Federal extraterritorial prosecutions of terrorists and arms dealers and even narcotics traffickers have become an integral part of modern American criminal justice. But extraterritorial prosecutions raise foundational legal questions—about the fairness of forcing foreign defendants to stand trial in our courts and about the outer boundaries of American power. And extraterritorial prosecutions foreground a puzzling inconsistency in constitutional law. A foreign defendant can invoke due process to challenge the court’s jurisdiction when he or she is sued on a tort or contract claim—but not when the defendant is charged with a crime. This Article argues for a sea change in American law: requiring due process curbs on personal jurisdiction in criminal cases, no less than in civil ones. Due process limits the sovereign’s coercive power, and this justifies the long-established limits, in civil cases, on the jurisdiction of the sovereign’s courts. But if the lesser and more diffuse coercive power brought to bear in everyday civil cases requires due process limits on jurisdiction, then the more forceful and more direct coercive power at issue in criminal cases should also require due process limits. What sort of limits? Rethinking the underlying theory of jurisdiction, this Article argues that the often-invoked “burdens” of litigating in a distant forum are a proxy for a deeper concern—for the kinds of declines in adjudicative accuracy that can foreseeably result when a defendant is forced to litigate in a far-off location. Adjudicative accuracy is a concern in criminal cases just as it is in civil ones. And all the more so in federal extraterritorial prosecutions, where there is a rarely noticed structural gap between federal criminal law (which often reaches abroad) and federal courts’ subpoena power (which virtually never does). In an extraterritorial prosecution, important evidence and witnesses will almost necessarily be outside of the United States. But the defendant cannot use the court’s power to get them in front of the jury. This systematic evidentiary deficiency increases the likelihood of convicting the innocent, and accordingly casts doubt on the legitimacy of extraterritorial prosecutions. A robust due process doctrine can help solve the problem. Treaties empower prosecutors to obtain evidence internationally, and due process limits can incentivize prosecutors to press these treaties into service for defendants, gathering evidence abroad on their behalf. This would ameliorate extraterritorial prosecutions’ accuracy deficit—at no very large cost to public safety, and in accord with some of the deepest commitments of U.S. constitutional law.
INTRODUCTION
- THE CONSTITUTIONAL CLEAN SLATE
- The Problem of Criminal Jurisdiction
- Article III
- Ker-Frisbie
- DUE PROCESS
- For Due Process
- Scope and Method
- THE CONSTITUTIONAL CLEAN SLATE
- The Problem of Criminal Jurisdiction
- Article III
- Ker-Frisbie
- STRUCTURE
- The Internal Approach
- The External Approach
- Conclusion—and Comity
- “BURDEN”
- The Adjudicative Accuracy Theory
- From Civil to Criminal
- Accuracy’s Added Impetus in Criminal Cases
- WHAT DUE PROCESS REQUIRES
CONCLUSION: JURISDICTION TO ADJUDICATE—AND TO LEGISLATE
Introduction
For decades now, the United States has turned to sprawling extraterritorial prosecutions as an instrument of national security policy—prosecutions of terror leaders for bombings in Libya and Algeria; of notorious weapons traffickers selling surface-to-air missiles in Russia and Spain; and of violent drug lords operating in Afghanistan and Colombia. These sorts of federal extraterritorial prosecutions have become an important—and entrenched—part of American criminal justice. And there is every reason to think that extraterritorial prosecutions will only become more common, perhaps even routine, during the coming years, as officials focus more on cybercrime and human trafficking, offenses that tend to have a transnational bent; and as the gathering pace of globalization more frequently lends an international dimension to crimes like securities fraud or even narcotics trafficking.
Federal extraterritorial prosecutions raise foundational questions—about the fairness of hauling defendants halfway around the world and into our courts and about the outer constitutional limits of American power. But the federal courts have hardly grappled with these issues. When a foreign defendant is sued in a tort action or on a contract or property claim, the defendant can challenge the court’s power—its jurisdiction—by invoking a century’s worth of due process jurisprudence, from the still-live parts of Pennoyer v. Neff, 1 95 U.S. 714 (1877). to the fine-spun doctrines of International Shoe 2 Int’l Shoe Co. v. Washington, 326 U.S. 310 (1945). and “minimum contacts.” 3 Id. at 316. But when a foreign defendant is criminally prosecuted, there is, simply, nothing directly comparable. With rare exceptions, the fundamental issue of the court’s criminal power has not been considered by judges or by commentators or by litigators. Jurisdiction is “the first and fundamental question,” 4 Steel Co. v. Citizens for a Better Env’t, 523 U.S. 83, 94 (1998) (quotation marks omitted) (quoting Great S. Fire Proof Hotel Co. v. Jones, 177 U.S. 449, 453 (1900)). and it has been asked and answered everywhere—except, it seems, in the context of extraterritorial prosecutions, where the question is most difficult and, increasingly, matters most.
This Article argues for a sea change in American law: requiring due process limits on personal jurisdiction in criminal cases. Due process limits the exercise of coercive power by a sovereign. And this justifies the familiar limits, in civil cases, on the jurisdiction of the sovereign’s courts. 5 See Goodyear Dunlop Tires Operations, S.A. v. Brown, 131 S. Ct. 2846, 2850 (2011) (“A state court’s assertion of jurisdiction exposes defendants to the State’s coercive power, and is therefore subject to review for compatibility with the Fourteenth Amendment’s Due Process Clause.”). But if the lesser and more diffuse coercive power brought to bear in run-of-the-mill civil cases requires due process limits on jurisdiction, then the more forceful and more direct coercive power imposed in criminal cases should also require due process limits. Constitutional text and history, as well as the logic of modern Supreme Court jurisprudence, buttress this conclusion.
