting conflict. (31) </pre> <p>Shklar thus adopts a legalistic solution to the problem of political trials: acceptable during wars with a foreseeable end, and not otherwise. This position is inconsistent with her critique of legalistic thinking. She ought to approve or disapprove of political trials just to the extent that they promote liberal values such as tolerance. Such liberal values, she concedes, can be promoted only in a society that is secure against external and internal threats. (32) It follows that a political trial that improves security (such as at time of war, but not only then) may be justified. She disagrees with Judge Hand about the nature of the threat posed by the American Communist Party; it is only this disagreement about the facts, and not a philosophical or analytical demonstration that the political trial during times of peace is inconsistent with liberal values, that drives her critique of Dennis. (33)

Indeed, one could argue that the Dennis case was less a political trial than the Nuremberg case. As I will discuss below, the promoters of the Nuremberg trial sought to persuade the world of the evils of the Nazi system and the virtues of the allied states that opposed it. By contrast, the Dennis trial was motivated less by partisan goals than by the belief that the American Communist Party was providing assistance to a dangerous foreign enemy.

Subsequent writings of political theorists have not departed much from Shklar's conclusions. (34) Professors Charles Abel and Frank Marsh argue that political trials may generate good political outcomes--the Supreme Court's contraction of libel law and expansion of rights of criminal defendants are among their examples--but they define the political trial so broadly as to encompass virtually any case in which the court's political views may play a role in the decision, hence virtually all constitutional cases decided by the Supreme Court. (35) In doing so, they lose sight of the classic trial against political dissent, which is the source of so much discomfort and which they do not seem to defend. Professor Ron Christenson argues that a political trial brings "social contradictions" out in the open, where they can be discussed and acknowledged--but, whatever one thinks of this theory, it is hardly a reason for a government to conduct a political trial when it otherwise would have no such inclination. (36)

The political science literature is valuable but excessively general for my purposes. My concern is with how government can lower process protections when justified by security concerns without generating suspicions that it is targeting its political opponents. This is a question of legal and institutional design. (37)

C. Legal Background

Political trials need not violate domestic or international law. (38) Consider U.S. law. The U.S. Constitution expressly authorizes one type of political trial--the impeachment. Although Article 2, Section 4 says that the president and other civil officers may be impeached for "high crimes and misdemeanors," (39) and one could argue that this means that they can be impeached only for criminal violations, the general view is that impeachment may be based on political expediency. (40) The articles of impeachment of both Presidents Andrew Johnson and Bill Clinton included political as well as legal claims. (41) And many impeachments of judges have been based on political charges, such as that of undermining public confidence in the court. (42)

The U.S. Constitution does not ban political trials of other public officials, nor does it say that such political trials can occur only through impeachment proceedings. It also does not ban political trials of ordinary citizens. A number of provisions, however, limit the government's ability to conduct political trials. (43)

The First Amendment is the basic constraint. It prohibits Congress from passing laws against political opposition in general and thus from authorizing the prosecution of individuals solely on the ground of their opposition to government policy. However, in times of stress, the courts have relaxed First Amendment constraints. During the Cold War, the Supreme Court permitted a crackdown on the American Communist Party, in part because of its connection with the Soviet threat. (44) The Supreme Court subsequently repudiated this line of cases, but only after the Cold War had begun to thaw. (45) Current doctrine holds that the government can target political opponents only if they cause harm in the course of their political activities (such as robbing a bank to finance their party) or pose a threat to public order via "imminent lawless action" (such as inciting mob violence). (46)

A further constraint on political trials is the Due Process Clause, which requires the judge to grant the defendant process rights. The great importance of the Due Process Clause lies in its restrictions on simple fraudulent (or aggressive) conduct on the part of the executive. If the defendant has not violated any laws, not even generally applicable laws, the government may be tempted to cut procedural corners. At the extreme, the government fabricates evidence and bullies witnesses. None of this is unknown, but in the more common case the defendant's legal guilt is ambiguous. The government may then strengthen its case by withholding information from the defendant's lawyer, forcing the defendant to confess to a crime he did not commit or manipulating him into such a confession, playing on the fears of the jury, and so forth. The various due process and related constitutional rights--to have a lawyer, to have a jury, to call witnesses, to examine evidence, to cross-examine witnesses--limit this kind of prosecutorial abuse.

The Due Process Clause does not prohibit prosecutorial discretion: prosecutors are free to bring cases against X rather than Y, even though they committed the same crimes, or even if Y's crime is worse. (47) The prosecutor might have any number of motives: resource constraints and difficulty of proof, the value of making an example of one defendant rather than another, and so forth. The Supreme Court has held only that the prosecutor's motive cannot be invidious, and this typically means that the prosecutor cannot have racist motives (for example) or want to punish a defendant for asserting a legal right in a prior case. (48) Although bringing prosecutions for political ends would probably violate the Equal Protection Clause, courts are so deferential--they require proof of the government's motive rather than just a pattern of prosecuting political opponents who happen to violate general laws--that there is no discernible restriction on this practice. (49)

A final important constraint on the political trial in the U.S. is structural. The separation of powers requires, in most cases, some degree of cooperation among all three branches if a political trial is to succeed. Congress can block political trials by declining to enact laws that target political opponents, refusing to enact very general laws that can be selectively enforced against political opponents, and defunding or otherwise constraining executive branch officials and judges who favor political trials. (50) The executive can undermine political prosecutions authorized by Congress by refusing to pursue them with zeal, or, when authorized, by establishing regulations that constrain its own officials. (51) The judiciary can undermine political prosecutions by throwing up procedural barriers such as burdens of proof, refusing requests for secrecy, and interpreting the constitutional provisions mentioned above in an expansive fashion. (52) As each institution wants to protect itself, and as each institution (except for the presidency) is always staffed by members of both parties, political trials (at the federal level) of members of mainstream parties are rare. Instead, political trials have been conducted mainly against people whom both parties regard as political opponents, usually extremists at either end of the political spectrum. (53)

All of what has been said so far is confined to the American experience; political trials in other constitutional democracies are usually not as heavily regulated by constitutional provisions or tradition. The impeachment power was frequently used by the British parliament in its historical efforts to rein in the King; these political trials are generally thought to have enhanced liberty and the rule of law because they were used against the person who posed the chief threat to them. Impeachment was also a powerful tool against official corruption in all parts of government. This history influenced the drafters of the American Constitution, who included the impeachment power in part as a bulwark against the feared monarchical tendencies of a president and in part as a lever against corruption. (54)

Much has been written about impeachment recently, (55) and so my focus will be on the use of the normal criminal trial for political ends in the American system and on international and transitional trials.

II. THEORY

A. Liberal Legalism

Legalism is the view that courts should resolve social conflicts by applying preexisting rules to the conduct of individuals, who are given an opportunity to defend themselves. The defendant has a right to make a defense, call witnesses, cross-examine witnesses, have an impartial judge and jury, and so forth. Collectively, this package of rights is known as the right to due process. Legalism can be understood as the view that the right to judicial process is paramount and that due process should never be violated, or violated only in the most unusual conditions. (56)

Legalism is not incompatible with laws against political opposition. A law that bans criticism of the state or government is such a law; a court could enforce the law without violating the right to judicial process as long as the defendant is given the opportunity to mount a defense. The joint commitment to legalism and political tolerance is "liberal legalism," which includes the idea that courts should not permit the government to ban political speech or opposition except when it causes immediate harm--for example, incitement to riot. (57)

Political trials cannot occur in a regime of liberal legalism as long as legal institutions uphold this ideal. But real democracies only approximate liberal legalism; they sometimes enact laws against political opposition. Moreover, the more common type of political trial occurs when process is relaxed in violation of the ideal of legalism so that general laws that do not target political opponents can be used as a pretext for doing just that.

From the perspective of liberal legalism, such trials are illegitimate unless, as Professor Shklar suggests, there is something like a state of war or civil insurrection. Shklar allows liberal states to relax their own principles to defend themselves, but as previously noted, she tries to place this exception within a legalistic framework. (58) If a court determines that a state of war exists, then it might relax the rules of process or defer to executive action such as the establishment of military commissions. (59) Why shouldn't courts relax process if a substantial threat short of war exists? Shklar insists that such thinking is "utopian" and unpragmatic and that courts should not be involved in deterring mere threats. (60) But, she does not persuasively explain why relaxing process in this way violates liberal principles.

B. An Instrumental Theory of Liberal Legalism

Academic defenders of liberal legalism normally provide philosophical justifications for this system, arguing that liberal legalism--also called liberal democracy, or constitutional democracy, or the rule of law, depending on whether more emphasis is put on liberalism or legalism--promotes welfare or fairness, or respects human dignity, or maintains social peace more effectively than alternative systems. (61) Without expressing a view on these approaches, I will take a different approach that emphasizes rational choice on the part of individuals or groups with power. I approach liberal legalism not as a system of values imposed on the government, but as a reflection of the principles and attitudes that would be taken by a rational government in a democratic system, one that seeks to maximize its political support. (62)

A government, as I use the term, consists of the people who control the policy and activities of the state. In a parliamentary system, the government is typically controlled by a single party or a coalition of parties; the opposition, then, consists of the party or parties that are out of power. In a presidential system, the government is typically controlled by the president's party, but the president may be forced to share power with opposition parties if they control the legislature. In any event, I want to distinguish, very roughly, the "majority party" or "party in power" from the (mainstream) "opposition party." In liberal democracies, the various mainstream parties compete for power within a legal framework; the opposition party is not outlawed or forced to suffer legal disabilities. In some democracies, extremist parties may be outlawed or regulated; in others, they may be able to share power.

I assume that the government's main goal is to stay in power, and that a party's main goal is either to maintain power (if it has it) or obtain power (if it does not). All political actors know that they cannot maintain power unless they implement policies desired by the general public, including (but not exclusively) their political base. One such policy is security, broadly conceived. Virtually every member of the public seeks security both against internal threats such as those posed by criminal activity and against external threats such as invasion by hostile foreign countries.

It might seem that the best way to deter crime is to deny all rights to criminals, and simply seize and punish anyone who has committed a crime. Once the police have satisfied themselves that a particular person committed a crime, they would punish the individual without going through the risky and tedious business of a trial. The usual objections to this approach are that trials promote fairness and accuracy and that they prevent the government from arresting and convicting people who are vulnerable but did not commit any crime, so as to make a show of responding to the public's fear of crime without having to expend resources on a criminal investigation. But these objections are not persuasive. If law enforcement routinely convicts the wrong people, then criminals will be encouraged rather than deterred. If the government cannot keep criminal behavior at a low level, it will lose public support.

The main problem with denying procedural protections to criminal defendants is that without such protections the government can use its monopoly on force to harass, detain, or eliminate its political opponents. Authoritarian countries, in fact, frequently do this, but liberal democracies do not. Why not?

The tempting answer is "the courts": independent courts prevent governments in liberal democracies from suppressing political opposition. The problem with this answer is that many authoritarian countries do have courts, and the courts of many liberal democracies are not independent. (63) The answer also begs the question why governments bent on suppressing political opposition do not push courts out of their way; this is exactly what happens in weak democracies. The question, then, can be reframed as follows: why do governments in liberal democracies with weak courts restrain themselves from suppressing political opposition, and why do governments in liberal democracies restrain themselves from undermining strong courts so that they can suppress political opposition?

The better answer is that a government that depends on the consent of the public cannot take the risk of allowing the public to think that the government eliminates political opponents who enjoy the support of at least some of the public. Any particular criminal defendant may be an ordinary criminal, but the defendant may also be an attractive political target because of his or her leadership of, or membership in, the opposition party, or because the defendant's activities have symbolic importance for the opposition party. The public, especially the leaders and members of the opposition political party, will sometimes not know with confidence whether the government targets a particular criminal defendant because the defendant has actually committed crimes or poses a threat to security or because he or she poses a mere political (or partisan) threat to the government or party in power. If the public does not know whether the government uses its monopoly on power to target political opponents, or if it believes that the government does, it may withdraw support from the existing government and look for alternatives. The reason is that a government that uses force against opponents rather than criminals is not providing maximum security, and indeed may be pursuing policies that benefit the government itself or its circle of supporters rather than the public at large.

The problem is one of asymmetric information, and the historic solution in Western states is liberal legalism. There are two points here. First, liberalism forbids the criminalization of political opposition, and it manifests itself in formal law as freedom of speech, freedom of association, and the other basic political rights. A government that voluntarily consents to laws that protect opposition parties has taken the first step toward showing that it is a government that serves the public interest. Such a government maintains its power not by intimidating political opponents, but by creating good policy that pleases the public, which will reward the government by returning it to power.

Second, legalism ensures that the government will not circumvent the basic political rights through subterfuge. The judicial process forces the government to show that the defendant is an actual criminal or public threat, not just a political opponent. The government must show that the defendant has violated a law--that is, a rule with democratic credentials. The government must persuade an independent judge and jury that the defendant violated the law. Rules of evidence and publicity ensure that the public can evaluate the government's case. Legalism prevents the typical subterfuge by which a government targets political opponents not by eliminating them or outlawing their parties but by accusing them of committing crimes that they did not commit or of crimes that are not generally enforced.

None of this suggests that a government will always adopt liberal legalism, or that liberal legalism is necessarily self-perpetuating. If a government believes that its political opponents are likely to win the next election, the government might think that it has little to lose by prosecuting them. In the United States, this occurred only once--and while it was still an unstable quasi democracy--during the Sedition Act trials of the late eighteenth century, when Federalists used the judicial process to fend off political attacks by Republican newspapers. (64) In stable democracies, the reason that such trials do not occur more often is that the reputational cost is so high: a party that uses its control over government to prosecute its political opponents will lose public support. Indeed, the Sedition Act prosecutions backfired, made martyrs of Republican writers and editors, and contributed to the defeat of the Federalists. The experiment would not be repeated. (65)

In sum, governments grant judicial process and refrain from banning political opposition as a way of showing that their policies are in the public interest. When this kind of self-restraint becomes entrenched in a society, the state is both liberal and legalistic. A government that goes to the trouble of eliminating its political opponents does so only because it fears that these opponents are likely to attract followers, which can be the case only if a large segment of the public can be persuaded that the government's policies do not benefit it or are otherwise wrong or unjust. If this is the case, the elimination of political opponents--however attractive for narrow political reasons--is likely to give rise to the inference that the government's policies are bad and thus to result in a loss of political support.

This theory is not incompatible with the philosophical view that political rights and judicial process are necessary because of fairness or to show respect for human dignity. Indeed, the instrumental theory of liberal legalism shows why a power-maximizing government will adopt liberal policies that many people find attractive on normative grounds. It thus shows why liberal legalism is politically robust, why governments sometimes voluntary introduce liberal reforms, and why liberal legalism can be attractive to governments in societies (such as Japan) that do not have a long liberal tradition but instead emphasize the collective good. The theory also shows why liberal legalism faces limits, the subject of the next Section.

C. Departures from Liberal Legalism: Political Trials

If liberal legalism has instrumental value for governments in the way that I have described, then governments will be tempted to depart from liberal legalism under two conditions. First, the government faces an unusually dangerous threat that cannot be adequately addressed within the existing legal framework. Second, the government enjoys an unusually high level of trust among citizens, so that it need not worry too much about creating suspicions by denying process in selected cases.

As to the first point, a government knows that if it cannot protect the people, they will eventually withdraw support. So its priority is security. Threats to security can be purely internal but can also be external. The normal internal threat is everyday crime. Many governments can keep crime at tolerable levels without departing from liberal legalism. To be sure, the norms of legal liberalism are not rigid and are relaxed or tightened incrementally as circumstances warrant. When crime increases as a result of an exogenous shock--new drugs, new technologies--authorities almost always pass laws or take actions that depart incrementally from liberal legalism. The drug crisis stemming from the spread of crack cocaine led to a relaxation of liberal legalism across several dimensions: (1) vague laws that enabled prosecutors to target the most dangerous criminals, (2) broad complicity rules that enabled prosecutors to reach all members of a drug gang, and (3) antiassociation laws that enabled police to prevent congregation of gang members on the street. (66) But the more significant test for liberal legalism is terrorism or domestic insurgency, and here most liberal states have departed much farther from liberal legalism, usually for the duration of the crisis, by claiming broad powers to be exercised only against the terrorist threat. (67)

An even more important test of a government's commitment to liberal legalism is the external threat. During wartime, virtually all legal protections may be suspended and military rule imposed, depending on the extent of the threat. In the United States, the Civil War resulted in the suspension of habeas corpus, World War I in aggressive sedition laws, and World War II in martial law in Hawaii and the relocation of Americans of Japanese ancestry on the mainland. (68) Soviet-led international Communism furnished ample reason, in the minds of American authorities, for relaxing liberal legalism in the 1920s and again in the 1950s. Today, Islamic terrorism is the chief external threat to American security and the excuse for relaxing process protections.

As to the second point, when people believe that the government does not seek to eliminate its opponents, they are more likely to tolerate reductions in process: although reducing process may enhance error, this by itself will not disadvantage opponents or entrench the existing authorities. One common method that governments use to enhance trust during emergencies is to invite political opponents into the government itself. Parliamentary systems often produce war cabinets with representatives from the party that is out of power. In the United States, the most famous example is the participation of the Republicans Henry L. Stimson and Frank Knox in Franklin Roosevelt's cabinet during World War II. Members of the opposition political party with knowledge of the internal workings of the government can be expected to raise a fuss if they discover the government is using its emergency powers to persecute their colleagues and supporters. (69)

Let me put the argument in a more stylized form. Suppose that the possible defendant in a criminal trial--if not simply an ordinary criminal--may be either a "public threat" or a "political opponent" of the government. A public threat is a person, such as a terrorist, who is likely to harm the general public or the constitutional system; a political opponent is a person who poses a threat to an existing government but not to the public--the case of normal political opposition. Because the public threat has not committed any crime, the defendant cannot be convicted of a crime if given normal, that is, "high," process. Assume that conviction is possible if the defendant is given "low" process.