What, precisely, should be the nature of personal jurisdiction limits in criminal cases? Rethinking the underlying theory of jurisdiction, this Article argues that the often-invoked burdens and inconveniences of litigating in a “distant” forum are best understood as a proxy for a deeper concern—a concern for the decline in adjudicative accuracy that can result when a defendant is forced to litigate in a far-off location. 6 s to the meaning in this context of adjudicative accuracy, see infra text accompanying footnotes 223–230. Adjudicative accuracy is a concern that due process must address in criminal cases no less than in civil ones. And all the more so in the context of federal extraterritorial prosecutions, where there is a structural gap between federal criminal law (which extends abroad) and federal courts’ subpoena power (which virtually never does). In an extraterritorial prosecution, evidence and witnesses are certain to be abroad, but the defendant cannot use the court’s power to put hands on them and get them in front of the jury.
This systematic evidentiary deficiency is a fundamental problem. It raises the specter of convicting the innocent and, unremedied, casts doubt on the legitimacy of globally important extraterritorial prosecutions. A robust due process doctrine can help to solve the problem. Courts lack certain powers outside of the United States. But a web of treaties empowers prosecutors to obtain evidence internationally. Due process can incentivize prosecutors to press these treaties into service for the defendant, gathering evidence abroad on his or her behalf. This would ameliorate extraterritorial prosecutions’ accuracy deficit—at no very large cost to public safety, and in accord with some of the deepest commitments of U.S. constitutional law.
The argument is developed in five parts. Part I cleans the slate. Due process checks on federal criminal jurisdiction are thought to be superfluous because Article III already imposes tight venue limits on where prosecutions can be brought. But under Article III, extraterritorial prosecutions—crimes “not committed within any State” 7 U.S. Const. art. III, § 2, cl. 3. —are subject to their own venue requirements, and those are extraordinarily loose. Similarly, due process limits on personal jurisdiction are said to be unavailable in criminal cases because of the Supreme Court’s Ker–Frisbie 8 Frisbie v. Collins, 342 U.S. 519, 523 (1952) (holding court need not divest itself of jurisdiction when defendant’s presence before court was secured by kidnapping); Ker v. Illinois, 119 U.S. 436, 443–44 (1886) (same). jurisprudence. But Ker–Frisbie speaks to the narrow question of how a criminal defendant came to be before the court—not to the broader question of whether jurisdiction can be maintained over a defendant who is before the court.
Part II develops the argument that due process should be understood to limit personal jurisdiction in criminal cases. Given the absence of strong historical or textual guideposts, Part II argues that the specific contours of criminal due process doctrine should be derived by analogy from the more familiar due process doctrines that control personal jurisdiction in civil cases. Parts III and IV begin that work.
As Part III argues, in civil cases personal jurisdiction doctrine is thought to be shaped in part by structural concerns: first, for whether a sovereign’s exercise of power is constitutionally justified; and second, for whether exercising that power negatively impacts other sovereigns. But as to the first point, federal extraterritorial prosecutions are already constitutionally justified, because the federal criminal law they apply must itself pass muster under Article I; accordingly, due process does not need to be brought to bear. And as to the second point, sovereigns can usually protect themselves from any negative impact that might flow from the United States prosecuting one of their nationals—simply by declining to extradite that national to the United States.
In addition to a concern for government structure, in civil cases personal jurisdiction doctrine is molded by a concern for the “burden” imposed on the defendant of having to litigate in a distant forum. Part IV advances the thesis, alluded to above, that this focus on “burden” is largely a surface indication of a deeper concern for the kinds of declines in adjudicative accuracy that foreseeably begin to seep in when cases are brought in far-off locales.
Part V shows what this conception of “burden” would mean in practice for extraterritorial prosecutions, proposing a concrete due process standard that would subject such prosecutions to roughly the same standards of adjudicative accuracy as domestic prosecutions—and that would induce prosecutors to make use of existing treaties to assist defendants in gathering certain foreign evidence.
The Article concludes by showing that the proposed due process standard not only helps to ensure adjudicative accuracy—it is also consistent with some of the bedrock values of our law.
* * *
The time is ripe to consider constitutional criminal jurisdiction. In civil cases, in the decades before World War II, the old strictly territorial conception of jurisdiction, exemplified by Pennoyer v. Neff, 9 95 U.S. 714 (1877). gave way. This happened under pressure from new forms of litigation, 10 See Burnham v. Superior Court, 495 U.S. 604, 617 (1990) (plurality opinion) (describing this process). and then, in 1945, in International Shoe, 11 Int’l Shoe Co. v. Washington, 326 U.S. 310 (1945). the Supreme Court laid down due process rules of the road, explaining what would be a jurisdictional bridge too far. 12 Id. at 316, 319–20 (setting forth “minimum contacts” standard). In the criminal law, today, the story is repeating. The old territorial approach, exemplified by the classic ideal of “local” criminal jurisdiction, 13 See Joseph Story, Commentaries on the Conflict of Laws, Foreign and Domestic 840 (Lawbook Exchange, Ltd. 2008) (8th ed. 1883) [hereinafter Story, Conflict of Laws] (describing crimes as “altogether local, and cognizable and punishable exclusively in the country where they are committed”). has been stretched to the breaking by new forms of litigation—including extraterritorial prosecutions. But we are still in the interlude. Numerous important criminal cases have been percolating in the lower courts. 14 See, e.g., United States v. Beyle, 782 F.3d 159, 162 (4th Cir. 2015), cert. denied, 136 S. Ct. 179 (2015) (affirming conviction of defendant who committed piracy on high seas, “beyond the territorial sea of any nation”); United States v. Fawwaz, No. S7 98-cr-1023 (LAK), 2015 WL 2114914 (S.D.N.Y. May 6, 2015), appeal docketed (2d Cir. 2015) (reflecting conviction of defendant in major extraterritorial terrorism prosecution). But the Supreme Court has not yet had its International Shoe moment; it has not yet explained what due process does or does not permit. This Article endeavors to supply an answer.
I. The Constitutional Clean Slate
A. The Problem of Criminal Jurisdiction
Judicial jurisdiction is the power of a sovereign’s court to hear a case. 15 See Restatement (Third) of Foreign Relations Law of the United States § 401(b) (Am. Law Inst. 1987) (defining “jurisdiction to adjudicate”). Note that throughout this Article, “sovereign” is used as an inclusive catchall, to refer to a state (like New York) or a country (like the United States or France). Legislative jurisdiction, in contrast to judicial jurisdiction, is the power of a sovereign to prescribe substantive rules; tort law, for example, or criminal law. Id. § 401(a). In another article, this author considers due process limits on criminal legislative jurisdiction. See Michael Farbiarz, Extraterritorial Criminal Jurisdiction, 114 Mich. L. Rev. 507 passim (2016). This author’s views on the relationship between due process limits on criminal legislative jurisdiction and due process limits on criminal judicial jurisdiction are summarized below, in the conclusion. Generally speaking, it has two components: first, the power to adjudicate a given class of claims (subject matter jurisdiction); and second, the power to adjudicate a given claim against a given defendant (personal jurisdiction). 16 See, e.g., Ins. Corp. of Ir. v. Compagnie des Bauxites de Guinee, 456 U.S. 694, 701–03 (1982) (describing subject matter and personal jurisdiction). As to subject matter jurisdiction, only federal courts can hear cases that allege a violation of federal criminal law. 17 18 U.S.C. § 3231 (2012) (describing U.S. district courts’ “exclusive” jurisdiction over “all offenses against the laws of the United States”). And as to personal jurisdiction, one core principle is equally straightforward: Criminal jurisdiction can only be exercised over a defendant who is physically present before the court. 18 Fed. R. Crim. P. 43(a)(2) (“[D]efendant must be present at . . . every trial stage . . . .”); accord, e.g., Lewis v. United States, 146 U.S. 370, 372 (1892) (“A leading principle that pervades the entire law of criminal procedure is that, after indictment found, nothing shall be done in the absence of the prisoner.”). Because of this physical presence rule, there are no default judgments in criminal cases, or trials in absentia—features of our legal culture so basic that we take them for granted. 19 See generally James G. Starkey, Trial in Absentia, 53 St. John’s L. Rev. 721, 721–33 (1979) (describing long history of Anglo-American aversion to trials in absentia). And the physical presence rule seems not just natural, but necessary. The Sixth Amendment gives criminal defendants the right to “confront” witnesses. 20 U.S. Const. amend. VI. And that has generally been taken to mean the defendant must be physically present in the courtroom, there to watch the witness testimony. 21 See, e.g., Illinois v. Allen, 397 U.S. 337, 338 (1970) (“One of the most basic of the rights guaranteed by the Confrontation Clause is the accused’s right to be present in the courtroom at every stage of his trial.”). It bears noting that the Supreme Court has held that it may sometimes be constitutionally permissible for a remote witness to testify by video—the witness’s testimony electronically piped into the courtroom, while the defendant (and judge and jury) watch the testimony from inside the courtroom. Maryland v. Craig, 497 U.S. 836, 841–42, 857 (1990). This might be taken to imply that roles can be reversed, that the Confrontation Clause could potentially allow a defendant to participate in his or her trial remotely—by live two-way video from outside of the courtroom, with the witnesses (and the judge and jury) physically present in the courtroom. If that were the rule, Guantánamo Bay detainees could potentially be given Article III trials in federal court in New York or Washington without their ever having physically to enter the United States. But all of this is abstract for now. Whatever else might be said about the constitutionality of this approach, the current version of Rule 43 forbids it. See supra note 18 (quoting Rule 43).
But if physical presence is a constitutionally necessary ingredient of criminal personal jurisdiction, is it also constitutionally sufficient? For most of Anglo-American legal history, the question hardly needed asking. Criminal cases were generally regarded as “altogether local, and cognizable and punishable exclusively in the country where they are committed.” 22 Story, Conflict of Laws, supra note 13, at 840. On this understanding, the question of where a prosecution could go forward (“exclusively” where the crime was committed) all but resolved the question of whether the prosecution could go forward in that place (because sovereigns can very clearly punish crimes committed within their borders). If a bank was robbed in Illinois, Illinois (and Illinois alone) could prosecute the case. And all would agree that it should be allowed to do so. 23 See, e.g., Mayor of N.Y. v. Miln, 36 U.S. (11 Pet.) 102, 139 (1837) (“[A] state has . . . undeniable . . . jurisdiction over all persons and things within its territorial limits.”).