When the government uses high process, people who are public threats, or are suspected of being public threats, are acquitted and set free. If they engage in terrorist attacks or support subversion, the public will react by saying to the government, "If you cannot protect us, we'll find another (less scrupulous) government that will." The dilemma faced by the government is that it might, with the acquiescence of the courts, (70) rationally grant a low amount of process in response, allowing it to convict public threats. But at the same time, the government knows that if it uses low process, the public will begin to suspect that the government may be targeting political opponents. As long as the public assumes that the government is using low process to eliminate political opponents, the government has nothing to lose and much to gain from actually doing so. (71) And the public might rationally tolerate low process if the public threat is serious enough. So, the government must choose between maintaining high process, addressing the threat inadequately, and then risking public support because of security concerns, or reducing process, addressing the threat properly, but risking public support because of the perceived reduction in political competition. And although the ability to target political opponents using low process may help the government maintain its power, this advantage may not compensate for the loss of public confidence.

There may be other ways for government to finesse these difficulties. Instead of granting low process to all criminal defendants, it could offer high process to "ordinary" criminals and low process to a class of people whom the public believes more likely to pose a real threat. The American government, in fact, has done this quite frequently, granting lower process to aliens, people who openly identify themselves with extremist groups (Communists, Islamic fundamentalists), and enemy soldiers. (72) These people may also be subject to greater surveillance than ordinary citizens. The government can also grant higher-than-normal process to people who belong to mainstream opposition parties; this helps avoid the inference that the government's motives are narrowly political.

To summarize the argument thus far, one can imagine the following sequence of events. First, some emergency or apparent emergency occurs, and the public demands protection against the real or imaginary threat. Second, the government responds by reducing procedural protections. At one extreme, it might suspend habeas corpus and declare martial law. But the reduction of procedural protections could take subtler forms: the enactment of new laws, or the invocation of long dormant ones, that target seditious, disloyal, or dangerous behavior; reliance on newly broad interpretations of existing laws allowing their use against the perceived threats; refusal by judges and juries to give certain types of defendants the benefit of the doubt; relaxed evidentiary standards; restrictions on defense lawyers' access to their clients; and so forth. Third, the government now has greater freedom of action, which it can use against political opponents as well as the people who pose the new public threat. A rational, power-maximizing government will use its freedom of action to pursue both types of person. Fourth, the public realizes that the government can use its freedom of action against partisan opponents as well as public threats. The public may partially or fully withdraw support from the government because it fears political persecution of marginal or even mainstream political opponents of the government, but it also may accept this reduction in political competition as an acceptable price to pay for enhanced security. Defendants in criminal trials will exploit this public unease and claim to be political opponents (when such a claim is plausible) whether or not they in fact are. Critics of the government will call these "political trials."

At this point, it might be useful to return to the definition of the political trial. "Political trial" is usually used as an epithet, and so it is tempting to stipulate that a trial is political only if the government uses its freedom of action to target political opponents rather than genuine public threats. Partisan trials are almost always objectionable because they violate the principle of political competition at the heart of liberal democracy. One could adopt this narrow definition of political trial, but then one would need a word for criminal trials of defendants who are not merely political opponents but are in fact also public threats--threats to the entire constitutional system or to the well-being of many people--who have not committed ordinary crimes. The better approach is to use a broad definition of political trial, a definition that encompasses both the partisan trial of political opponents and the more public-spirited trial of public threats. The reason is that the latter type of trial violates the rule of law: it is political, not legal, albeit political in the broader, less objectionable sense--a matter of (possibly wise) policy rather than a vindication of the law. A further reason for using the broad definition is that most relevant trials fall between the two extremes: anarchist, Communist, or Islamic terrorists are both political opponents in the narrow sense and also threats (even if remote and long term) to the constitutional system. The final and decisive reason for using the broad definition is that the political trial (in the broad sense) creates institutional design challenges that the ordinary criminal trial does not. For example, a judge who presides over political trials might appropriately change the rules of process to interfere with partisan prosecutions while permitting convictions of public threats.

Political trials can be contrasted with show trials, such as those conducted by Nazi Germany, the Soviet Union, and many Soviet satellites. Defendants were tortured or threatened offstage, then at trial would confess to whatever crimes the government charged them with, so as to avoid being tortured or shot afterwards and to spare their families the same fate. The defendants were, in effect, unpaid actors in a propaganda film. Show trials cut the Gordian knot: governments eliminate partisan opponents as well as public threats without losing public support through the simple expedient of only pretending that they grant process protections. If the public believes the government, the government's problems are solved. (73) But the pretense cannot be maintained indefinitely even in an authoritarian state, and show trials usually stop after a few years. Show trials are not an option in an open society because they would require the collaboration of people with different political views and goals-prosecutors, judges, lawyers, juries--or else the wholesale destruction of existing institutions, which itself would alert people to the government's intentions. (74)

D. Summary, Evidence, Implications

Liberal legalism is an instrumental strategy used by governments to maximize political support in societies that have a general interest in security but tolerate normal political opposition. Liberal legalism enables the government to minimize internal threats to public security to the largest extent possible, consistent with the need to reassure the public that it will not maintain its power by harassing political opponents. When security threats increase, the government departs incrementally from liberal legalism because the public in such times is willing to tolerate a marginal increase in the harassment of (usually extreme) political opponents in return for greater security. As a result, political trials occur. These trials may target both authentic public threats and partisan political opponents. If the gain in security is large enough that the public as a whole benefits, even though the government can increasingly use political trials to enhance its power, then erosion of liberal legalism is likely to be tolerated, at least for the duration of the emergency. (75)

The history of political trials in the United States supports the thesis that in a liberal democracy, political trials are more likely to be (politically) successful when defendants are extremists than when they are mainstream opponents. The Sedition Act trials of Jeffersonian Republicans were a spectacular failure: rather than destroy the Republicans, they destroyed the Federalists. (76) The impeachment of Federalist Justice Samuel Chase in 1805, this time at the instigation of Republicans, was another failure. (77) These failures helped establish the legitimacy of political competition between mainstream parties in the United States; (78) the implicit bargain--that the judicial process will not be used against mainstream partisan opponents--has held, more or less, for two hundred years.

Subsequent political trials can be divided into two categories. First, there were trials of people who had virtually no mainstream political support: anarchists, Nazi sympathizers, and Communists. Although the trials of these people often took place in a circuslike atmosphere, the evidence suggests that the public approved of the trials and convictions and that the political standing of the government improved as a result of them. (79) Second, there were trials of people whose views were somewhere between moderate and extreme: opponents of the Civil War, World War I, and the Vietnam War. These trials were only moderately successful, as one might expect. Civil War-era military trials of dissenters may have maintained order but were highly controversial and politically damaging. (80) World War I-era espionage and sedition prosecutions were popular and may have helped the war effort, bur they also enhanced the prestige of radical politicians like Eugene Debs. (81) Vietnam War--era prosecutions like the Chicago 8 trial seem to have both discouraged violent protests and weakened support for the national government and its policies. (82)

Use of political trials only against serious public threats, and not as a routine weapon against political opponents, can reflect self-restraint by prosecuting authorities, and need not be imposed by third parties such as courts. The history of Great Britain supports this proposition, as does self-restraint in the use of the impeachment power in the United States. (83) For that matter, so does the self-restraint of executive branch officials, even when there has been short-term public support for political trials. However, in the United States, the judiciary has a great deal of prestige, and it can interfere with political trials that the government is inclined to pursue. Thus, it is useful to consider the perspective of the judge and ask how judges manage a criminal trial that is, or might be, motivated by the political goals of the government.

III. DESIGN PRINCIPLES

A political trial balances two values: the value of convicting a defendant who poses a risk to the government or the public and the value of maintaining the public's confidence that the government does not target political opponents. Ordinary judicial process, involving relatively specific laws that prohibit harmful behavior, reflects the weight of the second value. The question raised by the political trial is whether process should be relaxed (or enhanced), and in what ways. This Part illustrates the trade-offs involved, focusing on laws against political opposition; charges and evidence; the roles of the judge, jury, and defense lawyer; and control of publicity. When lawyers think of these design elements, they usually focus on the trade-off between accuracy in determining guilt (convicting the guilty, not the innocent) and administrative cost (perfect accuracy can be purchased only at infinite expense). The focus here is instead on the trade-off between accuracy in evaluating the threat (convicting dangerous people, not mere political opponents or critics) and maintaining the credibility of the government.

A. Laws against Political Opposition

The most recognizable political trial is an ordinary trial for violation of a law that prohibits political opposition to the government. Authoritarian states have often enacted such laws. The law might prohibit the formation of political parties aside from the ruling party, any kind of political activity that opposes the government, subversive activity, advocacy of policies that are contrary to government policies, and so forth. A trial in which all the forms of legality are respected would nonetheless result in the conviction of a defendant because of political views or activities.

Liberal democracies do not have laws prohibiting mainstream political opposition: tolerance of formal political opposition is the key distinction between the liberal democratic system and the authoritarian system. But the amount of tolerance is not absolute. Turkey--whose democratic credentials are solid but not perfect--bans fundamentalist Islamic parties. (84) Germany prohibits parties that oppose its constitutional system. (85) Other democracies have similar bans on extremist parties and subversive activities that are contrary to the constitutional order. (86) Thus, they make a distinction between mainstream political dissent, which is tolerated, and constitutional dissent, which is not.

In the United States, there have been five overt attempts to suppress political dissent. The Sedition Act of 1798 prohibited "false, scandalous and malicious" statements about the government, but it was interpreted broadly so that it could be used to prosecute mainstream Republican opponents of the John Adams administration. (87) Martial law during the Civil War permitted the military to try and punish people who criticized the Lincoln administration's conduct of the war. (88) The Espionage and Sedition Acts of 1917-1918 were directed against obstruction of recruitment and interference with the military, but they were broadly interpreted to prohibit criticism of American participation in World War I. (89) The Smith Act of 1940 prohibited advocacy of violent revolution against the government and was also interpreted broadly until 1957. (90) The Internal Security Act of 1950 and the Communist Control Act of 1954 "effectively criminalized the Communist party." (91) Of these cases, only the Sedition Act of 1798 was, as interpreted by judges, a clear effort to suppress dissent by a mainstream group. The Espionage and Sedition Acts and the Smith Act targeted extremists, although these extremists did include prominent people (such as Eugene Debs) who had large followings. The Civil War case is ambiguous.

Why would a government prosecute members of fringe parties or people with idiosyncratic political beliefs? By assumption, these people do not pose a threat, or much of a threat, to the political dominance of the government. One reason is that such people may be dangerous to the public. The U.S. government prosecuted Communists not because they posed an electoral threat but because they were loyal to America's enemy, the Soviet Union. (92) The U.S. government currently pursues al Qaeda sympathizers because they might provide money, shelter, or other support to actual terrorists. If the evidence of criminal behavior is not strong, the government moves against these people based on an assessment of the risks. A Communist ideologue might be a spy or know a spy. (93) An al Qaeda sympathizer--especially one with a great deal of wealth and ties to fundamentalist Islamic groups--is a risk even if he or she has not committed a crime or, given the standards of criminal law, cannot be proven to have committed a crime. Such a sympathizer may also demoralize the public by cheering on terrorist attacks, thereby creating an atmosphere of insecurity. Thus, people with extreme antigovernment beliefs are more likely to be public threats than people without such beliefs--even ordinary criminals--and for this reason governments may seek to prosecute them.

To prosecute such people without violating due process, the government would need to rely on laws that directly prohibited such activity. As discussed earlier there are, and have been, many such laws--against subversion, conspiracy to violate the law, and the like--but these laws have proven to be unpopular. (94) The problem with criminalizing membership in a particular organization like the American Communist Party is that members can easily evade the law by disbanding the proscribed organization and setting up a new one. If broader laws are used, and all seditious organizations or activities are prohibited, then mainstream organizations can too easily be swept within their net, a possibility that inevitably provokes widespread political opposition--understandably so, as the government may find itself unable to resist the temptation to enforce the laws against mainstream political opponents.

If a government cannot enact laws against political opposition, then it will find itself hampered in its efforts to prosecute public threats who have not engaged in clearly illegal or violent acts. Its best hope is to bring charges under a general law against disorderly or subversive behavior, (95) or even unrelated laws against, say, wire fraud or extortion, (96) and then to persuade the judge to acquiesce in restrictions on process. These restrictions are the hallmark of political trials in liberal democracies and the focus of the next several sections.

B. Charges, Defense, and Evidence

Legalism requires that defendants be charged with the violation of an existing law; be informed of the charges against them, so that they may prepare a defense; and be given access to evidence, so that they may prove their case.

All of these elements of normal process interfere with the prosecution of public threats. If the government does not have laws against political or ideological opposition, it will not be able to apply generally applicable rules against criminal behavior to people who have not yet caused a harm or who are not on the verge of doing so. If the government must candidly inform defendants that it has no legal case against them, they will be able to make a plausible argument that the trial is political. And the government may not be able to reveal evidence that the defendants are a public threat without compromising intelligence assets and harming security. In the Rosenberg case, for example, some of the government's evidence came from secret cable intercepts that, if revealed, would have permitted the Soviet Union to destroy valuable intelligence assets. (97) The problem for the government is that if it denies process--for example, if it tries defendants in secret to determine whether they are a public threat--then it risks losing its credibility.

The government reduces these tensions in several ways.

Selective prosecution. First, the government prosecutes public threats, when possible, for violating generally applicable laws--laws against conspiracy, disorderly conduct, subversion, trespass, incitement to riot, and so forth--that are not usually enforced against ordinary people who do the things that the actual defendant did. This approach is very much a compromise. On the one hand, the public will be suspicious of the government because selective prosecution can be used against political opponents. On the other, the harm to the government's reputation is mitigated by the facts that the generally applicable laws have received public approval and that political opponents can maintain their freedom by complying with these laws. There may be a special hardship in complying with nanny taxes, sodomy laws, and conspiracy laws to which no one else pays attention, bur it is not as bad as prosecution unconstrained by the law. In addition, general laws often have lower sentences precisely because they can be applied against so many people, so the corresponding risk to political opposition is lessened.

Consider the following examples. In the prosecution of LeRoi Jones in 1967, "it was uncertain whether Jones was on trial for a stated or implied charge--for having possessed [two revolvers], or for having been responsible, in some mysterious way, for the riots that had engulfed Newark." (98) The Chicago 8 were tried for conspiracy to incite riots, but the prosecution's real motivation was to suppress the defendants' vigorous and effective opposition to government policy and to make an example of them. The East German head of intelligence--to take an example from a transitional trial--was tried for a murder he committed sixty years earlier; his real crime was his leadership of East German intelligence. (99)

An extreme example comes from the Haymarket trial. The defendants had advocated violent revolution, but no evidence linked them to the bomb thrower (never caught) who killed the police officers. (100) The judge instructed the jury that the defendants could be convicted if they "by print or speech advised, or encouraged the commission of murder, without designating time, place or occasion at which it should be done." (101) The government sought to disrupt the anarchist movement, and the murder became the occasion for eliminating several of its leaders and frightening its members.

Governments can rarely be completely candid in political trials because they do not want to admit that the trial violates due process, even if the violation is justified for reasons of public security. Instead, governments accuse the defendant of violating a general law, while also arguing that the acute danger posed by the defendant justifies a harsh sentence. Defendants might complain that, if they do not know the real reason that the government is prosecuting them, they cannot mount an effective defense. LeRoi Jones could have argued that he could not defend himself if he thought the government was prosecuting him for gun possession, when in fact the judge and jury would have convicted him if they thought be caused the Newark riots.

Partial sharing of evidence. Second, the government may be willing to reveal classified evidence to the judge or the defense lawyer as long as it is not shared with the defendant. The defendant could legitimately object that be or she will not be able to mount a sufficient defense without having access to the information; he may not be able to provide relevant mitigating evidence to his lawyer or the judge unless he knows about the apparently inculpatory classified evidence. Further, the public might believe that the evidence is not inculpatory and that the defendant is merely a political opponent. One solution is to shift the burden to the judge (or to the defense lawyer) in the hope that the public will believe that the judge will evaluate the evidence impartially and can evaluate it correctly without hearing the response of the defendant. This solution can be effective--in the sense of maintaining the government's credibility while allowing it to convict the defendant--only if the public believes that the judge is impartial and the defendant's inability to respond to the evidence does not undermine the defendant's ability to mount a defense. But then the question becomes, why should the public trust the judge? I return to this question in Section C.

Political defenses. Third, the court, with or without the government's acquiescence, may allow the defendant to assert a political defense. Ordinary criminal defendants are rarely permitted to argue that their crimes were justified because the government is evil. There is no reason for an ordinary criminal trial to become a forum for evaluating the government's policies. But when the public suspects that the defendant is being prosecuted for political views, it may make sense to allow the defendant to mount a political defense. If his or her views are extreme--for example, the defendant is an anarchist who thinks that terrorism is justified--then the public will be more likely to support the prosecution. If his or her views are moderate, then it will be more likely that the defendant is not a public threat and that the government's motives are partisan. Thus, by allowing defendants to make political statements, governments may be able to show that a prosecution is appropriately directed toward a public threat rather than motivated by partisanship.

The judge in the Debs case permitted the defendant to make a speech defending his actions--opposition to American participation in World War I--on political grounds. (102) The judge in the Dennis case prevented the defendants from arguing that the Communist Party had an appealing political program and limited them to the question whether the party had ever advocated violent revolution. (103) Both trials were successes for the government and the judge; however, the Debs trial was less disruptive--even though Debs was a more politically popular figure.