But with today’s extraterritorial prosecutions, we are at the far end of the road. The United States prosecutes Venezuelan soccer officials for conduct in Europe 24 See Matt Apuzzo, Stephanie Clifford & William K. Rashbaum, FIFA Officials Arrested on Corruption Charges; Blatter Isn’t Among Them, N.Y. Times (May 26, 2015), http://www.nytimes.com/2015/05/27/sports/soccer/fifa-officials-face-corruption-charges-in-us.html (on file with the Columbia Law Review) (discussing arrests and extradition). and Algerian terrorists for murders in North Africa. 25 See Amended Complaint at 3, United States v. Belmokhtar, No. 13 MAG 522 (S.D.N.Y. July 19, 2013) (describing factual basis of charges). And it prosecutes Colombian drug barons 26 See Farbiarz, supra note 15, at 512–13 (describing prosecutions). and European arms traffickers 27 See, e.g., United States v. Bout, 731 F.3d 233, 236–37 (2d Cir. 2013) (describing prosecution of Russian arms trafficker); United States v. Al Kassar, 660 F.3d 108, 115 (2d Cir. 2011) (describing prosecution of Spanish arms trafficker); United States v. Viglakis, No. 12 CR 585 (KBF), 2013 WL 4477023, at *1 (S.D.N.Y. Aug. 14, 2013) (describing prosecution of Greek arms trafficker). —who had never set foot on American soil until they were extradited here. These are among the most high-stakes prosecutions of the modern era. Some involve hundreds of murders. 28 E.g., United States v. Ghailani, 733 F.3d 29, 36 (2d Cir. 2013) (describing prosecution of defendant implicated in al Qaeda bombings that killed over 200 people); In re Terrorist Bombings of U.S. Embassies in East Africa, 552 F.3d 93, 102 (2d Cir. 2008) (same). And some function as important tools of American statecraft. 29 See Farbiarz, supra note 15, at 512–13 (describing ways in which national security prosecutions are used to advance U.S. foreign policy goals). But all of these extraterritorial prosecutions are controversial. In virtually every one, there is a potentially available—and undoubtedly legitimate—alternative to a U.S. prosecution. The FIFA soccer case, after all, could have gone forward in Switzerland or France. The Algerian terrorist could have been prosecuted in Algeria.
Because of all this, federal extraterritorial prosecutions open up a deep problem. 30 This Article does not consider prosecutions of high-seas piracy. Such prosecutions are rare, and may very well be sui generis. See W.E. Beckett, The Exercise of Criminal Jurisdiction over Foreigners, 6 Brit. Y.B. Int’l L. 44, 45 (1925) (“Piracy stands on such an exceptional basis that it throws no light on the question of penal jurisdiction generally.”). This Article also gives no consideration to prosecutions of U.S. citizens for their conduct abroad. This is because virtually all important contemporary extraterritorial prosecutions are of non-U.S. citizens and, more fundamentally, because prosecutions of U.S. citizens are distinct, from a jurisdictional perspective, from prosecutions of non-U.S. citizens. See, e.g., Skiriotes v. Florida, 313 U.S. 69, 73 (1941) (describing “duty of the citizen in relation to his own government” even when abroad). On the one hand, they raise difficult questions about jurisdiction. On the other hand, in thinking through these questions, the old answers will not do. This is because the necessary premise of modern extraterritorial prosecutions (that a criminal violation can be prosecuted in multiple places) is wholly at odds with the traditional ideal of “local” criminal jurisdiction (that prosecutions can go forward “exclusively” in one place, where the crime was committed). 31 The basic tension described in the text between the classic approach to criminal jurisdiction and extraterritorial prosecutions may suggest to some that, as a class, extraterritorial prosecutions are unlawful. But that argument is not viable in light of, among other things, the various constitutional provisions that specifically contemplate extraterritorial prosecutions. Farbiarz, supra note 15, at 515 n.40.
B. Article III
If old answers will not do, what of modern ones? Contemporary civil litigation frequently spins off complicated cross-border issues, and those issues are managed by due process limits on personal jurisdiction. 32 See generally Ins. Corp. of Ir. v. Compagnie des Bauxites de Guinee, 456 U.S. 694, 702 n.10 (1982) (stating due process is “only” part of Constitution that limits personal jurisdiction of state courts). In state court, the Due Process Clause of the Fourteenth Amendment sets the outer limits of personal jurisdiction, and the focus of the inquiry is on connections between the defendant and the state where the court sits. See, e.g., Int’l Shoe Co. v. Washington, 326 U.S. 310, 319 (1945) (noting corporate litigant conducted business in state). Putting aside diversity jurisdiction, in federal court, the outer limits of personal jurisdiction are set by the Due Process Clause of the Fifth Amendment, and the focus is on connections between the defendant and the United States as a whole. See generally 4 Charles Alan Wright, Arthur R. Miller & Adam N. Steinman, Federal Practice and Procedure § 1068.1, at 691–98 (4th ed. 2015) (describing Fifth Amendment minimum contacts jurisprudence); see also J. McIntyre Mach., Ltd. v. Nicastro, 131 S. Ct. 2780, 2789 (2011) (plurality opinion) (explaining “personal jurisdiction requires a forum-by-forum . . . analysis” and suggesting United States, as “distinct sovereign,” is one such forum). The due process clauses of the Fifth and Fourteenth Amendments use the same operative language, and courts applying the Fifth Amendment routinely apply the Fourteenth Amendment’s more developed jurisprudence to determine what constitutes a sufficient connection to the United States. See Wright, Miller & Steinman, supra, 701 n.31 (collecting cases). Due process curbs on jurisdiction in civil cases, it would seem, might be applied in criminal cases as well.
But Article III of the Constitution guarantees, “The trial of all Crimes . . . shall be held in the State where the said Crimes shall have been committed.” 33 U.S. Const. art. III, § 2, cl. 3. Article III’s venue provision is essentially interchangeable with the Sixth Amendment’s vicinage provision. See 2 Charles Alan Wright & Peter J. Henning, Federal Practice and Procedure § 301, at 324 (4th ed. 2009) (noting “technical distinction” between these provisions “has been of no real importance”); cf. United States v. Cabrales, 524 U.S. 1, 6 (1998) (treating provisions together). And this venue provision, it is said, “may render [due process] minimum contacts analysis superfluous,” at least with respect to the federal criminal proceedings that Article III controls. 34 Lea Brilmayer, An Introduction to Jurisdiction in the American Federal System 331 (1986) [hereinafter Brilmayer, Introduction to Jurisdiction]. “Minimum contacts” of course refers to the standard set out in Int’l Shoe Co., 326 U.S. at 319.
This, though, is not persuasive. Article III’s venue requirement can make due process “superfluous” only if the venue standard is as high as (or higher than) the due process “minimum contacts” standard. But the opposite is the case. Criminal venue, for example, is permissible in a given location if it was merely foreseeable to the defendant that a relevant act would occur there. 35 See, e.g., United States v. Rowe, 414 F.3d 271, 279 (2d Cir. 2005) (rejecting challenge to New York child pornography conviction of Kentucky defendant; venue was based on advertisement in internet chat room and defendant “must have known or contemplated that the advertisement would be transmitted by computer to anyone the whole world over who . . . entered the chat room” (citation omitted)). But on a due process “minimum contacts analysis,” jurisdiction would typically be impermissible in precisely that situation. 36 See, e.g., World-Wide Volkswagen Corp. v. Woodson, 444 U.S. 286, 295–96 (1980) (“‘[F]oreseeability’ alone has never been a sufficient benchmark for personal jurisdiction under the Due Process Clause.”). Criminal venue can be sustained based on the actions of the defendant—but also on the actions of others. 37 See, e.g., United States v. Gonzalez, 683 F.3d 1221, 1227 (9th Cir. 2012) (affirming venue based on government informant’s location when defendant spoke to informant over telephone); United States v. Cordero, 668 F.2d 32, 43–44 (1st Cir. 1981) (Breyer, J.) (affirming venue based on undercover agent’s location when defendant spoke to agent over telephone); see also Hyde v. United States, 225 U.S. 347, 362–63 (1912) (upholding venue on basis of coconspirator’s actions); United States v. Royer, 549 F.3d 886, 896 (2d Cir. 2008) (upholding venue on basis of actions coconspirator caused another person to take). Personal jurisdiction, by contrast, is generally based only on the actions of the defendant. 38 See Walden v. Fiore, 134 S. Ct. 1115, 1122–25 (2014) (holding personal jurisdiction “must arise out of contacts that the defendant himself creates with the forum State” (internal quotation marks omitted)) (quoting Burger King Corp. v. Rudzewicz, 471 U.S. 462, 475 (1985)); Helicopteros Nacionales de Colom., S.A. v. Hall, 466 U.S. 408, 417 (1984) (“[U]nilateral activity of another party or a third person is not an appropriate consideration when determining whether a defendant has sufficient contacts . . . to justify . . . jurisdiction.”); World-Wide Volkswagen, 444 U.S. at 298 (noting actions of third parties “‘who claim some relationship with a nonresident defendant’” cannot form basis for jurisdiction over defendant (quoting Hanson v. Denckla, 357 U.S. 235, 253 (1958))). Criminal venue can be maintained where an item simply passed through a forum on its way to somewhere else. 39 See, e.g., United States v. Shearer, 794 F.2d 1545, 1551 (11th Cir. 1986) (concluding mere movement of items through area is sufficient to establish venue); United States v. Mayo, 721 F.2d 1084, 1090–91 (7th Cir. 1983) (same); see also Armour Packing Co. v. United States, 209 U.S. 56, 63 (1908) (“The court had jurisdiction of the alleged crime for the reason that the transportation was conducted through the district . . . .”). Under due process, though, jurisdiction is not so easily established. 40 See J. McIntyre Mach., Ltd. v. Nicastro, 131 S. Ct. 2780, 2792 (2011) (Breyer, J., concurring in the judgment) (describing case law). Examples can be multiplied. But the point remains the same—given the comparatively low standard it sets, venue law cannot be said to render “minimum contacts analysis superfluous.” 41 Brilmayer, Introduction to Jurisdiction, supra note 34, at 331.