The problem with allowing defendants to mount a political defense is that they may persuade the public to take their side; even if they do not, they may be able to undermine the public's confidence in the justice system by converting the trial into theater, preferably farce. Mockery of the judge, grandstanding, and delay become the defendant's most powerful tools. Such disruptions may provoke the judge to take harsh measures against the defendants, further showing that the judge is complicit in the government's effort to suppress political dissent.

This strategy succeeded spectacularly in the trial of Elizabeth Dilling and her codefendants--a group of Nazi sympathizers prosecuted under the Smith Act during World War II--whose lawyers objected to every act of the prosecutor and disputed every ruling of the judge. The trial dragged on for months and then ended abruptly with the death of the trial judge--from exhaustion, it was said. A retrial, more than a year later, was dismissed. (104)

To deal with these problems, judges need great skill and patience. The judges in the trial of Eugene Dennis and other members of the American Communist Party in 1949, and in the trial of the Chicago 8 in 1969, were considerably less tolerant of courtroom theatrics than the Dilling judge was. The Dennis judge frequently cut off the defendants and their lawyers. The Chicago 8 judge jailed defendants and their lawyers for contempt. In taking these steps, the judges opened themselves up to the accusation that they were depriving the defendants of a fair trial. Numerous rulings of the Chicago 8 judge were reversed on appeal. Although both judges survived the ordeal, (105) the Dennis judge was more successful; the reason was almost surely that Dennis was, at the rime of the trial, a less sympathetic figure than were the Chicago 8. In 1949, America was unified in its opposition to the Soviet Union, and therefore Dennis was unpopular except among fringe groups. In 1969, by contrast, America was divided over Vietnam, and the Chicago 8, although politically extreme, enjoyed some mainstream support for their stand against American militarism.

Judges can interrupt defendants who do not follow the rules and hold defenses out of order. But defendants can complain about these rulings, and so jurors and other witnesses might conclude that the government's motives are partisan, that the judge is complicit, and that the defendants are political opponents rather than public threats. Thus, like the other devices I have discussed, limiting the defense can have ambiguous effects. It can increase the probability of convicting a public threat by depriving the defendant of a defense, but it can also cause the jury to acquit the defendant, or the public to withdraw support from the government, because they suspect that the defendant is merely a political opponent.

C. The Judge

Judges are supposed to be impartial: they enforce the rules without bias toward the prosecution or the defense. For ordinary criminal trials, the ideal of the impartial judge is attainable because judges, whatever their hostility toward criminals, can enforce the rules of due process and ensure that people likely to have committed crimes are locked up in jail. These ordinary rules of process function mainly to ensure that innocent people are not inadvertently convicted.

Normal process no longer functions smoothly when the defendant is a public threat who has not committed any crime. If no law against political dissent or opposition exists, then the judge can ensure conviction of the public threat only by relaxing the rule of law. In this way, the judge must be complicit in the government's effort to selectively apply vague, general laws against particular defendants, or even in the trumping up of charges when no such laws can be used.

This leads to a familiar dilemma. If judges relax process when they think that a defendant is a public threat, governments may take advantage of this opportunity and bring charges against partisan opponents. Eventually, the public, including the mainstream opposition, will realize that process protections have been relaxed, and the government will lose its credibility. If people believe that the government targets its political opponents by persuading judges that they are public threats, they will--on the theory I have advanced--withdraw their support from the government. They would likely withdraw their trust from the judiciary as well.

Several design features of the judiciary mitigate this tension. I divide them into two categories: selection of judges and incentives of judges. I then discuss how judges relax process.

Selection of judges. In the United States, virtually every judge is a member of one of the two major political parties and is selected on the basis of two criteria: competence and proved partisan loyalty. In most other advanced countries, judges are members of the government bureaucracy but are trained and treated as experts, rather than partisans.

The American system functions properly as long as the parties alternate in power or government is occasionally divided, so that judicial appointments are, individually or in the aggregate, the product of compromise between the two parties. (106) As most judges are the product of the patronage system, they will refuse to allow the other party in power to convict members of their own party on trumped-up charges. To be sure, frequently a (say) Republican government will be able to bring a case before a Republican judge, but there is always the chance that the appellate panel will be dominated by Democrats, who will be sure to draw attention to partisan elements in the trial if there are any. By contrast, neither a Republican nor a Democratic judge will have much sympathy for a radical who seeks to destroy the constitutional system under which the judge exercises power. Thus, any judge is more likely to relax the rules of process in such cases.

The selection system in foreign countries is not quite as effective. Judges are trained as technocrats, and therefore they are more likely than their American counterparts to apply process rules in a mechanical fashion, regardless of the political views of the defendant. (107) This may explain why legislatures in some of these countries--especially Germany--are more likely to pass laws that prohibit extreme political dissent inconsistent with the constitutional underpinnings of the state.

Incentives of judges. Civil law systems make up for the weak selection mechanism of judges with more powerful incentives to comply with government policy. Judges are bureaucrats, and although they have some civil service protections, they are vulnerable to sanctions meted out by the government. In Japan, for example, judges who displease the government may find themselves assigned to remote rural districts. (108) To avoid such sanctions, judges may be willing to relax process rules when the defendant is a public threat. But why would such judges not also permit convictions of partisan opponents? The answer is likely that the judges fear that the mainstream opposition party of today will be the party in power tomorrow, armed with the power to exile the judge to remote districts or show their displeasure in other ways. The alternation of parties maintains an incentive that enables the prosecution of people whom both parties dislike--genuine public threats--but prevents the prosecution of people whom only one party dislikes--members of another mainstream party.

By contrast, it is harder for the American government to punish judges who fail to relax process in trials of public threats. Federal judges have independence under the Constitution that, as a practical matter, makes punishment impossible. Still, the government can reward compliant judges by elevating them. Indeed, the judges in the Rosenberg and Dennis cases were elevated to the court of appeals. (109) Judge Hand, who ruled against the government in an Espionage Act case during World War I, was subsequently denied elevation to the court of appeals that may have been his due. (110)

As a practical matter, then, the American and foreign systems may have the same effect. They either select judges or give them incentives such that process rules are likely to be maintained for trials of mainstream partisan political opponents, but not for trials of public threats or of people with fringe views. Judges might expect to be rewarded when they conduct trials that convict those widely regarded as public threats--either with popular acclaim, elevation, or similar benefits. In the first week after the Dennis case concluded, the presiding judge received fifty thousand letters from grateful citizens, many urging him to run for office. (111) And, of course, judges may share the public's fear of public threats, and, for that reason, be willing to relax process rules to convict them.

How judges relax process. How do judges relax process without destroying the rule of law as a device for maintaining political peace between mainstream groups? The primary answer has been that judges relax process mainly during wartime and other emergencies. (112) In doing so, they ensure that political competition will occur during normal times, at which time judges will enforce process protections. They also seem to relax process when the defendant is a political extremist.

As noted above, judges learned from the Sedition Act prosecutions of President John Adams' administration the dangers of trying mainstream political opponents. Since then, there have been no American political trials of mainstream political opponents. All subsequent political trials have targeted extremists--Communists, anarchists, Nazis, KKK members, members of Islamic fundamentalist groups. When the government prosecutes extremists, the public is not as likely to assume that the government is trying to obtain partisan advantage--these extremists are just too weak and unpopular to be a political threat to the mainstream parties, and they are feared because of their tendency toward violence--except, of course, when the prosecution takes advantage of public fears or misunderstandings for political gain (though this is no different from ordinary prosecutions for political gain).

This is not to say that trials of extremists have been uncontroversial. The public's political views fall along a spectrum, and trials against extremists alarm people who hold views between the extremes and the mainstream. But the point is that American judges have allowed these trials to proceed, and doing so has been politically possible and even advantageous.

Indeed, some judges have enthusiastically facilitated political prosecutions. The Federalist judges in the Sedition Act cases instructed juries in such a way that shifted much of the burden of proof onto the defendants. (113) Most judges in Espionage Act cases during World War I read the statute broadly, so that the government would not need to provide evidence that the defendant's statement caused a direct harm such as obstruction of military recruitment. (114) The judge in the Haymarket case allowed the bailiff to stack the jury with middle-class, mostly native-born citizens hostile to the anarchist, working-class, foreign-born defendants, and be gave the prosecution much more latitude than the defense. (115) The judge in the Chicago 8 case jailed many of the defendants and their lawyers for contempt, errors that were reversed on appeal. (116) The judge in the LeRoi Jones case made clear, by his questioning and demeanor, that he believed that Jones was guilty. (117)

An abiding concern for judges is that if they identify too closely with the government, they will lose their reputation for impartiality. If they are not considered impartial, they will both lose much of their public support and also the support of the government itself, which can benefit from judges who constrain it somewhat, rather than too much or too little.

For this reason, judges encourage governments to create specialized courts that are not operated by regular (Article III) judges. Military courts and commissions are examples; these tribunals do not eschew process altogether but do reduce it. (118) Military judges and lawyers are loyal to the military, but they are expected to act with some independence and can be trusted to keep secrets. (119) Allowing the government to use the military does not protect partisan opponents, but it does preserve the integrity of the judiciary (except to the extent that permitting military trials undermines it), so that it can credibly reassert its impartiality as between the mainstream parties when the emergency ends.

The main example is the Civil War, when military rule enabled authorities in the North to prosecute people who expressed political sympathy with the Confederacy. These trials sometimes led to significant political disturbances in the Union and President Lincoln, who was more politically sophisticated than the generals to whom he had to delegate military rule, was not happy with them. (120) However, to avoid the obstructionist efforts of mainstream judges like Justice Taney, (121) Lincoln had to rely on military rule and accept its attendant risks.

Similar factors have been at work in the Bush administration's creation of military commissions to try al Qaeda members and Taliban soldiers. The establishment of these commissions has apparently not hurt the government politically (122) because the American public appears to believe that emergency conditions justify a relaxation of due process. (123) Moreover, at the present time, it is not plausible to think that these commissions are being used against mainstream partisan opponents.

D. The Jury

Scholars today usually think of juries as fact-gathering institutions. Because jurors bring diverse experiences and expectations to the trial, they can combine their perspectives, enabling them to sift evidence more effectively than even a highly experienced judge. The assumption that juries are necessary for accuracy has stimulated a large literature that investigates the extent to which jurors really do make correct decisions about guilt and innocence. Although this literature has not reached firm conclusions, evidence suggests that cognitive biases and social influences may cause jurors to make worse decisions than judges do. (124)

This scholarly focus has obscured another function of juries, which is not so much to contribute to the accuracy of the fact-gathering process as to present a barrier against government oppression with judicial complicity. The jury's entrenchment in American jurisprudence is due to its pre-Revolution popularity, when jury nullification derailed prosecutions of revolutionaries and other critics of the British government. The judges, who owed their position to British authorities, took the side of the prosecution and were frustrated by the recalcitrant juries. This history implanted in the American mind the conviction that juries, not judges, are the bulwark against political prosecutions. (125)

The history suggests that the jury could be an ideal device for permitting political prosecutions against public threats but preventing those against partisan opponents. After the American Revolution, the jury could no longer regard the government as presumptively a hostile force. And if the government can make a plausible case that a particular defendant poses a public threat, the jury may be willing to convict even though the legal basis of the conviction is weak. In addition, as long as the jury is politically diverse--in the sense of having at least one or two members who belong to, or sympathize with, the opposition party--the unanimity rule ensures that partisan convictions will not be possible. To be sure, extremists who find their way onto juries may be able to block the conviction of a public threat, but judges and lawyers are careful to prevent such people from being assigned to the jury. Thus, as a general matter, juries ought to be able to hinder partisan prosecutions, but not prosecutions of public threats.

American history provides only ambiguous evidence for this hypothesis. Juries' propensity to return convictions does not appear to depend on whether the defendant belonged to an extremist or mainstream group. Thus, although it is true that Wobblies, members of the Communist Party, and foreign spies have been routinely convicted, (126) juries also returned convictions almost without exception under the Sedition Act of 1798, which targeted mainstream opponents of the government. (127) Juries did not interfere with Sedition Act prosecutions because jurors were selected by political appointees such as marshals, (128) and many of the judges instructed the juries in an aggressive fashion. (129) However, jury manipulation became a political issue that was exploited by the Republicans; (130) the fear of jury nullification may explain why there were not more trials in the Republican-dominated South. (131)

The best evidence for the hypothesis that juries could block political convictions comes from the Civil War. With the suspension of habeas corpus and military rule, political opponents could be tried without a jury or, for that matter, without an independent judge. It seems clear that the reason that Lincoln and then Congress suspended habeas corpus was that they expected juries to acquit Southern sympathizers and others who were conspiring to impede troop movements or engage in sabotage but who had not committed a provable crime. (132)

The costs and visibility of manipulating juries to ensure conviction may prove too high for many governments. The more mainstream the political opponent, the more difficult it is to manipulate the jury--because it is more likely that a member of the opponent's party will end up on the jury unless manipulation takes place. And even if manipulation is successful, it may be blatant and thus good fodder for the defense. The right to a jury may in fact have resulted in fewer partisan trials than otherwise would have occurred.

E. The Defense Lawyer

Good legal process grants criminal defendants the right to competent and independent defense counsel. Competence is a straightforward requirement; independence is more complex. Defense lawyers are officers of the court, and they are not permitted to help the defendant engage in perjury. But, even if paid by the government, they are, as a matter of custom, law, and professional self-understanding, antagonistic to the prosecution, to the point that obtaining an acquittal of a guilty client may seem a positive duty and a badge of honor. And, of course, defense lawyers attract clients by obtaining acquittals.

When the defendant is a public threat, the independence of the defense counsel may create problems. At one extreme, defense lawyers may belong to groups that share the goals of the defendant. If so, revealing classified information to defense lawyers becomes an unacceptable risk for the government. Even allowing defense lawyers to have private contact with defendants may pose unacceptable risks because defense lawyers may carry messages between defendants and their organizations. (133) Even a defense lawyer who does not share the defendant's goals may inadvertently reveal sensitive information.

At the other extreme, the defense lawyer's good-faith zeal on behalf of a client may hinder the prosecution of a public threat. Defense lawyers demand process; if the government relaxes process, the defense lawyer will draw attention to the government's efforts, causing public embarrassment, and perhaps will persuade the jury to acquit. All of this may be tolerable, but it is far from ideal.

Denying the defendant a lawyer is hardly a solution, as it encourages the public to think that the defendant is being tried for partisan reasons.

Various intermediate mechanisms have been developed. First, the weakest constraint is to require defense lawyers to abjure any connection with, or sympathy for, extremist groups. In the United States, this constraint arose in a decentralized way when bar associations decided that their members could not belong to the Communist Party. (134)

Second, governments may replace civilian defense lawyers with military lawyers. This requires either a suspension of habeas corpus, as in the Civil War, or else classifying the defendant as an enemy combatant. (135) Military procedure does not usually dispense with lawyers, but these lawyers are soldiers and therefore can be assumed to be more loyal to the state than ordinary defense lawyers.

Third, governments can give more or less assistance to lawyers; more or less access to their clients; and so forth. In some of the enemy combatant cases arising from the prosecution of the war on terror, the defendants were initially denied access to a lawyer, and then given limited access under supervision. (136)

To the extent that defense lawyers feel loyalty to the government, or have internalized norms of judicial process--such that they will attack the prosecutor but not the system--they may have limited value for the political defendant. The American Communist Party instructed its members not to use lawyers, or to limit them to the technical aspects of the case, and trained its members to use the courtroom as a platform for espousing their opposition to capitalism. The goal was not to persuade the jury to acquit the defendants--though that would be welcome--but to persuade workers that the capitalist justice system could not do justice. (137)

F. Publicity

Political trials are often, but not always, public. Publicity serves the cause of the government when it believes that the defendant is a public threat and that the public, persuaded by the prosecution that the defendant is dangerous, will forgive any bending of the rule of law and feel gratitude to the government for protecting it.

But publicity also protects the defendant, who can use the trial as a platform to denounce the government. The Chicago 8 trial is the best example of this phenomenon in recent memory, (138) but there have been many other trials in which the government dropped or settled charges after the political danger of the trial became clear, or pardoned or granted clemency to the defendants. Eugene Debs's stature rose after his conviction for sedition during World War I. He received almost one million votes for president while in jail, and subsequently his sentence was commuted by President Warren Harding. (139)

It publicity can protect the political opponent, it can also endanger the prosecution of the public threat. Many political trials have prosecuted accused spies, such as Ethel and Julius Rosenberg and Alger Hiss, and in conducting such trials, the government does not want to reveal secret information to the world. The usual practice is to keep relatively low-level trials secret when doing so is permitted by law; proceedings of military tribunals may be kept secret, for example. But, this creates the risk that the public will infer that the defendants are not public threats but political rivals of the government.

Political trials are often farcical, and, for many, this shows that they are a bad idea. At the Chicago 8 trial, two of the defendants, Abbie Hoffman and Jerry Rubin, appeared in court one day wearing judicial robes, and then took them off and wiped their feet on them. (140) At the Dilling trial, defense lawyers repetitiously lodged identical objections on behalf of each of the defendants to delay the trial and frustrate the judge. (141) But the elements of farce are the result of the specific strategy of the defendant to risk everything to make the government back down, rather than working within the legal process in the hope of gaining a regular acquittal or reduced sentence. In some cases, like the Chicago 8 trial, the farce worked in favor of the defendants, who managed to convince large segments of the public that the justice system or the government was unjust. (142) But, in other cases, like the Dennis case, the farce left the public unmoved, the strategy failed, and the movement petered out. (143) AS messy as these trials were, it is far from clear that they were failures, from the perspective of the government.

G. Summary

In political trials, the government and the defendant battle for public opinion. The government seeks to persuade the public that the defendant is a public threat rather than a mere political opponent. If this is true, the government's best strategy is to publicize the trial and give the defendant ample time to make a political defense (for example, "Justice is not possible without a proletarian/Islamic revolution.") Any defects in process may be forgiven by the judge, jury, and public, who are glad to see the threat extinguished.

Sophisticated defendants in cases like these, however, will downplay their revolutionary ardor and draw attention to the defects in process. If, for example, the government relies on a general law, the defendant will point out that the government could use the law against mainstream political opponents as well as radicals or extremists. If the judge or some jurors belong to the out-of-power party, they may be persuaded to acquit to protect themselves and their party. Moderates in the public may become suspicious that the government is establishing a precedent with the extremists before turning its attention to mainstream political opponents. If so, they may put political pressure on the government to drop the charges.

The government, then, has the task of persuading the judge, jury, and public that the relaxation of process (or the enforcement of explicit laws against political dissent) is limited to cases involving authentic public threats. This can be done by limiting the law or enforcement action to aliens, people with connections to enemy foreign countries or movements, people with connections to groups that engage in violence, and people with highly sensitive positions (soldiers, spies, government employees), and by granting as much process as is compatible with the requirements of secrecy. As I have argued, allowing political defenses may also be useful.

Complicating matters for the government, the logistical difficulties of running an ordinary criminal trial become political problems in a political trial. The everyday compromises of a normal criminal trial--the judge and prosecutor's influence over the composition of the jury; reliance on secret evidence or testimony when victims or informers face retaliation; the need to cut off defendants, witnesses, and lawyers in the interest of time; the exclusion of morally relevant but legally irrelevant arguments; the blunders of subpar lawyers and judges--take on heightened significance at a political trial, in which a skeptical public may misinterpret these normal compromises as a special effort by the government to deprive the defendant of what the public thinks are the standard protections. When the public is suspicious enough about the government's motives, the government might be well advised to give the defendant heightened process and focus on convicting the defendant for violations of normal laws (if any) rather than making vague "dangerousness" the focus of the trial.

The amount of process granted in a trial is a function of all the factors that I have discussed--the independence of the judge, the availability of defense lawyers, the extent of the right to make a case and cross-examine, and so forth. It is not clear whether each element is essential or whether granting more of one element can compensate for less of another; this is perhaps true in some cases but not others. Each factor is context specific. It may be necessary to deny an articulate and charismatic defendant the amount of time available for testimony that is given to a regular defendant. Publicity and even lawyers may have to be dispensed with when the evidence consists of secret materials; in this case, a judge from an out-of-power party may nonetheless serve to constrain the government appropriately. Again, an independent judiciary may be unavailable when it is tainted--a common problem for trials in transitional regimes--and a dependent judge will be used; but then other protections (defense lawyers, for example) may compensate for the subservient judge.

IV. VARIATIONS ON THE POLITICAL TRIAL

This Part examines trials that are variations on the type of political trial that I have discussed so far. Trials of enemy combatants are an even purer form of the politically colored domestic criminal trial: their main purpose is identifying people who are threats but who have not committed crimes. Transitional trials are efforts to eliminate the influence of politically dangerous people left over from an earlier regime. Pedagogical trials are trials whose purpose is to educate the public. And international trials are politically colored because international law is very weak and, in its practical application, is heavily influenced by the geopolitical agendas of powerful states.

A. Trials of Enemy Combatants

Nations have traditionally detained captured soldiers for the duration of hostilities and tried and punished those enemy soldiers who committed war crimes. These trials have elements of both political trials and ordinary trials. On the one hand, enemy soldiers tried for war crimes are usually granted the same process that one's own soldiers are granted, and the laws of war are generally accepted by nations. (144) On the other hand, prosecutions of soldiers reflect military and political goals. Immunity is granted to enemy soldiers with intelligence value, technical expertise, managerial skills, and other characteristics that are important to some new war effort or to the return to peace. For example, the U.S. granted immunity to Japanese researchers who had conducted experiments on prisoners of war (POWs), a price deemed worth paying for the information they had to offer. (145) Moreover, the armies rarely prosecuted their own soldiers for war crimes.

The laws of war extend their protection only to soldiers who meet certain qualifications. They must wear uniforms, carry their weapons openly, belong to a regularly constituted military unit, and meet related criteria. (146) Soldiers who violate these rules are considered spies, guerillas, or, in the current phrase, "unlawful combatants," and deprived of the rights enjoyed by POWs. Shortly after its entry into World War II, the American military set up a military commission to try German soldiers--including one American citizen--who had sneaked onto American territory to engage in sabotage. (147) This type of military commission need not comply with the requirements of the Geneva Convention for example, it may grant less process than the tribunals used for POWs who commit crimes. After the 9/11 terrorist attacks, the U.S. government has claimed the right to classify Americans and foreigners as unlawful combatants and to detain them until the end of hostilities. (148) The government also has established military commissions that have the authority to try enemy combatants--members of al Qaeda and affiliates--for war crimes. (149)

The U.S. government's post-9/11 legal strategy has received a great deal of criticism. Many critics argue that al Qaeda terrorists should be treated as criminal suspects and, when captured, given the same process as regular criminal defendants. (150) Although the courts have not gone this far, they have demanded that the government grant unlawful combatants more process than the government was initially willing to give them. (151) The debate, within the courts and outside them, has weighed the unfairness of denying process against the demands of military exigency. (152)

The theory that I have advanced sheds light on this debate, albeit from a different angle. One question--why the post-9/11 strategy reflects a military rather than law enforcement approach, when similar actions during earlier wars were undertaken mainly by law enforcement--can be easily answered.

First, the public believes that the threat posed by al Qaeda is enormous; for that reason, it will tolerate a reduction of political competition to enable the government to counter the threat. By contrast, the public has had more mixed views about the threats posed by domestic Communists, anarchists, and members of other fringe movements. World War I and the Vietnam War had less public support than the war on terror because America's enemies did not pose as palpable a threat.

Second, members of al Qaeda are, for the most part, ethnically and religiously distinct from Americans. As long as the government focuses its investigative efforts on Arabs, Muslims, and aliens from Arab and Muslim countries, and the American public believes that the government's attention is confined to these types of people, the vast majority of the public itself will feel unthreatened by politically motivated prosecutions. By contrast, World War I-era prosecutions often targeted ordinary Americans, especially those affiliated with the labor movement, which enjoyed widespread support.

In these two ways, the closest precedent for the effort against al Qaeda is the internment of Japanese Americans during World War II. Both cases involve high and palpable risks to American civilians on American territory; and both cases involve an unassimilated, ethnically distinct, and politically weak group, which could be targeted for special measures without creating the risk that such measures would be used against mainstream Americans. (153)

Third, the war against terror is a bipartisan effort. The government cannot credibly accuse the opposition party--the Democrats--of being supporters of al Qaeda, so the government cannot plausibly classify Democrats as enemy combatants. The current situation bears no resemblance to the Sedition Act prosecutions, which were brought by Federalists who believed that the Republicans had allied themselves with France, then America's enemy in an undeclared war. (154) Nor does it resemble the Cold War, when Republicans could castigate the Democrats for being soft on Communism but never tried to bring criminal charges against mainstream Democratic opponents.

All of this suggests that reduced process protections during the war against terror do not provide the Bush administration with real opportunities for targeting political opponents; thus, few people regard the trials and detentions as motivated by partisan political objectives. To be sure, these trials and detentions may be mistaken, unfair, opportunistic, and political in the sense of being designed to show the public that the government is doing something when it really is not doing much. But this does not distinguish them from ordinary law enforcement practices. As long as al Qaeda and similar groups pose a genuine public threat, and as long as these groups are unable to acquire significant support from mainstream Americans, the reduction in process protections tolerated by the courts so far can be explained using the instrumental theory of liberal legalism.

B. Transitional Trials

Transitional trials occur when a newly democratic state tries officials of the old regime for acts that were lawful at the time they were performed. The transitional government faces a delicate problem. It must, on the one hand, satisfy demands for substantive justice against the old regime and eliminate the influence of remnants that continue to hold important positions in the military, bureaucracy, and economy. On the other hand, the transitional government must avoid the accusation that it is using transitional trials to do what the old regime did: eliminate political opponents through a perversion of the judicial process. Transitional governments have balanced these considerations in several ways.

First, the government may rely on the exceptional nature of the transition and the crimes of the members of the old regime. Their chief crime (in a moral sense) was the rejection of liberal legality; they are being punished, in part, for that offense. The public watching the trial need not infer that ordinary individuals will be similarly denied due process once the transition has been completed and the regime has entered the phase of normal politics. Retroactivity, in a paradoxical way, reduces concerns of political motivation by indicating that younger people untainted by the past have nothing to fear from the courts.

Second, the government may grant the defendant as much process as is compatible with the retroactive nature of the trial. This is not trivial and may include the assignment of a lawyer, the chance to testify and cross-examine, publicity, and so forth. A government bent on eliminating its political opponents would not give defendants these rights. (155) The lesson is that the government may, and should, exercise restraint when pursuing its political opponents. This may seem like a modest lesson, but it is an important one for states emerging from decades of totalitarianism.

Third, governments may rely heavily on legal fictions to conceal the political motivation of the prosecutions. I will say more about this tactic in Section C.

Finally, some governments may use administrative proceedings rather than trials. In Czechoslovakia, for example, administrative proceedings were used to identify and "lustrate" former officials and collaborators, who were deprived of positions in the government but could otherwise resume a normal life. (156) These proceedings reduced both process and punishment. The government thus could not take advantage of the limited process in these proceedings to eliminate its political opponents but could only embarrass them; the embarrassment was, in any case, connected to the source of the transitional government's legitimacy--the decisive public rejection of the old regime.

C. Pedagogical Trials

Much discussion of political trials concerns a second-order issue: the educational message that they send to the public. Uncomfortable with the notion that a trial could ever be justified by public demands for retributive justice or the dangers posed by a public threat, scholars have instead focused on how political trials may teach the public liberal values. Four examples will illustrate this argument.

First, Professor Jon Elster and other scholars argue that transitional trials may be justified as a way to educate the public about the rule of law. (157) Observers of the transitions feared that liberal democracy would not take hold in societies in which the rule of law had been repudiated for decades. By conducting fair trials of members of the old regime, transitional trials would dramatically show that the rule of law extends even to enemies of the new government.

Second, Professor Shklar argues that the Nuremberg trial was justified as a device for creating new norms of international illegality. (158) Prior to that trial, international law did not prohibit genocide and other crimes against humanity; the trial helped established that this behavior was criminal. (159) In doing so, the Nuremberg trial helped extend international law beyond its traditional application to states and into the realm of human rights.

Third, several scholars have pointed out that the Nuremberg trial placed "on the record" thousands of archival documents that showed how the Nazis engineered the Holocaust. (160) Truth is a liberal virtue, and documenting atrocities is thus a liberal duty. This argument was used by those who supported the establishment of the tribunals charged with trying perpetrators of international crimes in the former Yugoslavia and Rwanda. (161)

Fourth, many domestic political trials have been intended to teach the public about emerging threats. The trials of Jeffersonian Republicans were intended to show that the defendants were in league with America's enemy, France. (162) The trials of Nazi sympathizers and Communists were intended to show Americans that the Nazis, then the Communists, posed a threat to American security. (163)

The educational purpose of political trials is different from, but not necessarily inconsistent with, their main purpose, which is to counter public threats or to do substantive justice. However, in practice, the main purpose of the trials takes precedence and obscures their educational impact. As several critics have noted, the transitional trials might be interpreted not as a vindication or even illustration of the rule of law, but as the opposite: as demonstrations that judicial process and the rule of law can be disregarded when the defendants are political opponents. (164) After all, these trials resulted in the conviction of people for activities that did not violate any law at the rime that they occurred. This apparent dilemma has also bothered critics of international war crimes trials, who fear that the educational purposes of these trials may corrupt the judicial process. (165)

The criticism is too simple, but it does have an element of truth. Transitional trials are meant to show, in part, that the old regimes' great sin was their disregard of liberal legalism; not to punish members of the old regime would also send the wrong message, as it would indicate that tyrants are not punished for their tyranny. The lesson of the transitional trials is that, if political opposition may be tolerable within the constraints of a liberal constitutional order, the rejection of liberal democracy is never tolerable.

This message has not, however, been delivered in the most candid way. Rather than forthrightly announcing that certain political beliefs and systems would not be tolerated, the governments have surrounded the judicial proceedings with legal fictions. The fictions implied that the defendants were being tried and punished for committing crimes recognized as such during the old regime.

In the trials of East German border guards, for example, many jurists claimed, and some judges held, that the conviction could be based on prior law--including international treaties such as the International Covenant on Civil and Political Rights (ICCPR), international jus cogens norms against human rights violations, West German law (on the theory that East German law was never valid because East Germany never had sovereignty), natural law, and East German law that on its face could be interpreted as prohibiting the border killings. (166) These were all subterfuges: the ICCPR and jus cogens norms were never thought to apply to domestic prosecutions like the border guard cases, and East German law, whose preunification domestic validity was acknowledged by West Germany in the unification treaty, authorized the border guards to kill people who were trying to escape. (167) Academics who recognize the force of retroactive justice in the transitional setting nonetheless find the lack of candor intolerable. Let the governments prosecute the remnants of the old regime for the misery it inflicted, says Professor Elster, but make them admit what they are doing. (168) Hannah Arendt anticipated this reaction with nearly identical comments about the Eichmann trial: don't pretend that Eichmann broke the law; admit that the motivation for the trial was revenge. (169)

These criticisms reflect the worry that the trials will not educate people if they send muddy or mixed messages. But one can see why it is unwise to use political trials for educational purposes by asking why the German government and courts did not follow Elster's recommendation. Surely, the government's main concern was obtaining convictions--both to appease longstanding outrage in West Germany at the actions of the border guards and to show East Germans that such behavior would no longer be tolerated. The judges appeared to have seen the force of these goals, but their jobs were to enforce the law, not implement official policy. It is hard to imagine them candidly announcing that they would imprison the border guards even though they broke no laws. The educational goal of the trials ran up against the bureaucratic realities of a modern liberal state.

The same problem exists for the goal of creating a historical record. One purpose of the Nuremberg trial was to provide a record of the Holocaust. (170) The extensive German documentation of the concentration and death camps could be entered into the record of the trial; films of the atrocities were also shown. (171) The French referred to their trial of Klaus Barbie as a "pedagogical trial" because its purpose was to teach the public about French complicity in the Holocaust. (172)

But as the Nuremberg trial shows, the efforts to provide a record of history conflicted time and again with the more important goal of proving the charges. The Nuremberg and Tokyo prosecutors sought representatives of important segments of the population to stand as defendants. The German and Japanese publics needed to learn that the rot had spread throughout the military, the bureaucracy, and the industrial elite. A trial of four army generals would not have been as effective as a trial of one general, one admiral, one diplomat, and one industrialist. At Nuremberg, Justice Jackson sought to implicate German industry; but, when a natural choice--Adolf Krupps--turned out to be too ill to stand trial, Justice Jackson's demand that his son Gustav be substituted outraged the judges, and they refused. (173) In Japan, the prosecutors made sure to charge not only representatives of each component of the Japanese government, but also to find defendants to represent each time period--the attack on Manchuria as well as the attack on Pearl Harbor--although government officials during the various periods were not the same. (174) These choices all contributed to a satisfying narrative arc, but they also made the government vulnerable to the argument that the trial was intended for propaganda, not to establish the truth.

Nuremberg illustrates other compromises. Because the trial had to be conducted with dispatch, many atrocities were not discussed: the records had not yet been discovered. (175) Because the prosecutors relied on documentation rather than witnesses, undocumented but amply witnessed events were also downplayed. Most important, each victor state had different preferences about which events to emphasize and which to suppress. (176) For the Russians, the goal was to show the world the extent of the sacrifice of the Russian people, and to conceal Russian atrocities and Russian aggression against Poland and other innocent states. This was not the aim of the Americans, who preferred to emphasize the guilt of German leaders for waging aggressive war. A similar problem arose over the charge that the German invasion of Norway was part of its conspiracy to engage in aggressive war; the Nazis responded that they had merely preempted a British and French invasion of Norway--which was true, and embarrassing to the British, who refused to release secret documents that would have confirmed the defendants' claim. (177) Events that placed the victors in a bad light were suppressed as much as possible. Indeed, even though everyone suspected the Russians, the indictment charged that the Germans had massacred the Polish officers at Katyn Forest. (178) As is always the case in trials, the establishment of the truth, in all its nuance and complexity, was subordinated to the immediate goals of proving or refuting the charges. (179)

Even when the truth comes out, it does not necessarily follow that the right lessons will be learned. The Allies initially hoped that the Nuremberg trial would teach the Germans and their own populations that the German leaders were evil and that there was no possible justification for their behavior. The dual goals were the education of the Germans about their own leadership, so that they could reconcile themselves to a new order in which Germany would either be ruled by foreign powers or else have greatly reduced international status, and vindication of the political leadership of the Allied countries. (180)

Throughout the trial, the prosecutors and judges feared that the defendants would argue that (1) the war crimes committed by the Nazis were matched by war crimes committed by the Allies; and (2) Germany's behavior was justified by the Versailles Treaty and was abetted during the 1930s by the British and the French, who ratified many of the formal violations of the Versailles Treaty by the Germans, and by Russia, which cooperated with Germany in many ways. (181) Although not many citizens of Allied nations would have been receptive to these arguments, German citizens might have been; thus, the trial, like the Versailles Treaty itself, could have become a rallying point for unreconstructed Nazis and nationalists in Germany.