And the point is stronger yet with respect to extraterritorial prosecutions. As to crimes “not committed within any State,” Article III indicates that “the Trial shall be at such Place or Places as the Congress may by Law have directed.” 42 U.S. Const. art. III, § 2, cl. 3. Based on that provision, the Crimes Act of 1790 stipulated that “the trial of crimes committed . . . out of the jurisdiction of any particular state, shall be in the district where the offender is apprehended, or into which he may first be brought.” 43 Act of Apr. 30, 1790, ch. 9, § 8, 1 Stat. 112, 114. That statute, with small modifications, remains on the books. 44 18 U.S.C. § 3238 (2012). It allows the United States to arrest a defendant overseas for an extraterritorial offense; fly him or her into the country, in custody, on a government plane; and then take venue and initiate the proceedings wherever prosecutors want the plane to land. 45 See, e.g., United States v. Yousef, 327 F.3d 56, 114–15 (2d Cir. 2003) (construing relevant venue provision); United States v. Erdos, 474 F.2d 157, 160 (4th Cir. 1973) (same). This is done with frequency. 46 See, e.g., Yousef, 327 F.3d at 114–15 (describing foreign arrest and flight, in custody, to New York for trial); United States v. Abu Ghayth, 945 F. Supp. 2d 511, 512 (S.D.N.Y. 2013) (same). And it is a barely-there check on where a criminal trial might go forward—too faint, by far, to “render [due process] minimum contacts analysis superfluous.” 47 Brilmayer, Introduction to Jurisdiction, supra note 34, at 331. Two points bear brief mention. First, the constitutionality of the referenced venue statute is not in doubt. It is a direct response to Article III’s invitation for congressional legislation. U.S. Const. art. III, § 2, cl. 3. And the fact that the First Congress originally passed the venue statute is a large thumb on the scale in its favor. See Eldred v. Ashcroft, 537 U.S. 186, 213 (2003) (giving significant weight to early congressional action); Myers v. United States, 272 U.S. 52, 175 (1926) (same). See generally Michael Bhargava, The First Congress Canon and the Supreme Court’s Use of History, 94 Calif. L. Rev. 1745, 1748–62 (2006) (“The Supreme Court . . . continues to invoke the actions of the First Congress almost as the received word of the Founders.”). Second, it is often said that open-ended constitutional commands cannot impose added restrictions when more specific, on-point constitutional standards already occupy the field; due process, for example, does not set the standard for police seizures because the Fourth Amendment already does. Graham v. Connor, 490 U.S. 386, 396–97 (1989) (“Because the Fourth Amendment provides an explicit textual source of constitutional protection . . . that Amendment, not the more generalized notion of ‘substantive due process’ must be the guide . . . .”). But the fact that Article III is fairly specific with respect to venue does not preclude due process from having a role with respect to jurisdiction, because venue and jurisdiction are distinct. In federal criminal cases, there is all the difference between venue (can the United States prosecute in the Southern District of California or in the Northern?) and jurisdiction (can the United States prosecute at all?).
C. Ker-Frisbie
Another theory given to explain why due process cannot limit personal jurisdiction in criminal cases is grounded in the Ker–Frisbie doctrine. The Ker–Frisbie doctrine is named for two Supreme Court cases, each of which holds that jurisdiction need not be relinquished when the criminal defendant is physically present before the court solely because he or she was kidnapped and brought there. 48 Frisbie v. Collins, 342 U.S. 519, 522 (1952) (rejecting challenge to state criminal jurisdiction where defendant was kidnapped into state); Ker v. Illinois, 119 U.S. 436, 441 (1886) (same). The Ker–Frisbie doctrine is commonly taken to mean that physical presence is the last word, such that once a criminal defendant has appeared before the court, nothing more can be said about personal jurisdiction. 49 This author served as a federal national security prosecutor for more than a decade, and knows about the conventional understanding of Ker-Frisbie from extensive conversations with senior criminal defense lawyers. Because the doctrine acts to dissuade defense lawyers from making certain arguments, the conventional understanding of Ker-Frisbie rarely finds its way into the case reports. But there are occasional exceptions. See, e.g., In re Vasquez, 705 N.E.2d 606, 609 (Mass. 1999) (noting criminal defendant pressed jurisdictional argument before arriving in state where he was to be tried based on idea that, once he arrived, jurisdictional challenge would become unavailable). But this misunderstands the doctrine.
In civil cases, there has long been a “force-or-fraud” rule as a matter of state law. 50 See Ex parte Johnson, 167 U.S. 120, 126 (1897) (collecting cases). This rule prevents jurisdiction from being exercised based on the defendant’s physical presence if the defendant was tricked into being before the court or was kidnapped. 51 See id. (“[I]f a defendant in a civil case be brought within the process of the court by a trick or device, the service will be set aside, and he will be discharged from custody.”). But criminal cases have always been different. At common law, the fact that kidnapping secured a criminal defendant’s physical presence was no bar to jurisdiction. 52 The canonical early American cases establishing this proposition are State v. Brewster, 7 Vt. 118, 121 (1835) and State v. Smith, 17 S.C.L. (1 Bail.) 283, 290 (S.C. 1829). See also Ex parte Scott (1829) 109 Eng. Rep. 166, 166; 9 B. & C. 445, 445 (“[W]hen a party is liable to be detained on a criminal charge, the Court will not inquire into the manner in which the caption was effected . . . .”). And the Ker–Frisbie doctrine simply establishes that due process does not compel the creation of a criminal force-or-fraud doctrine analogous to the already existing civil force-or-fraud rule. As the Supreme Court has put it:
[I]f a defendant in a civil case be brought within the process of the court by a trick or device, the service will be set aside . . . . The law will not permit a person to be kidnapped or decoyed within the jurisdiction for the purpose of being compelled to answer to a mere private [civil] claim, but in criminal cases the interests of the public override . . . . 53 Johnson, 167 U.S. at 126 (emphasis omitted).
In short, both the criminal Ker–Frisbie doctrine and the civil force-or-fraud rule purport to answer the same question, though in different ways—whether jurisdiction can be undone based on how the defendant came to be before the court.