Although the defendants did make these arguments, they did not have the expected effect. The Germans at first ignored the trial, regarding it as irrelevant or as an exercise in Allied propaganda. But then something surprising happened: the German people began to feel that they were themselves on trial. The trial made clear the vast participation of ordinary citizens in the Nazi extermination machine, even though "[t]hat had never been Jackson's intention. He, like many others, had hoped to prune out Nazism and induce healthy growth in remaining Germany. The idea of German guilt had not appeared in the indictment. It had emerged during the trial." (182)

But a guilty nation was not a nation that could rejoin the world community as a liberal democracy. If there was something wrong with Germans, how could they be given political responsibility? The notion of German guilt grew particularly difficult for the Americans as it became clear, partway through the trial and to the premature delight of the defendants, that the Germans would be America's allies in the gathering Cold War against the Soviet Union. (183) Polling data suggest that the Germans initially thought that the trial was fair, but changed their minds after a few years; by the 1950s, the dominant view was that the trials were unfair and the convictions were victor's justice. (184)

The Tokyo war crimes tribunal had the opposite but equally unwelcome effect. The Japanese did not consider themselves on trial, or guilty, though they were horrified when they learned about the atrocities committed by their soldiers. But as bad as their soldiers were, they did not seem any worse than the Americans, who had killed hundreds of thousands of civilians through firebombing and atomic bombing. The trial gave the defendants an opportunity to defend Japan's militarism, an opportunity seized by Marshal Tojo with great success. (185) Tojo argued that Japan had acted in self-defense against colonial aggression. "Only victor nations sat on the Bench, many of them colonial powers with far longer records of imperialism than Japan--and they allowed the colonies they were intent on regaining no place among the judges." (186) The trial suggested the moral equivalence of the victors and vanquished with respect to aggression, but the victors had a monopoly on hypocrisy.

A great problem for the Tokyo trials was the decision by the U.S., made prior to the initiation of the trial, to allow the Japanese emperor to retain his throne. As a result, he could not be tried, even though he was the one person who had formal--and probably personal--responsibility for all aspects of Japan's aggression. (187) Trying to tie together disparate defendants in a conspiracy when the one person they had in common was absent was a nearly impossible task. (188) The U.S. had its reasons for immunizing the emperor: American officials believed that a cooperative emperor was their best hope for governing Japan. In Tokyo, as at Nuremberg, the need to cooperate with defeated officials and leaders warred with the desire to do justice, and the result was a message with little educational value, at least little that would directly serve the interests of the U.S. The trial was widely regarded as a political failure. (189) Many Japanese citizens did not see the trial as a vindication of the rule of law but as victor's justice. The trial contributed to a resurgence of nationalism during the postwar years. (190)

However, nationalism did not reassert itself in all quarters, and one historian argues that the trial--because it was a caricature of justice--contributed to postwar Japanese pacifism. "[T]he crimes revealed by the trial, compounded by the perception that this was a world gone mad with violence and that such crimes against peace and humanity were not unique to Japan, reinforced the deep aversion to militarization and war that had come with defeat." (191) If this was the lesson that the Japanese drew from the trial, it certainly was not the lesson intended by the Americans, who sought to remilitarize Japan as an ally against the Soviet Union and who not only abandoned further war crimes prosecutions after the Tokyo trial was over, but (as in Germany) cooperated with and supported war criminals, including a future prime minister who the U.S. thought could be useful in the Cold War. (192) Political trials may have an educational function, but this function is often out of the hands of those who conduct them.

The educational performance of political trials thus has a sorry history. Whether or not they result in convictions, they never teach exactly the kind of lesson that the government has in mind. The question, then, is why a government might be willing to undergo the risk of a trial to educate the public: why not send educative messages through propaganda, official statements, and other routine channels? Or, if the government does want to discover and publicize the truth about historical events, why not use truth commissions? (193)

One answer is that the government believes that it can win a public political contest with its enemies. In such cases, the trial becomes a forum, like a legislative house, in which a policy debate is engaged. The value of the message is heightened if the defendant has a fair chance to refute it but fails to win over the public--especially when life or freedom are at stake, making the defendant a natural object of sympathy. To be sure, the government cannot present its message in as unadulterated a form as a propaganda message, but the message of the trial, because of the presence of the defendant, is more credible.

This argument is similar to the instrumental theory that political trials occur because governments need to maintain their credibility-they serve the public interest and do not merely eliminate partisan opponents--when attempting to counter public threats. In the current argument, the defendant is not necessarily a real threat, but the defendant's message is. The government gives the defendant the chance to publicize political views because the government believes that the public will reject them.

The latter argument helps to explain why international political trials are almost always promoted and conducted by liberal democracies--which might seem paradoxical, given the rarity of domestic political trials in liberal democracies and their frequency in authoritarian states. The explanation is that the liberal democratic system has greater international appeal than any particular authoritarian system, because authoritarian systems always elevate the interests of some national or religious or ideological group; democratic systems are, in principle, universalistic. Every international political trial has been intended as a story about how an authoritarian government led its people astray, and, by implication, about how liberal democracies are superior. The creators and managers of trials are not always correct; sometimes, a trial does not convince anyone of anything. But the greater appeal of trials to democratic governments is easily understood. (194)

D. International Criminal Trials

1. General Considerations. The international criminal trial (195) is a special type of political trial. The "defendant" is now a soldier or former leader of a (usually) defeated state; the prosecuting "government" is now a foreign power, or a coalition of foreign powers, which vanquished the other state. The victor's main goal is to eliminate the hostile government of the defeated state--its leaders and its supporters. The purpose of the trial is to define the category of individual, government, or state behavior that will not be tolerated by the victors, so that the rest of the world will understand what kind of behavior will provoke an international military response and what kind of behavior will not.

The Nuremberg trial, for example, was an ambitious effort to define new rules of international conduct that would make a repetition of the two world wars impossible. The world wars were blamed in part on great power rivalries and the understanding that states were permitted to go to war for political objectives. Thus, one of the Nuremberg charges was the crime of aggressive war. (196) And the conduct of World War II, with the German government's mass slaughter of its own citizens as well as foreign citizens, led to the invention of "crimes against humanity." (197) These two crimes, and the notion that individuals (as opposed to states) could be charged for violating them, were all innovations. (198) Ideally, if individuals in the future knew that they could be punished for the crimes of aggressive war, crimes against humanity, and also ordinary war crimes (also a subject of the Nuremberg trial, but not an innovation), they would be deterred from starting wars or conducting them too brutally.

The Nuremberg trial had many of the elements of domestic political trial. The victorious governments, led by the United States, sought to accomplish two things. First, they wanted to eliminate the major Nazi figures. This could have been accomplished with summary execution--an option that was widely discussed and seriously considered. But there was a second goal as well. This was to show the world--including the citizens of Germany, the citizens of the victorious nations, and the governments of other countries--that there would be a new international order, one in which governments would not be permitted to engage in aggressive war (that is, use military force to disturb the status quo), commit war crimes, and commit atrocities against their own citizens (such as genocide). The penalty for violation would be criminal prosecution. The flip side of this aggressive stance was that nations that did not do any of these things had nothing to fear from the United States and the Soviet Union, or from any other major power. Or, to shoehorn this analysis into the earlier categories, legitimate international political competition could no longer involve aggressive war or crimes against humanity.

The advantages of holding international criminal trials following the conclusion of World War II can now be seen. The trials would show that the victorious powers were not interested in eliminating any person who happened to be a threat to their international ambitions, nor even in exercising the traditional victor's prerogative of taking revenge. The victorious powers sought to punish only those who had, through their actions, endorsed the view that aggressive war, genocide, and similar actions were legitimate forms of international action. To persuade the world of their good faith, the victorious powers had to give the defendants an opportunity to be heard, to show that they had not engaged in the actions that, under the new order, would be considered crimes. Otherwise, the world would have no reason to believe that the victors sought to create a new international order that involved some self-restraint on their own ambitions.

This was the ideal, but the managers of the trial ran into trouble from the beginning. None of the victors had acted consistently with the new rules that they were now pressing on the Germans, so the world might understandably react with skepticism to the victors' claims that the rules were sustainable. Russia had engaged in aggressive war against Poland, Finland, and the Baltic states; the United States had dealt with its own internal racial problems with highly oppressive laws that could be likened to crimes against humanity; and all sides had committed atrocities in their conduct of the war. Within limits, the charges could be qualified; the Americans, for example, insisted that an element of the crime against humanity was that the atrocities occurred in a wartime setting, so as to avoid the charge that America's racial laws were crimes against humanity. The defendants could, and did, try to make an issue of these inconsistencies; they could, and were, shushed by the courts (the London charter banned the "tu quoque" defense); bur the fact remains that the rest of the world knew about the inconsistencies. The problem was not so much one of victor's justice, but a more basic problem about world order. If the states that were pressing for new norms of international conduct had never complied with them, what reason was there to think that they could be sustainable?

At roughly the same time, the managers of the Tokyo trial were running into similar problems. The goals of this trial were the same as the goals of the Nuremberg trial: to eliminate powerful and influential people in the Japanese government, and to show that their elimination was tied to their violation to new norms of international legality, which would provide the basis for a peaceful world order. The problem was, again, that the United States as well as Japan had committed war crimes, and although Japan had clearly brought war to the United States, the roots of this war could be traced to a competition between two great powers for colonial influence in the west Pacific. No clear norms differentiated the conduct of the United States and Japan--Japan had Nanking, the United States had Hiroshima. These points were made by one of the dissenters, Justice Pal. (199) For Pal and other critics such as Professor Shklar, the difference between Nuremberg and Tokyo was that the Germans engaged in mass extermination of their own citizens; the Japanese did not (though they did kill thousands of foreign citizens). (200)

The literature considers Nuremberg a success and Tokyo a failure. One might doubt this judgment and conclude that both were failures. As noted in Section C, neither trial had clear, beneficial effects on public opinion; neither trial established much of a precedent. International criminal trials would not be used again for almost fifty years. In contrast to domestic political trials, we might infer that international political trials are unlikely to succeed because there is no international consensus on the line between legitimate (international) political competition and illegitimate international political competition. A consensus of states was not willing to accept the premise, for example, that a great power should not start a war to protect access to resources, as Japan did.

If Nuremberg and Tokyo were failures, or if Nuremberg alone was a success because of the uniquely evil behavior of the Nazis, it can be readily understood why subsequent international criminal trials have been rare and not particularly successful. The problem has always been that the U.S. and other states have been complicit in the reign of the defeated leader or that they have been unable to obtain a complete surrender. Manuel Noriega, Saddam Hussein, Slobodan Milosevic--these are people whom the U.S. at one time or another dealt with as legitimate leaders. (201) Thus, the claim that they should be prosecuted for engaging in illegitimate behavior rings hollow. The judges have tried to deflect charges of bias by granting defendants an extremely high level of process, with the result that the trials drag on for years while the defendants use their trials as platforms for stirring up resentment and xenophobia at home. (202) But no amount of process can overcome the fundamental problem of bias, which has nothing to do with the procedures, but rather with the bate fact that the trials are used by strong nations to assert their international ambitions at the expense of weak nations.

2. The International Criminal Court. If only the strongest states have the power to establish international tribunals, determine their memberships, set their agendas, and thus influence the development of international criminal law, predictably the resulting norms of international criminal law will reflect the interests of the strong states, not the weak ones. Thus, at Nuremberg, crimes against humanity were tied to war to immunize the United States from claims that its racial policies violated international criminal law. But weaker states could hardly be expected to submit to such an international order, and their efforts to gain influence have resulted in the establishment of the International Criminal Court (ICC).

The ICC has an independent prosecutor and tribunal, and jurisdiction over international crimes committed by nationals of states parties and over international crimes committed on the territory of states parties. (203) In theory, the ICC prosecutor decides on prosecutions in the same impartial way that ordinary prosecutors are supposed to: on the basis of the seriousness of the crimes and the amount of resources available. What would this mean? Prosecutors would need to make the case that the crimes that they investigate are more serious than the crimes that they do not investigate; judges would need to persuade the world that convictions and sentences flow impartially from the extremely vague rules of law.

This structure avoids the obvious forms of victor's justice, but it introduces its own set of problems. On the one hand, because the U.N. Security Council no longer chooses which crises will be subject to investigation and prosecution, (204) and because ICC members can no longer immunize themselves from prosecution, defendants cannot argue that the ICC is a tool of the great powers to suppress weak states they do not like. On the other hand, the ICC prosecutor and judges are human beings with their own national origins and biases, eager to satisfy whichever governments determine whether they retain their positions or obtain new ones. International politics still determines who is prosecuted and tried, which means that defendants can argue that their trials are politically motivated.

One might argue that the new politics of international criminal adjudication under the ICC will reflect something like a global consensus on international crime, rather than the views of the United States, Russia, China, Britain, France, and a few other members of the Security Council. Defendants before the ICC will thus be arguing against the world, not a few great powers, and the claim of political motivation will be weaker. However, this benefit is purchased at a high price. Because the great powers no longer see themselves as having dominant influence over the development of international criminal law under the ICC, they have no reason to support it. (205) Thus, the main military and economic powers--predominantly the U.S., but also Russia and China (206)--have made it clear that they will have nothing to do with the ICC, and the U.S. has gone to great lengths to undermine it. (207) Deprived both of the political support of the most powerful countries and the military means that these countries supply, the ICC is a frail institution with a dim future.

International criminal trials, then, are analogous to domestic trials in a divided state. When the public is sufficiently divided, all trials will look like efforts by the government to eliminate its political opponents rather than like vindications of community norms incorporated in the law. In the international setting, international criminal tribunals will similarly look like efforts by the governments that influence the prosecutor and judges--whether the Security Council (in ad hoc cases) or the members of the ICC--to harass or embarrass states with contrary foreign policy objectives. The states whose nationals are being tried will always make this charge, however faithfully the prosecutor and judges try to carry out their duties.

This charge will be hard to deflect, not only because it is difficult to prove to the world that it is hard to convict a criminal who has not been tried, but also because one's judgment of the seriousness of a crime is unavoidably political. Suppose, for example, that leaders in a war of national liberation are responsible for the deaths of thousands of civilians who have been inadequately separated from the battlefield, whereas the authoritarian government on the other side has killed only a few hundred. Are the revolutionaries' crimes more serious because more extensive or less serious because in the pursuit of a good cause? Would it matter if the rest of the world sided with the authoritarian government, or with the rebels; or what if the rest of the world is split between them?

The American experience with domestic political trials is instructive. Domestic political trials can be politically successful only when they target extremists who can plausibly be considered threats--in the long or short term--to American political institutions and mainstream values. When domestic political trials are used against mainstream political opponents, as during the Federalist period, they contribute to political division rather than overcome it. The Nazis committed extraordinary atrocities, threatened all of the world's major powers, and were thoroughly defeated; in these unique circumstances, there was enough international political consensus about the undesirability of the Nazi system to make an international trial possible--albeit one that was heavily controlled and quite controversial. Every other international trial has been a political failure--or, at least, cannot claim visible evidence of political success--either because it has introduced further divisiveness into an arena in which consensus was absent, or because such divisions prevented it from accomplishing its goal.

CONCLUSION

Political trials owe their uneasy status in liberal democracies to their paradoxical role, which is to eliminate enemies of a system devoted to political tolerance. In a political trial, the normal constraints of liberal legalism yield, with judicial and public acquiescence, to the political imperative of self-preservation; this relaxation also creates an opportunity for the government to stifle legitimate political competition. Why do governments, then, not conduct political trials more often? The reason is that people understand that political trials can be used for partisan purposes, and a government that conducts political trials takes the risk that the public, or a large portion of it, will withdraw its support. For this reason, governments in liberal democracies conduct political trials only when they can plausibly claim a national emergency, and the public tolerates and approves of such trials only when it believes that the national emergency justifies giving the government powers that it can that it can potentially misuse for partisan gain. Governments can further enhance their political support, however, by granting what one might call "political process" to the defendants in political trials--the power to make a political defense of their activities. Whether the trial is conducted for partisan gain or public safety itself becomes an issue in the trial, and by allowing this issue to be addressed, the government shows that it believes that its policies enjoy the support of the public and that the defendant's views do not. The political trial is a high-stakes exercise. If the government persuades the public that the defendant or the defendant's views pose a threat to the nation, then its legitimacy will be enhanced, and the threat will be removed. If it does not, then it may lose public support and provoke a constitutional crisis.

If this view is correct, it is too simple to criticize a political trial on the grounds that the defendant is being punished for particular political beliefs, or that the judge has unfairly relaxed normal process protections. The problem with this criticism is that sometimes departure from ordinary due process is justified. Due process standards have evolved to address normal criminal activity and do not reflect the special circumstances of an emergency or other crisis. Consider the contemporary example of the Guantanamo Bay military commissions. It is not appropriate to criticize them on the ground that they do not provide detainees with the process enjoyed by ordinary criminal defendants under domestic law or even with the level of process enjoyed by enemy soldiers who are tried for committing war crimes. The special circumstances created by the threat of further terrorism by al Qaeda call for special procedures that are tailored to that threat.

I cannot provide a complete analysis of the problem here, but a few observations may be useful. Relaxed procedures are justified by the magnitude of the threat and the importance of secrecy. At what point should process be set, then? It seems to me that the answer is mainly one of international politics: to the extent that foreign states object that their citizens are being convicted for crimes without adequate process, the U.S. will need to take these objections seriously (which it has). There are also moral limits on what a government can properly do to an alien (or its own citizens) in the course of protecting itself, but if international law reflects the standard, these limits are minimal. (208) Judicial involvement should be limited because there is no reason to think that the executive will not make the proper trade-offs; at least, there is no more reason to think so than in any other life-and-death foreign policy decision that judges refrain from reviewing. So far, it is clear that the Bush administration is not using military commissions to target legitimate political opponents.