This “how” question is a narrow one, and there is no reason to think that it crowds out all other inquiries related to personal jurisdiction. Take, for example, the Supreme Court’s decision in Burnham v. Superior Court, 54 495 U.S. 604 (1990). a civil action in which personal jurisdiction was established, at least initially, by the defendant’s physical presence in California. 55 Id. at 608 (plurality opinion). There were no issues with respect to how the defendant had come to be there—no allegations, for example, of abduction across state lines. But none of the Justices suggested that ended the inquiry. The fact that the Burnham defendant had no viable “how” claim (because he was not kidnapped into California) did not purport to foreclose broader inquiries concerning whether jurisdiction was proper. 56 These broader inquiries turned, for some justices in Burnham, on questions of tradition and, for others, on questions of fairness. Compare id. at 608–28 (plurality opinion) (focusing on “continuing traditions of our legal system that define the due process standard”), with id. at 628–40 (Brennan, J., concurring in the judgment) (maintaining Court should “undertake an ‘independent inquiry into . . . fairness’” (citation omitted) (quoting id. at 621 (plurality opinion))). So too in criminal cases. The fact that a criminal defendant can have no viable “how” claim (because Ker–Frisbie takes it off the table) does not purport to occupy the whole field, nor to imply that there is no possible room left to ask other questions about the legitimacy of jurisdiction. 57 This analysis is not changed by a dictum in the Frisbie case, which indicates that “due process of law is satisfied when one present in court is convicted of crime after having been fairly apprized of the charges against him and after a fair trial in accordance with constitutional procedural safeguards.” Frisbie v. Collins, 342 U.S. 519, 522 (1952). That broad statement may have made sense in the context of Frisbie itself, where the crime was committed in the state where the court sat—such that even if due process required a “minimum contacts”-type jurisdictional test, that test would necessarily have been satisfied. See, e.g., Mayor of N.Y. v. Miln, 36 U.S. (11 Pet.) 102, 139 (1837) (“[A] state has . . . undeniable . . . jurisdiction over all persons and things, within its territorial limits . . . .”). Moreover, the Frisbie dictum can hardly be taken for all it is worth. If it were, it would suggest that whole areas of modern constitutional law are invalid—due process limits on criminal legislative jurisdiction, for example. See Farbiarz, supra note 15, at 516–17 (describing “extraterritorial due process doctrine”).
II. Due Process
A. For Due Process
It has been argued that due process need not limit personal jurisdiction in federal criminal cases (citing Article III) and cannot limit personal jurisdiction in any criminal cases (citing Ker–Frisbie). But these arguments are not persuasive, 58 See supra sections I.B–C (challenging both theories). and without them the slate is clean. This section begins to fill it, arguing that due process should be affirmatively understood to check personal jurisdiction in all criminal cases, including federal extraterritorial prosecutions.
Due process, of course, limits jurisdiction in civil cases. It is not hard to see why. When a sovereign resolves a dispute through its courts, it is bringing to bear its coercive power, and due process generally limits exercises of sovereign coercive power. A “court’s assertion of jurisdiction exposes defendants to the State’s coercive power, and is therefore subject to review for compatibility with the . . . Due Process Clause.” 59 Goodyear Dunlop Tires Operations, S.A. v. Brown, 131 S. Ct. 2846, 2850 (2011). On this unsentimental account, power is power; due process generally checks legislative and executive power, and judicial power must be so checked too:
The Due Process Clause protects an individual’s right to be deprived of life, liberty, or property only by the exercise of lawful power. This is no less true with respect to the power of a sovereign to resolve disputes through judicial process than with respect to the power of a sovereign to prescribe rules of conduct for those within its sphere. 60 J. McIntyre Mach., Ltd. v. Nicastro, 131 S. Ct. 2780, 2786–87 (2011) (plurality opinion) (citations omitted); accord, e.g., Missouri v. Dockery, 191 U.S. 165, 171 (1903) (noting appellee’s “rights under [Fourteenth] [A]mendment turn on the power of the State, no matter by what organ it acts”); Murray’s Lessee v. Hoboken Land & Improvement Co., 59 U.S. (18 How.) 272, 276 (1856) (holding due process constrains judicial power as well as legislative); cf. Shelley v. Kraemer, 334 U.S. 1, 14–18 (1948) (affirming “action of state courts and judicial officers in their official capacities is to be regarded as action of the State within the meaning of the Fourteenth Amendment”).
Due process, in sum, usually limits the reach of the sovereign’s coercive power. And, the Supreme Court has suggested, this general principle applies to the particular sort of coercive power that the sovereign brings to bear when it employs the “judicial process” 61 Nicastro, 131 S. Ct. at 2786–87 (plurality opinion). —when it empowers an official (the judge) to enter orders and to enforce the parties’ subpoenas, to interpret substantive law and to apply rules of evidence, and generally to provide a neutral forum for “resolv[ing] disputes.” 62 d. At 2787.
For three reasons, this rationale for imposing due process limits applies to criminal cases no less than to civil ones.
First, “[e]very trial involves the exercise of judicial power,” 63 Ex parte Milligan, 71 U.S. (4 Wall.) 2, 121 (1866). whether the trial sounds in civil law or criminal. The jury may be instructed on causation, a witness may be required to testify, or a stay may be entered pending appeal—and these familiar features of the judicial process are as much an example of a sovereign exercising coercive power in one class of cases (civil cases) as in another (criminal ones). Courts determine whether a document is privileged, whether venue lies, or whether a certain set of facts constitutes fraud as a matter of law. In civil cases, making such determinations is a routine aspect of “the power of a sovereign to resolve disputes through judicial process” 64 Nicastro, 131 S. Ct. at 2786–87 (plurality opinion). —and in criminal cases, too.