In the case of domestic (including transitional) political trials, the real issue is whether the government makes a plausible case that the defendant--alone, as part of a group, or as a symbol for a certain message--poses a genuine threat to public safety or the constitutional system. If not, it is reasonable to object to the trial on the ground that the government is using the judicial process to eliminate legitimate, even if extreme, political opponents. But what makes the question difficult in any case is that the government will have legitimate reasons not to disclose all its evidence and can point to heightened security risks that justify public and judicial deference to its claims. In the case of international political trials, the real issue is whether the defendants or their worldview pose a threat to the international order, or at least are likely to make it worse rather than better. On this view, the conventional wisdom about Nuremberg is incorrect. It was not justified because it created new norms of international crime; those norms could have been created, without a trial, via treaty. Indeed, they were, in part. It was justified because, or to the extent that, it persuaded people around the world to abandon fascism and embrace liberal democracy (if in fact it did do this).

The role of judges in both the domestic and international cases is, to a large extent, political. In the domestic cases, judges will not always know whether a prosecution is a regular criminal action or a political action, but they will have their suspicions. When they believe that the prosecution is political but are unsure whether the defendant is a public threat or not, they often can--and perhaps ought to--relax legal process and demand political process. Relaxing legal process may include broadly interpreting statutes, permitting selective prosecution or retroactive lawmaking, limiting the choice of defense lawyer, preventing the defendant from seeing classified evidence, or acquiescing in military trials. Granting political process includes, chiefly, allowing the defendant to mount a political defense--typically, that the government's motives are partisan and the judicial system is corrupt. As the decorum of the judicial forum yields to the circuslike atmosphere of democratic politics, many people will condemn the trial as a farce. But allowing politics into the courtroom may be preferable to the stark alternatives: preventing the government from countering genuine public threats or allowing the government to eliminate its political opponents under cover of judicial process.

(1.) See generally FROM NUREMBERG TO THE HAGUE: THE FUTURE OF INTERNATIONAL CRIMINAL JUSTICE (Philippe Sands ed., 2003). The Nuremberg trials--and especially the first trial of major war criminals--are the subject of a large literature. See generally ROBERT E. CONOT, JUSTICE AT NUREMBERG (1983); JOSEPH E. PERSICO, NUREMBERG: INFAMY ON TRIAL (1994); ANN TUSA & JON TUSA, THE NUREMBERG TRIAL (1983).

(2.) Accounts of many of these trials are collected in 2 TRANSITIONAL JUSTICE (Neil J. Kritz ed., 1995).

(3.) See, e.g., JUDITH SHKLAR, LEGALISM: LAW, MORALS, AND POLITICAL TRIALS 155-70 (1964).

(4.) See generally, e.g., MICHAL R. BELKNAP, COLD WAR POLITICAL JUSTICE: THE SMITH ACT, THE COMMUNIST PARTY, AND AMERICAN CIVIL LIBERTIES (1977); THE CONSTITUTION IN WARTIME (Mark Tushnet ed., 2005); VICTOR S. NAVASKY, NAMING NAMES (2003); H.C. PETERSON & GILBERT C. FITE, OPPONENTS OF WAR 1917-1918 (1986); SHKLAR, supra note 3; JAMES MORTON SMITH, FREEDOM'S FETTERS: THE ALIEN AND SEDITION LAWS AND AMERICAN CIVIL LIBERTIES (1956); PETER L. STEINBERG, THE GREAT "RED MENACE": UNITED STATES PROSECUTIONS OF AMERICAN COMMUNISTS, 1947-52 (1984); GEOFFREY STONE, PERILOUS TIMES: FREE SPEECH IN WARTIME (2004). Hannah Arendt's criticisms of the Eichmann trial reflect the conventional wisdom about political trials, though she was apparently not opposed to a different kind of political trial, that is, one conducted by an international tribunal rather than in an Israeli court. See HANNAH ARENDT, EICHMANN IN JERUSALEM 269 (1994) (arguing that charges against Eichmann were crimes against humanity that should have been heard by an international court). See also Akhil Reed Amar, The Bill of Rights as a Constitution, 100 YALE L.J. 1131 (1991), although Amar emphasizes the role of the jury.

(5.) I will discuss many examples of the first in Part III.C, infra; Lincoln's refusal to obey Justice Taney's order in Ex parte Merryman, 17 F. Cas. 144 (C.D. Md. 1861), is the preeminent American example of the second.

(6.) Thus, this Article can be placed with recent work discussing how the president's powers should (or should not) change during emergences. Compare Oren Gross, Chaos and Rules: Should Responses to Violent Crises Always Be Constitutional?, 112 YALE L.J. 1011, 1023-24 (2003) (arguing that the president should act lawlessly and then seek public ratification), with Bruce Ackerman, The Emergency Constitution, 113 YALE L.J. 1029, 1030-32 (2004) (advocating a statutory scheme that would grant the president special powers during an emergency). My more modest point is that any time security concerns are heightened (whether or not there is a true "emergency"), process protections will be relaxed, and this relaxation can be done in a manner that is consistent with basic liberal, democratic principles, as long as one pays attention to the difference between suspects who are public threats and suspects who are mainstream political opponents.

(7.) See, e.g., STONE, supra note 4, at 542-50.

(8.) My approach is thus to treat liberal principles--like legalism and the rule of law--as instrumental rather than ideological; this is in the spirit of Stephen Holmes, Lineages of the Rule of Law, in DEMOCRACY AND THE RULE OF LAW 19 (Jose Maria Maravall & Adam Przeworski eds., 2003), and Adam Przeworski, Why Do Political Parties Obey the Results of Elections?, in DEMOCRACY AND THE RULE OF LAW, supra, at 114. The approach is to ask why the people with power--the wealthy, the well born, those with guns--would be willing to commit themselves to the rule of law (Holmes), elections (Przeworski), and other liberal democratic principles. The answer is assumed to be that it lies in their self-interest, not in the "inherently binding power of norms" or "legitimacy." Holmes, supra, at 24; see also Barry Weingast, The Political Foundations of Democracy and the Rule of Law, 91 AM. POL. SCI. REV. 245 (1997) (presenting a model of politics in which the government exercises self-restraint to avoid sanctions from opposition groups). For a related approach by a legal scholar, see Daniel A. Farber, Rights as Signals, 31 J. LEGAL STUD. 83 (2002) (arguing that when states make constitutional commitments to protect human rights, and grant the judiciary independence to enforce them, they signal that they have a long time horizon, and therefore are unlikely to expropriate investments).

(9.) In particular, the Hiss trial does not seem political. Hiss was prosecuted for committing perjury, and his trials (the first ended in deadlock) seem to have been fair. See ALLEN WEINSTEIN, PERJURY: THE HISS-CHAMBERS CASE 412-502 (1978) (describing the trials). The same is true for the Sacco and Vanzetti trial and the trial of the Rosenbergs. But many people regarded these trials as political at the time because the defendants had radical political views, and they assumed that the government had trumped up charges to weaken political opposition. See DAVID CAUTE, THE GREAT FEAR: THE ANTI-COMMUNIST PURGE UNDER TRUMAN AND EISENHOWER 61 (1978) (describing the public reaction to Hiss: "obviously guilty not only to conservatives but also to Cold War liberals, obviously innocent, the victim of a frame-up, to almost all who deplored the purge").

(10.) For international tribunals, see generally RICHARD H. MINEAR, VICTOR'S JUSTICE: THE TOKYO WAR CRIMES TRIAL 31 (2001) (Tokyo); TUSA & TUSA, supra note 1, at 504 (Nuremberg); James Meernik, Victor's Justice or the Law?, 47 J. CONFLICT RESOL. 140, 154 (2003) (Yugoslavia tribunal). For transitions, see generally 2 TRANSITIONAL JUSTICE, supra note 2; MICHAEL WALZER, REGICIDE AND REVOLUTION (1974). For American domestic trials, see generally AMERICAN POLITICAL TRIALS (Michal R. Belknap ed., 1994); POLITICAL TRIALS IN HISTORY: FROM ANTIQUITY TO THE PRESENT (Ron Christenson ed., 1991). On Sedition Act trials, see generally SMITH, supra note 4. On Civil War trials, see generally J.G. RANDALL, CONSTITUTIONAL PROBLEMS UNDER LINCOLN (rev. ed. 1951). On World War I-era trials, see generally PETERSON & FITE, supra note 4. On Smith Act trials, see generally STEINBERG, supra note 4. On the Haymarket trial, see generally PAUL AVRICH, THE HAYMARKET TRAGEDY (1984); JOHN F. BANNAN & ROSEMARY S. BANNAN, LAW, MORALITY AND VIETNAM: THE PEACE MILITANTS AND THE COURTS (1974). Other sources are cited infra.

(11) When multiple charges were made, the main charge or a representative charge is listed.

(12.) Excluded are ordinary military trials of captured enemy soldiers or noncombatants who have committed war crimes. In some ways, these are political trials, but in the main they are not: they are based on laws for which there is an international consensus.

(13.) GARY BASS, STAY THE HAND OF VENGEANCE 106-46 (2001).

(14.) The statute authorizing the trial passed only with the approval of the U.S., and the U.S. has sent a team of attorneys to participate in the preparation of the trial. See Neil A. Lewis & David Johnston, The Struggle for Iraq: War Crimes; U.S. Team Is Sent to Develop Case in Hussein Trial, N.Y. TIMES, Mar. 7, 2004, at 1.

(15.) Professor Martha Minow argues that Nuremberg inspired the trials of Adolf Eichmann in Israel, Argentina's prosecution of members of its military, Germany's border guard trials, and the trial of Jaruzelski in Poland, but, as she also notes, this claim is in tension with "the enormous gap in time between the Nuremberg trials and any comparable effort to prosecute war crimes in international settings." MARTHA MINOW, BETWEEN VENGEANCE AND FORGIVENESS: FACING HISTORY AFTER GENOCIDE AND MASS VIOLENCE 27 (1998). See also the valuable discussion of Argentina's experience in CARLOS SANTIAGO NINO, RADICAL EVIL ON TRIAL (1996), which notes the relationship between the Nuremberg trial and the trial of Argentine leaders. Whatever the truth about Nuremberg, I have found no support in the historical literature for Professor Minow's claim that the Tokyo tribunal has had similar positive influence. MINOW, supra, at 27. Most accounts of it are decidedly negative, e.g., the chapters in THE TOKYO WAR CRIMES TRIAL: AN INTERNATIONAL SYMPOSIUM (C. Hosoya et al. eds., 1986), especially B.V.A. Roling's lucid Introduction, in id., at 15. See also the historical accounts cited elsewhere.

(16.) There are a few other ambiguous cases, including the Rwanda and Sierra Leone tribunals.

(17.) I exclude trials that involve controversial laws or defenses, like the battered-spouse defense, which have political resonance, and trials in which the government tries to enforce a generally applicable law (say, against murder) but the jury nullifies the law and acquits a guilty defendant for political reasons. In so doing, the jury is licensing the murder of its political opponents. See, e.g., DALLIN H. OAKS & MARVIN S. HULL, CARTHAGE CONSPIRACY: THE TRIAL OF THE ACCUSED ASSASSINS OF JOSEPH SMITH 184-86 (1975) (describing the acquittal of Smith's murderers by a jury consisting of anti-Mormon citizens). However objectionable, this is not a problem of the government trying to maintain its power by eliminating its political opponents and is thus outside the scope of this paper.

(18.) For a discussion of the complex relationship between political trials and trials used to determine whether a person is an enemy combatant, see infra Part IV.A.

(19.) My definition falls between the two extremes in the literature. Professor Otto Kirchheimer limits political trials to trials that have partisan motivations. OTTO KIRCHHEIMER, POLITICAL JUSTICE 49 (1967). Professor Michal Belknap, on the other hand, includes the trial that</p>

<pre> is intended to affect the structure, personnel, or policies of government, that is the product of or has its outcome determined by political controversy, or that results from the efforts of a group within society having control of the machinery of government to use the courts to disadvantage its rivals in a power struggle which is not itself immediately political or to preserve its own economic or social position. </pre> <p>Michal R. Belknap, Introduction, in AMERICAN POLITICAL TRIALS, supra note 10, at xvi. This definition would require classifying almost any trial as political. The definition I use excludes trials (fairly) based on laws that derive from the constitutional bargain. Of course, if one thinks of the constitutional bargain as itself "political," then all trials are political, but then one cannot make a useful distinction between routine criminal trials and the kind of troublesome trials that dominate the literature.

(20.) See, e.g., sources cited supra note 4.

(21.) See SHKLAR, supra note 3, at 170-79.

(22.) By contrast, she views the Tokyo trial as a "dud," because, in her view, Japanese traditions and culture were not receptive to Western style legalism, id. at 181, and the Japanese simply did not behave as badly as the Germans--no crimes against humanity, only war crimes and crimes of aggression--and thus could portray themselves as moral equivalents of the victors. The first point is simplistic: Japanese ethical traditions differ in complex ways from Western ethical traditions, but if her claim is that the Japanese were less likely to condemn the behavior of fellow citizens than the behavior of foreigners, this feature of their moral system would hardly distinguish them from the Germans or the Americans.

(23.) Id. at 156; see also Bernard D. Meltzer, "War Crimes": The Nuremberg Trial and the Tribunal for the Former Yugoslavia, 30 VAL. U. L. REV. 895, 907 (1996) (describing how the Nuremberg trial provided closure after World War II by satisfying people's desire for judgment, limiting vigilante justice, and reintegrating Germany into the rest of Europe). Professor Shklar's arguments would be echoed in Professor Michael Walzer's defense of the trials of Charles I and Louis XVI, which, by symbolizing the end of monarchy, performed a valuable political function, the ushering in of democracy. See WALZER, supra note 10, at 5-6. For a contrasting view, see MINEAR, supra note 10, who attacks the Tokyo war crimes tribunal for violating the norms of legality. His view appears to be that enemy leaders should not be tried if they are "sincere men"; if they are not, summary execution would be appropriate. Id. at 179-80. He appears to condemn Nuremberg as well. Id. at 169.

(24.) SHKLAR, supra note 3, at 220.

(25.) 183 F.2d 201 (2d Cir. 1950).

(26.) Id. at 212.

(27.) Id. at 213.

(28.) See SHKLAR, supra note 3, at 217.

(29.) See Dennis v. United States, 341 U.S. 494, 568 (1951) (Jackson, J., concurring) (explaining that a "rule of reason" will distinguish between a "danger of substantive evil or a harmless letting off of steam"). That Communists ought to be granted a lower level of process was apparently a general view of the Supreme Court. See William M. Wiecek, The Legal Foundations of Domestic Anticommunism: The Background of Dennis v. United States, 2001 SUP. CT. REV. 375, 377. This was also the view of President Harry Truman, who "was willing, as were some other liberals, to destroy the CPUSA in order to maintain basic liberties of all other citizens." STEINBERG, supra note 4, at 289.

(30.) SHKLAR, supra note 3, at 215.

(31.) Id. at 219.

(32.) Id. at 210.

(33.) History has been kinder to Judge Hand than to Professor Shklar. The disclosure of the Venona cables and the opening of Soviet archives in the 1990s revealed that the CPUSA was dominated by the Soviet Union and used for espionage purposes, although by the end of World War II the effectiveness of the Soviet espionage network in the United States had been undermined by defections, counterintelligence, and purges. HARVEY KLEHR ET AL., THE SOVIET WORLD OF AMERICAN COMMUNISM 4-5 (1998); ALLEN WEINSTEIN & ALEXANDER VASSILIEV, THE HAUNTED WOOD 339-44 (1999).

(34.) Otto Kirchheimer's exhaustive book on political trials, POLITICAL JUSTICE, supra note 19, is mainly descriptive and analytical--he categorizes political trials, shows how they work, illustrates the tensions, and so forth. His few normative comments seem to reflect ambivalence, but the main tone is one of distaste. In this way, he anticipates Shklar, though be does not draw her dogmatic line around political trials in liberal democracies. Others have defended political trials in transitional settings. See, e.g., BASS, supra note 13, at 310 (arguing that war crimes tribunals may prevent victims from taking justice into their own hands); WALZER, supra note 10, at 86-89 (approving of the trials of Charles I and Louis XVI for establishing a symbolic break with monarchy).

(35.) See generally CHARLES F. ABEL & FRANK H. MARSH, IN DEFENSE OF POLITICAL TRIALS (1994).

(36.) POLITICAL TRIALS IN HISTORY, supra note 10.

(37.) The legal literature is also mostly unrelated to the arguments in this Article. The post-World War II debates about positivism were inspired by political trials, such as the famous postwar prosecution of a woman who informed on her husband under Nazi rule in order to get rid of him. However, this political trial was the vehicle for investigating the concept of law, and the chief figures in the debate--Hart and Fuller--did not discuss how political trials should be conducted. Oddly, they both appear to have thought that the trial would be morally justified, with the main difference being that Hart thought Germany should pass an ex post facto law before prosecuting the woman whereas Fuller thought that the courts should declare Nazi law void and enforce the law that existed at the rime of the Weimar Republic. Compare H.L.A. Hart, Positivism and the Separation of Law and Morals, 71 HARV. L. REV. 593, 618-21 (1958), with Lon L. Fuller, Positivism and Fidelity to Law--A Reply to Professor Hart, 71 HARV. L. REV. 630, 659-60 (1958). It is doubtful that their jurisprudential disagreements could explain this disagreement about means.

Another literature focuses on the choices faced by transitional governments. See generally, e.g., MINOW, supra note 15; CLAUS OFFE, VARIETIES OF TRANSITION: THE EAST EUROPEAN AND EAST GERMAN EXPERIENCE (1996); MARK OSIEL, MASS ATROCITY, COLLECTIVE MEMORY, AND THE LAW (1997); RUTI TEITEL, TRANSITIONAL JUSTICE (2002); Eric A. Posner & Adrian Vermeule, Transitional Justice as Ordinary Justice, 117 HARV. L. REV. 762 (2003). I will address some of the arguments in this literature in Part IV.