Second, while the sovereign’s “coercive power” 65 Goodyear Dunlop Tires Operations, S.A. v. Brown, 131 S. Ct. 2846, 2850 (2011). is brought to bear in both criminal and civil cases, that power is usually brought to bear much more directly in a criminal case than in a civil one. In both civil and criminal cases, the sovereign establishes a neutral forum for the resolution of disputes. But in criminal cases, those dispute resolution proceedings are saturated with pointed exercises of the sovereign’s coercive power. The prosecutor is present in every criminal case, directly deploying the sovereign’s coercive power. And in criminal cases, the judge all-but necessarily goes well beyond any hypothesized neutral or “umpire” role. In a typical tort or contract case, for example, it is the jury that generally fixes the remedy (damages) after it has found the defendant liable. In a criminal case, by contrast, it is the judge him or herself who decides on the remedy—personally selecting and imposing the sentence after the defendant has been found liable (that is, after the defendant has been convicted). As the Supreme Court has put it, “Unlike a criminal prosecution, in which the coercive power of the state is immediately brought to bear, the state’s involvement in . . . a private civil suit is minimal. The state’s role is simply to provide a forum for the resolution of a private dispute.” 66 Van Cauwenberghe v. Biard, 486 U.S. 517, 525 (1988). If due process limits personal jurisdiction even in run-of-the-mill civil cases (where “coercive power” is “involve[d]” in a relatively indirect way, as a matter of “simply . . . provid[ing] a forum”), then due process should limit personal jurisdiction in criminal cases (where the sovereign’s “coercive power” is brought to bear in a plainly more direct and “immediate[]” way). 67 The directness (or indirectness) of the exercise of sovereign coercive power that is emphasized in the text is of course no random variable. It is, rather, a central question of constitutional law. For example, when sovereign power is brought to bear in a merely diffuse or indirect way, constitutional protections may not apply in the first place. See, e.g., O’Bannon v. Town Court Nursing Ctr., 447 U.S. 773, 789 (1980) (“[T]he due process provision of the Fifth Amendment does not apply to the indirect adverse effects of governmental action.”); The Legal Tender Cases, 79 U.S. (12 Wall.) 457, 551 (1870) (“[The Fifth Amendment] has always been understood as referring only to a direct appropriation, and not to consequential injuries resulting from the exercise of lawful power.”); cf. Erwin Chemerinksy, Constitutional Law 77—92 (3d ed. 2001) (noting “taxpayer standing” is generally unavailable because potential plaintiffs have only “generalized grievances”).
Third and finally, while the sovereign’s “coercive power” 68 Goodyear Dunlop Tires Operations, S.A., 131 S. Ct. at 2850. is brought to bear in both criminal and civil cases, the sovereign’s power is usually brought to bear much more forcefully in a criminal case than in a civil one. An adverse civil judgment typically means that the defendant must make a payment or stop (or start) doing something. But a criminal judgment often means that the defendant goes to prison. Except in rare cases, the coercion implicit in spending a year or two or ten in prison utterly dwarfs the coercion implicit in satisfying a tort judgment. And prison is not the end of it. After the defendant has finished his or her sentence, “collateral consequences” are heaped on a criminal defendant—the defendant may not be able to vote or live in public housing; may be subject to routine searches by a probation officer; or may be deported or forced to register as a sex offender. 69 See Margaret Colgate Love et al., Collateral Consequences of Criminal Convictions: Law, Policy and Practice §§ 2:1–2:77 (2013) (providing overview of collateral consequences). After a civil defendant has paid the plaintiff, that is typically the road’s end. But “[e]ven when [a criminal] sentence has been completely served, the fact that a man has been convicted of a felony pursues him like Nemesis.” 70 Nat’l Council on Crime & Delinquency, Annulment of a Conviction of Crime: A Model Act 3 (1962). And this is to say nothing of the period before judgment is entered. Preliminary proceedings in a civil case may sometimes lead to, say, the freezing of assets. 71 See, e.g., Grupo Mexicano de Desarrollo, S.A. v. All. Bond Fund, Inc., 527 U.S. 308, 325 (1999) (discussing prejudgment equitable relief). In a criminal case, preliminary proceedings can lead to assets being frozen, too 72 See, e.g., 21 U.S.C. § 853©(1) (2012) (granting judges authority to “preserve the availability of property” in preliminary proceedings). —but also to much more severe consequences, including the defendant being jailed pending trial. 73 18 U.S.C. § 3142© (2012) (authorizing pretrial detention). The movements of a criminal defendant who is not detained pending trial will generally be limited. See, e.g., id. § 3142©(B)(iv) (authorizing travel restrictions); id. § 3142(c)(B)(vii) (authorizing imposition of curfew). More fundamentally, in a typical civil case, the coercive power that is brought to bear is judicial power—as when the court enters judgment. 74 Cf. Stern v. Marshall, 131 S. Ct. 2594, 2615 (2011) (describing “entry of a final, binding judgment by a court” as “most prototypical exercise of judicial power” (emphasis omitted)). But a sovereign can only prosecute violations of its own criminal laws in its courts. 75 Restatement (Second) of Conflict of Laws § 89 cmt. E (Am. Law Inst. 1971); accord, e.g., The Antelope, 23 U.S. (10 Wheat.) 66, 123 (1825) (“The Courts of no country execute the penal laws of another . . . .”); cf. Martin v. Hunter’s Lessee, 14 U.S. (1 Wheat.) 304, 337 (1816) (“No part of the criminal jurisdiction of the United States can, consistently with the constitution, be delegated to state tribunals.”). Accordingly, when it takes jurisdiction in a criminal case, a court is not only opening the door, as in a civil case, to the exercise of judicial power. Rather, the court is also opening the door—indeed, it is opening the only possible door—to the exercise of part of the sovereign