(38.) I will discuss international law in Part IV.

(39.) U.S. CONST. art. 2, [section] 4.

(40.) RAOUL BERGER, IMPEACHMENT: THE CONSTITUTIONAL PROBLEMS 56-59 (1973).

(41.) See Articles of Impeachment Against Andrew Johnson, art. X, Cong. Globe, 40th Cong., 2d Sess. 1615 (1868) ("[President Johnson] did attempt to bring into disgrace, ridicule, hatred, contempt, and reproach the Congress ... [and] to impair and destroy the regard and respect of all the good people of the United States for the Congress and legislative power thereof...."); Articles of Impeachment Against William Jefferson Clinton, art. IV, H.R. Res. 611, 105th Cong. (1998) ("William Jefferson Clinton has undermined the integrity of his office, has brought disrepute on the Presidency, has betrayed his trust as President, and has acted in a manner subversive of the rule of law and justice, to the manifest injury of the people of the United States.").

(42.) BERGER, supra note 40, at 53-59.

(43.) There are also political trials under the laws of the states, and various state constitutional and statutory constraints, which I will ignore. A few examples are discussed in BELKNAP, supra note 4, at 223-25, which describes the efforts of state bar associations, beginning with Maryland and New York, to secure counsel for accused Communists.

(44.) See Dennis v. United States, 341 U.S. 494, 517 (1951); see also Wiecek, supra note 29, at 406-07.

(45.) Yates v. United States, 354 U.S. 298, 303 (1957); see also BELKNAP, supra note 4, at 157.

(46.) The current doctrine requires "imminent" harm, Brandenburg v. Ohio, 395 U.S. 444, 447 (1969), bur that has not always been the case. See supra notes 44-45 and accompanying text. In other democratic countries, such as Germany, which proscribes the Nazi Party, and Turkey, which periodically proscribes Islamic parties, there is no such requirement. See John E. Finn, Electoral Regimes and the Proscription of Anti-democratic Parties, in THE DEMOCRATIC EXPERIENCE AND POLITICAL VIOLENCE 51, 70-74 (David C. Rapoport & Leonard Weinberg eds., 2001) (listing these proscriptions as well as those in other countries); Walter F. Murphy, Excluding Political Parties: Problems for Democratic and Constitutional Theory, in GERMANY AND ITS BASIC LAW 173, 180-87 (Paul Kirchhof & Donald P. Kommers eds., 1993) (comparing German and American policies on excluding certain parties from participation in the political process).

(47.) Teah R. Lupton, Prosecutorial Discretion, 90 GEO. L.J. 1279, 1279-83 (2002).

(48.) Id. at 1286-94.

(49.) Id. at 1285-87 & n.651; see, e.g., Wayte v. United States, 470 U.S. 598, 607 (1985) (upholding prosecution of draft resisters selected because they had expressed an intention not to register); United States v. Hastings, 126 F.3d 310, 311 (4th Cir. 1997) (upholding prosecution despite prosecution memo that identified the defendant's political party); see also United States v. Armstrong, 517 U.S. 456, 463-66 (1996) (describing the high standard for asserting a selective prosecution claim).

(50.) However, various administrations have argued that the president's Article II powers authorize him to detain and try enemy combatants without congressional authorization. The Supreme Court has, so far, declined to express a view on this argument; for its most recent statement, see Hamdi v. Rumsfeld, 124 S. Ct. 2633, 263940 (2004).

(51.) A good example of this is the foot-dragging of the acting secretary of labor during the Red Scare of the 1920s. Another is Attorney General Biddle's reluctance to enforce the Smith Act during World War II, except when prodded by Roosevelt. BELKNAP, supra note 4, at 38-40.

(52.) For example, Yates v. United States, 354 U.S. 298, 310 (1957), narrowed the Smith Act. Even before Yates, many trial judges declined to impose the maximum penalty for Smith Act violations. BELKNAP, supra note 4, at 158.

(53.) Such groups have included the American Communist Party, e.g., Dennis v. United States, 341 U.S. 494, 517 (1951), and the Ku Klux Klan, e.g., Bryant v. Zimmerman, 278 U.S. 60, 63 (1928) (upholding a state law that required KKK members to register with the state). The view that Reconstruction-era trials of KKK members were political is defended in Kermit L. Hall, Political Power and Constitutional Legitimacy: The South Carolina Ku Klux Klan Trials, 1871-1872, 33 EMORY L.J. 921 (1984). Professor Hall argues that the trials were used to consolidate Republican power in the South; however, Hall's definition of political trial is broader than that used here.

(54.) See BERGER, supra note 40, at 7-52; see generally PETER CHARLES HOFFER & N.E.H. HULL, IMPEACHMENT IN AMERICA, 1635-1805 (1984).

(55.) See generally WILLIAM REHNQUIST, GRAND INQUESTS (1992); Joseph Isenbergh, Impeachment and Presidential Immunity from Judicial Process, 18 YALE L. & POL'Y REV. 53 (1999); Cass Sunstein, Impeaching the President, 147 U. PA. L. REV. 279 (1998).

(56.) Legalism is more or less synonymous with the notion of the rule of law, though it has a pejorative connotation, suggesting excessive concern with following the rules. On this, see the essays collected in DEMOCRACY AND THE RULE OF LAW, supra note 8.

(57.) This seems to be Professor Shklar's view with respect to the domestic setting, which puts great weight on the harm principle. SHKLAR, supra note 3, at 60-70.

(58.) Id. at 219-20.

(59.) The lure of legalism remains, however, as can be seen in the judicial response to the War on Terror. For example, contrast the government's view that it can classify individuals, including Americans, as enemy combatants with virtually no judicial oversight (a view endorsed by Justice Thomas in his dissent in Hamdi v. Rumsfeld, 124 S. Ct. 2633, 2683 (2004)), and detain them for the duration of hostilities, and the Hamdi plurality's view that a person classified as an enemy combatant is entitled to due process protections if the classification is contested, id. at 2645-52.

(60.) SHKLAR, supra note 3, at 214-15, 217-19.

(61.) See generally ETERNALLY VIGILANT: FREE SPEECH IN THE MODERN ERA (Lee C. Bollinger & Geoffrey R. Stone eds., 2002).

(62.) This is roughly the approach of Holmes, supra note 8, and Przeworski, supra note 8.

(63.) See J. Mark Ramseyer, The Puzzling (In)dependence of Courts, 23 J. LEGAL STUD. 721, 721-22 (1994).

(64.) SMITH, supra note 4, at 186.

(65.) Id. at 431. Smith provides several examples that show that defendants convicted of Sedition Act violations often became heroes; in one case, the defendant was an elected official who was rewarded with reelection after he was released from prison. See, e.g., id. at 238, 241, 244, 274, 395.

(66.) There is a large literature on this topic. See, e.g., Dan M. Kahan & Tracey L. Meares, Foreword: The Coming Crisis of Criminal Procedure, 86 GEO. L.J. 1153 (1998) (gang loitering laws); Harry Litman, Pretextual Prosecution, 92 GEO. L.J. 1135 (2004) (vague laws).

(67.) See POLITICAL TRIALS IN HISTORY, supra note 10, at 20-23 (Baader-Meinhof Gang); id. at 168-71 (Irish Republican Army).

(68.) See generally STONE, supra note 4, for a recent discussion.

(69.) In the United States, the executive branch might also seek broad support from Congress, as emphasized by Professors Samuel Issacharoff and Richard H. Pildes, Between Civil Libertarianism and Executive Unilateralism: An Institutional Process Approach to Rights during Wartime, in THE CONSTITUTION IN WARTIME, supra note 4, at 161, 187-94, and by Professor Cass Sunstein, Minimalism at War, 2004 Sup. Cr. REV. 47, 75-77. However, given that the problem of distrust is partisan rather than institutional, I would argue that it was more important for Roosevelt to appoint Knox and Stimson to his cabinet than for him to obtain the acquiescence of the Democratic Congress.

(70.) I will discuss later the extent to which governments can expect judges to acquiesce in this way. See Part III.C. For now, assume that judges will do what they think the government wants them to do.

(71.) See Finn, supra note 46, at 66. Discussing the World War I trials, Professor Peterson notes that "it is a fact that almost immediately after the beginning of World War I people of the political right used the war as an excuse to attack people of the left. They did so by accusing leftists of being disloyal." PETERSON & FITE, supra note 4, at 45; see also id. at 213-21 (discussing the way the sedition law was used against leftist groups).

(72.) This was Justice Jackson's argument in Dennis v. United Sates, 341 U.S. 494, 567-69 (1951). See supra note 29 and accompanying text. On trials of enemy soldiers, see Part IV.A infra.

(73.) Some people might argue that the purpose of show trials is to instill fear and intimidate the public, which is supposed to know that the defendant's confession was the result of torture, and thus that torture is the punishment for political opposition. This is, at best, a small portion of the truth: disappearances or, for that matter, overt violence against political opponents would have served (and did serve) the purpose of intimidation. The great show trials in the Soviet Union had the specific purpose of discrediting Stalin's opponents, intended for foreign as well as domestic audiences (foreign journalists were invited to attend the trials). The trials did not fool everyone in the West, as the charges were often absurd, and some of the facts asserted in the trials could be checked out and disproved. Bur they did fool many influential people in the West, including politicians, journalists, artists, and intellectuals. See ROBERT CONQUEST, THE GREAT TERROR 91 (1990) (describing foreign observers at the trial of the old Bolsheviks); id. at 105-08 (describing the Western reception of the trial); id. at 463-76 (describing the Western reaction to all of the trials of the 1936-1938 period).

(74.) There is a sliding scale, and some devices used in a political trial can make it hard to distinguish from a show trial. In France, the "amalgam" was a device for associating a political opponent with ordinary criminals with whom the defendant never conspired but shared some superficial similarity. One trial in 1894 brought together some anarchists, whose offense was only political, with ordinary criminals, who justified their crimes using anarchist rhetoric but who otherwise had no association with the political defendants. KIRCHHEIMER, supra note 34, at 196 & n.40. The fiction here was more than the court could tolerate.

(75.) A complementary philosophical treatment can be found in JOHN E. FINN, CONSTITUTIONS IN CRISIS: POLITICAL VIOLENCE AND THE RULE OF LAW (1991). Professor Finn argues that, during emergencies, the commitment to constitutionalism can be maintained, even as a particular constitution's requirements are evaded, as long as certain elemental requirements of constitutionalism--reason, deliberation, etc.--are satisfied. Id. at 5-7.

(76.) SMITH, supra note 4, at 432-33.

(77.) A more ambiguous example is the trial of Aaron Burr for treason. See POLITICAL TRIALS IN HISTORY, supra note 10, at 47-50. Burr was a political enemy of Jefferson, but by 1807 he probably could not be considered a part of the mainstream opposition. However, Justice Marshall, who derailed the trial by defining "treason" narrowly, was. But see Robert K. Faulkner, John Marshall and the Burr Trial, 53 J. AM. HIST. 247, 247 (1966) (criticizing the view that Marshall's interpretation of treason law showed political bias).

(78.) See Richard E. Ellis, THE JEFFERSONIAN CRISIS: COURTS AND POLITICS IN THE YOUNG REPUBLIC 278-79 (1971).

(79.) See, e.g., AVRICH, supra note 10, at 280-85 (anarchists); BELKNAP, supra note 4, at 113 (Communists). However, it is important to note that these trials created a political backlash. Several of the Haymarket defendants were ultimately pardoned, and the trial radicalized many workers. See AVRICH, supra note 10, at 307-12 (after trial); id. at 409-14 (after executions); id. at 433-36 (long-term effect).

(80.) During the Civil War, a large number of Northerners suspected of Southern sympathies were detained by the military, with no judicial process. The precise number is unknown, but it was probably in the hundreds (if one limits oneself to the clearest cases) or thousands (especially if one includes draft resisters, unexplained arrests, and so forth). See MARK E. NEELY, JR., THE FATE OF LIBERTY: ABRAHAM LINCOLN AND CIVIL LIBERTIES 51-65, 113-38 (1991) (describing arrests and detentions). Detainees did receive military hearings, which involved regular procedures. Id. at 162-75.

(81.) Harold Josephson, Political Justice During the Red Scare: The Trial of Benjamin Gitlow, in AMERICAN POLITICAL TRIALS, supra note 10, at 154 ("[T]he trials ... created martyrs and enabled radicals to use the issues of freedom of speech and political liberty to rally broad liberal support to their cause.").

(82.) See James W. Ely, Jr., The Chicago Conspiracy Case, in AMERICAN POLITICAL TRIALS, supra note 10, at 249 ("[T]he defendants' effort to appeal over the court to the general public seemingly fell on deaf ears.").

(83.) See HOFFER & HULL, supra note 54, at 3-8 (discussing Great Britain), 256-63 (discussing early state and federal impeachments in the United States).

(84.) BEVERLEY MILTON-EDWADS, ISLAM & POLITICS IN THE CONTEMPORARY WORLD 73 (2004).

(85.) See Finn, supra note 46, at 56 (quoting Article 21(2) of the German Constitution: "Parties which, by reason of their aims or the behavior of their adherents, seek to impair or abolish the free democratic basic order or to endanger the existence of the Federal Republic of Germany, shall be unconstitutional....").

(86.) Id. at 70-74 (describing laws of Chile, Estonia, France, Germany, Ireland, Israel, Italy, Portugal, Romania, and Rwanda).

(87.) SMITH, supra note 4, at 94-95, 176-87.

(88.) See RANDALL, supra note 10, at 177-85.

(89.) POLITICAL TRIALS IN HISTORY, supra note 10, at 95-98.

(90.) See Yates v. United States, 354 U.S. 298, 337 (1957).

(91.) Finn, supra note 46, at 60.

(92.) STONE, supra note 4, at 410.

(93.) The Soviets preferred agents with an ideological motivation because they were more reliable than paid agents. See WEINSTEIN, supra note 9, at 419.

(94.) SMITH, supra note 4, at 431-37; POLITICAL TRIALS IN HISTORY, supra note 10, at 9598.

(95.) General laws against disorderly and similar behavior that could sweep in political activity are politically acceptable for familiar reasons: laws that are prospective and general cannot be easily used against political opponents as they might sweep in political supporters as well. See Adrian Vermeule, Veil of Ignorance Rules in Constitutional Law, 111 YALE L.J. 399, 408-15 (2001).

(96.) Many scholars have noted that many federal (as well as state) laws are so broad and vague that they can be used to criminalize almost any tort or even breach of contract. See, e.g., John C. Coffee, Jr., Does "Unlawful" Mean "Criminal"?: Reflections on the Disappearing Tort/Crime Distinction in American Law, 71 B.U.L. REV. 193, 202-13 (1991) (discussing wire fraud and the Hobbs Act). The effect of these laws is to give prosecutors a great deal of discretion. See Daniel C. Richman, Federal Criminal Law, Congressional Delegation, and Enforcement Decision, 46 UCLA L. REV. 757, 758-60, 789-814 (1999) (discussing institutional mechanisms for limiting prosecutorial discretion); William J. Stuntz, The Pathological Politics of Criminal Law, 100 MICH. L. REV. 505, 512-19, 529-46 (2001) (discussing the reasons why various political actors and interest groups prefer broad criminal laws).

(97.) RONALD RADOSH & JOYCE MILTON, THE ROSENBERG FILE xv-xxii (2d ed. 1997) (discussing the Venona project).

(98.) Kenneth M. Dolbeare & Joel B. Grossman, LeRoi Jones in Newark: A Political Trial?, in THEODORE LEWIS BECKER, POLITICAL TRIALS, 227-32 (1971).

(99.) Posner & Vermeule, supra note 37, at 817.

(100.) See AVRICH, supra note 10, at 268-75 (describing unsuccessful attempts to link the defendants to the unidentified bomb thrower).

(101.) Id. at 277.

(102.) PETERSON & FITE, supra note 4, at 252-54.

(103.) STEINBERG, supra note 4, at 223-24.

(104.) BELKNAP, supra note 4, at 40.

(105.) For the Dennis trial, see BELKNAP, supra note 4, at 77-116, and STEINBERG, supra note 4, at 157-77; for the Chicago 8 trial, see David J. Danelski, The Chicago Conspiracy Trial, in BECKER, supra note 98, at 178-80.

(106.) Cf. Ramseyer, supra note 63, at 728-30; Matthew C. Stephenson, "When the Devil Turns ...": The Political Foundations of Independent Judicial Review, 32 J. LEGAL STUD. 59, 63-64, 83-86 (2003) (arguing that roughly equally matched parties in a competitive party system will tolerate an independent judiciary so that they are protected from persecution when out of power).

(107.) Jed Rubenfeld, Unilateralism and Constitutionalism, 79 N.Y.U.L. REV. 1971, 1994-95 (2004).

(108.) See Ramseyer, supra note 63, at 727-28 ("[J]udges who decided politically sensitive cases according to non-LDP political preferences incurred a substantial risk that the Secretariat would assign them to a series of low-status positions.").

(109.) Irving Kaufman, who presided over the Rosenberg case, and Harold Medina, who presided over the Dennis case, were both elevated to the Second Circuit Court of Appeals. Both judges were highly regarded before these cases, so it is quite possible that they would have been elevated anyway.

(110.) See STONE, supra note 4, at 169-70. His elevation was delayed several years. Other judges who acted similarly were attacked in the press and ostracized. Id.

(111.) BELKNAP, supra note 4, at 113.

(112.) Eric A. Posner & Adrian Vermeule, Accommodating Emergencies, 56 STAN. L. REV. 605, 644 (2003).

(113.) See, e.g., SMITH, supra note 4, at 325-27 ("[T]he presiding judge relieved the jury of even [the duty of deciding on intent] by flatly stating that bad intent had been proved.").

(114.) PETERSON & FITE, supra note 4, at 17.

(115.) See AVRICH, supra note 10, at 263-67.

(116.) POLITICAL TRIALS IN HISTORY, supra note 10, at 78.

(117.) See Dolbeare & Grossman, supra note 98, at 232-34 ("At the end of the trial [the judge] charged the jury so as to leave no doubt of his belief that Jones was a liar and a scoundrel....").

(118.) See Hamdi v. Rumsfeld, 124 S. Ct. 2633, 2648-49 (2004).

(119.) For the Supreme Court's views of the trade-offs involved, see Johnson v. Eisentrager, 339 U.S. 763, 778-80 (1950), which upheld the conviction by military tribunal of nonresident enemy aliens.

(120.) See RANDALL, supra note 10, at 179 & n.16.

(121.) See Ex parte Merryman, 17 F. Cas. 144, 148 (C.C.D. Md. 1861). See generally Ex parte Milligan, 71 U.S. (4 Wall.) 2 (1866).

(122.) Democratic candidate Senator John Kerry did not make an issue of them during the 2004 presidential election campaign.

(123.) Fox News/Opinion Dynamics Poll, July 26-27, 2005, in War on Terrorism, POLLINGREPORT.COM, http://www.pollingreport.com/terror.htm (last viewed Oct. 1, 2005) (noting 65 percent of respondents favored "protecting [their] safety and surroundings from terrorism" over "protecting [their] civil liberties"); Pew Research Center for the People & the Press Survey, July 13-17, 2005, in War on Terrorism, supra (tracking the margin by which respondents' concern that the government "[had] not gone far enough to adequately protect the country" outweighed their concern that it "[had] gone too far in restricting the average person's civil liberties").

(124.) See, e.g., CASS R. SUNSTEIN ET AL., PUNITIVE DAMAGES : HOW JURIES DECIDE 79-95, 245-48 (2003); Kim A. Kamin & Jeffrey J. Rachlinski, Ex-Post [not equal to] Ex Ante: Determining Liberty in Hindsight, 19 LAW & HUM. BEHAV. 89 (1995).

(125.) See Amar, supra note 4, at 1150-51.

(126.) But not always: trials of some prominent radicals during World War I under the Espionage Act resulted in a hung jury. See STONE, supra note 4, at 170 & n.*. Some Vietnam-era trials also ended in acquittals or hung juries. Id. at 483.

(127.) See SMITH, supra note 4 passim. For one exception, see id. at 282.

(128.) Id. at 423; see, e.g., id. at 235-36, 321, 348 (identifying the role that political divisions played in the trials of Lyon, Cooper, and Callendar).

(129.) See id. at 325-27.

(130.) See id. at 321, 348.

(131.) Id. at 187. The preponderance of trials in northern and middle states was due to the greater influence of Federalists in those areas. Id. at 177.

(132.) See Farber, supra note 8, at 16-18; Frank L. Klement, The Indianapolis Treason Trials and Ex Parte Milligan, in AMERICAN POLITICAL TRIALS, supra note 10, at 97, 101-03.

(133.) See Julia Preston, Lawyer Is Guilty of Aiding Terror, N.Y. TIMES, Feb. 11, 2005, at A1 (describing the conviction for providing material aid to a terrorist of a lawyer who represented an al Qaeda member).

(134.) See KIRCHHEIMER, supra note 19, at 253-54 (citing Konigsberg v. State Bar, 353 U.S. 252 (1957)); see also Sacher v. United States, 343 U.S. 1, 38-39 (1951).

(135.) It remains unclear whether defendants must be given civilian lawyers. See generally Hamdi v. Rumsfeld, 124 S. Ct. 2633 (2004).

(136.) Neil A. Lewis, Sudden Shift on Detainee, N.Y. TIMES, Dec. 4, 2003, at A1.

(137.) BELKNAP, supra note 4, at 13-15 (discussing the activities of the International Labor Defense).

(138.) See Danelski, supra note 105, at 164.

(139.) POLITICAL TRIALS IN HISTORY, supra note 10, at 97-98. Harding also commuted the sentences of Abrams and his codefendants, but this was part of a bargain in return for which the defendants exiled themselves from the United States. RICHARD POLENBERG, FIGHTING FAITHS: THE ABRAMS CASE, THE SUPREME COURT, AND FREE SPEECH 336-38 (1987).

(140.) Danelski, supra note 105, at 164.

(141.) BELKNAP, supra note 4, at 40.

(142.) Danelski, supra note 105, at 177-80.

(143.) BELKNAP, supra note 4, at 168.

(144.) These principles are required by the Geneva Conventions. See Geneva Convention (III) Relative to the Treatment of Prisoners of War pt. I, art. III, [section] 1(d), Aug. 12, 1949, 6 U.S.T. 3316, 75 U.N.T.S. 135.

(145.) JOHN W. DOWER, EMBRACING DEFEAT: JAPAN IN THE WAKE OF WORLD WAR II 465 (1999).

(146.) See Hague Convention (II) on Laws and Customs of War on Land art. 1, July 29, 1899, 32 Stat. 1803, T.S. No. 403; Geneva Convention (III), supra note 144, at pt. I, art. IV, 75 U.N.S.T. 138.

(147.) See Ex parte Quirin, 317 U.S. 1, 2 (1942).

(148.) Rumsfeld v. Padilla, 124 S. Ct. 2711, 2715-17 (2004).

(149.) It is not yet clear whether this plan will be acceptable to courts; for an argument that the commissions are constitutional, see generally Curtis A. Bradley & Jack L. Goldsmith, The Constitutional Validity of Military Commissions, 5 GREEN BAG 2D 249 (2002). See also Hamdi v. Rumsfeld, 124 S. Ct. 2633, 2640-42 (2004).

(150.) E.g., Peter Spiro, Not War, Crimes, FINDLAW'S LEGAL COMMENTARY, Sept. 19, 2001, http://writ.findlaw.com/commentary/20010919_spiro.html.

(151.) See, e.g., Hamdi, 124 S. Ct. at 2648; Rasul v. Bush, 124 S. Ct. 2686, 2696 (2004).

(152.) See, e.g., Issacharoff & Pildes, supra note 69, at 181-87; David Luban, The War on Terrorism and the End of Human Rights, in THE CONSTITUTION IN WARTIME, supra note 4, at 219, 221-23.

(153.) However, the use of military commissions has received more criticism after September 11 than during World War II. See Jack Goldsmith & Cass R. Sunstein, Military tribunals and Legal Culture: What a Difference Sixty Years Makes, 19 CONST. COMMENT. 261, 274, 280-89 (2002) (arguing that the difference in reactions is due to greater suspicion of government, greater respect for civil rights, and the lesser magnitude of the threat).

(154.) See supra text accompanying notes 64-65.

(155.) For instance, Stalin's regime did not do so. See CONQUEST, supra note 73, at 92 (discussing trials during Stalin's rule in which sentences were set before the trials even began).

(156.) See Posner & Vermeule, supra note 37, at 767.

(157.) See, e.g., TEITEL, supra note 37, at 29; Jon Elster, Moral Dilemmas of Transitional Justice, in PRACTICAL CONFLICTS: NEW PHILOSOPHICAL ESSAYS (Peter Baumann & Monika Betzler eds., 2002). Professor Walzer makes an analogous argument about the executions of Charles I and Louis XVI. He claims that democracy could not be established until the subservient habits that evolved in a monarchy were eliminated; to him, the symbolism of the trials was of great significance. See WALZER, supra note 10, at 86-89.

(158.) SHKLAR, supra note 3, at 170-79.

(159.) See SAMANTHA POWER, A PROBLEM FROM HELL: AMERICA AND THE AGE OF GENOCIDE 31-47 (2002).

(160.) See, e.g., LAWRENCE DOUGLAS, THE MEMORY OF JUDGMENT: MAKING LAW AND HISTORY IN THE TRIALS OF THE HOLOCAUST (2001) (arguing that these trials can successfully reconcile the demands of legal process and of history).

(161.) This is emphasized by OSIEL, supra note 37, at 53-55, who argues that political trials in transitional settings are important for shaping the collective memory of past atrocities.

(162.) SMITH, supra note 4, at 14.

(163.) See POLITICAL TRIALS IN HISTORY, supra note 10, at 204-28 (discussing the blacklisting of the Hollywood 10 for their refusals to cooperate with the House Committee on Un-American Activities); id. at 271-74 (describing an indictment charging that individuals joined "the Nazi movement to destroy democracy throughout the world").

(164.) See the discussion and citations in Posner & Vermeule, supra note 37, at 762-65.

(165.) See ARENDT, supra note 4, at 135 ("He did his duty ... he not only obeyed orders, he also obeyed the law.").

(166.) Posner & Vermeule, supra note 37, at 793-95; see Kif Augustine Adams, What Is Just?: The Rule of Law and Natural Law in the Trials of Former East German Border Guards, 29 STAN. J. INT'L L. 271, 295-300 (1993).

(167.) A. JAMES MCADAMS, JUDGING THE PAST IN UNIFIED GERMANY 31-34 (2001).

(168.) See Elster, supra note 157, at 308-09; Hart, supra note 37, at 619-20.

(169.) See ARENDT, supra note 4, at 294-95 ("What we have demanded in these trials, where the [defendant] ha[s] committed 'legal' crimes, is that human beings be capable of telling right from wrong.").

(170.) Henry L. Stimson, The Nuremberg Trial: Landmark in Law, 25 FOREIGN AFF. 179 (1947).

(171.) TUSA & TUSA, supra note 1, at 101.

(172.) DOUGLAS, supra note 160, at 185-86.

(173.) TUSA & TUSA, supra note 1, at 138-39.

(174.) Solis Horwitz, The Tokyo Trial, 28 INT'L CONCILIATION 473, 495-97 (1950).

(175.) CONOT, supra note 1, at 87-88 (1983); see also MICHAEL R. MARRUS, THE NUREMBERG WAR CRIMES TRIAL, 1945-46: A DOCUMENTARY HISTORY 193 (1997).

(176.) See, e.g., TUSA & TUSA, supra note 1, at 104-07.

(177.) MARRUS, supra note 175, at 138-39.

(178.) Id. at 56-57. The Soviets subsequently presented witnesses testifying as to German responsibility for the massacre, but when they tried to prevent the Germans from providing their own witnesses, their motion was refused by the tribunal. Id. at 100-01.

(179.) For another example taken from the Klaus Barbie trial, see DOUGLAS, supra note 160, at 189, who discusses the way that the requirements of legal form caused prosecutors to downplay the most serious aspects of Barbie's crimes. Osiel provides further examples from the Barbie, Nuremberg, Eichmann, and Argentine junta trials. See generally OSIEL, supra note 37, at 79-141.

(180.) MINEAR, supra note 10, at 13-14 (discussing Justice Jackson's statements at the London Conference where the Nuremberg charter was drafted).

(181.) TUSA & TUSA, supra note 1, at 260, 304.

(182.) Id. at 223.

(183.) Churchill's "iron curtain" speech at Fulton, Missouri, was the turning point. See id. at 201.

(184.) PUBLIC OPINION IN OCCUPIED GERMANY: THE OMGUS SURVEYS, 1945-1949, at 93, 121, 138 (Anna J. Merritt & Richard L. Merritt eds., 1970); PUBLIC OPINION IN SEMISOVEREIGN GERMANY: THE HICOG SURVEYS, 1949-1955, at 101 (Anna J. Merritt & Richard L. Merritt eds., 1980); see also RICHARD L. MERRITT, DEMOCRACY IMPOSED: U.S. OCCUPATION POLICY AND THE GERMAN PUBLIC, 1945-49, at 160-73 (1995). The trials did not have measurable impact on American public opinion. See WILLIAM J. BOSCH, JUDGMENT ON NUREMBERG: AMERICAN ATTITUDES TOWARD THE MAJOR GERMAN WAR-CRIME TRIALS 87-116 (1970).

(185.) MEIRION HARRIES & SUSIE HARRIES, SHEATHING THE SWORD: THE DEMILITARIZATION OF JAPAN 163 (1987).

(186.) Id. at 175; see also DOWER, supra note 145, at 471.

(187.) See HERBERT P. BIX, HIROHITO AND THE MAKING OF MODERN JAPAN 581-612 (2000) (describing Emperor Hirohito's role and the efforts to conceal it during the trial).

(188.) DOWER, supra note 145, at 459-60.

(189.) HARRIES & HARRIES, supra note 185, at 175-76; see also OSIEL, supra note 37, at 181 n.43 (citing sources recognizing the trial's failure to "take[] root among the Japanese people").

(190.) DOWER, supra note 145, at 444.

(191.) Id. at 474; see also BIX, supra note 187, at 612-18 (describing the complex reactions of the Japanese public). For a general discussion of Japanese attitudes about the Tokyo trial, along with a comparison to German attitudes about the Nuremberg trial, see IAN BURUMA, THE WAGES OF GUILT 159-68 (1994).

(192.) DOWER, supra note 145, at 525-26.

(193.) For a survey of truth commissions, see generally TRUTH COMMISSIONS: A COMPARATIVE ASSESSMENT (Harvard Law School Human Rights Program ed., 1997).

(194.) Professor Gary Bass, by contrast, argues that international trials occur because of the power of the legalist ideal. Leaders of liberal democracies seek to impose liberal values on defeated countries; the trial suggests itself as one way to advance this goal. See BASS, supra note 13, at 8. There are two problems with Bass's argument. First, the story breaks down at the domestic level: both ordinary authoritarian governments and democratic governments rely on courts. Second, international criminal trials from Nuremberg on do not reflect the legalist idea because they do not conform to the rule of law. They are attempts to impose liberalism, not legalism, on other countries.

(195.) I use this terra to refer to trials of leaders and other major actors; for trials of ordinary soldiers and civilians, see supra Part IV.A.

(196.) POLITICAL TRIALS IN HISTORY, supra note 10, at 328.

(197.) SHKLAR, supra note 3, at 162-63.

(198.) Prosecutors claimed that the Kellogg-Briand Pact could provide the basis for the crime of aggressive war, but that agreement was not understood to create legal obligations, and certainly not criminal law applying to individuals. In any event, the Kellogg-Briand pact was a dead letter. The effort to create a crime of aggressive war continues today. The Rome Statute creating the International Criminal Court provides for further negotiation over the definition of such a crime, because states parties could not reach agreement. See Rome Statute of the International Criminal Court art. 5(2), July 1, 2002, available at http://www.un.org/law/icc/statute/english/rome_statute(e).pdf.

(199.) See DOWER, supra note 145, at 471-72.

(200.) Id.; see SHKLAR, supra note 3, at 188-90.

(201.) See, e.g., The U.N. Criminal Tribunals for Yugoslavia and Rwanda: International Justice or Show of Justice: Hearing Before the H. Comm. on Int'l Rel., 107th Cong. 107-71 (2002) (statement of Larry A. Hammond, Attorney at Law, Osborn Maledon, P.A.) (indicating that the U.S. provided assistance to the Croatian offensive against Serbia and may have been complicit in Croatian war crimes).

(202.) These problems were discussed by witnesses before a House hearing on U.N. criminal tribunals. See id. See generally Jacob Katz Cogan, International Criminal Courts and Fair Trials: Difficulties and Prospects, 27 YALE J. INT'L L. 111 (2002); Meernik, supra note 10.

(203.) Rome Statute of the International Criminal Court art. 12, 13(b), July 1, 2002, available at http://www.un.org/law/icc/statute/english/rome_statute(e).pdf.

(204.) It does have referral power, however, and only an extremely weak power to defer investigations. See id. arts. 13(b), 16.

(205.) For a discussion of these problems, see Jack L. Goldsmith, The Self-Defeating International Criminal Court, 70 U. CHI. L. REV. 89, 93-101 (2003).

(206.) A list of states parties can be found at International Criminal Court: The States Parties to the Rome Statute, http://www.icc-epi.int/asp/statesparties.html (last visited Oct. 1, 2005).

(207.) The U.S. has compelled numerous countries to agree not to refer Americans to the ICC. HUMAN RIGHTS WATCH, BILATERAL IMMUNITY AGREEMENTS 7-13 & tbl.1 (2003), http://www.hrw.org/campaigns/icc/docs/bilateralagreements.pdf.

(208.) The laws of war prohibit states from targeting enemy civilians, but states may kill them as incidental to lawful targeting of enemy soldiers and facilities, as long as civilian casualties are not "excessive in relation to the concrete and direct military advantage anticipated." Protocol Additional to the Geneva Conventions of 12 Aug. 1949 and Relating to the Protection of Victims of International Armed Conflicts (Protocol I) pt. IV, art. 51, [subsection] 1, 5(b), June 8, 1977, 1125 U.N.T.S. 3.

Copyright [c] 2005 by Eric A. Posner.

Eric A. Posner, Kirkland & Ellis Professor of Law, University of Chicago. Thanks to Dan Farber, Daryl Levinson, Richard Posner, Mark Ramseyer, Matthew Stephenson, Lior Strahilevitz, Bill Stuntz, and Adrian Vermeule, for comments on an early draft; to participants at a workshop at the University of Chicago Law School; and to the John M. Olin foundation for financial support.
COPYRIGHT 2005 Duke University, School of Law
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2005 Gale, Cengage Learning. All rights reserved.

 
Article Details
Printer friendly Cite/link Email Feedback
Author:Posner, Eric A.
Publication:Duke Law Journal
Date:Oct 1, 2005
Words:32076
Previous Article:Good faith business judgment: a theory of rhetoric in corporate law jurisprudence.
Next Article:A choice of rules in Title VII retaliation claims for negative employer references.
Topics:

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters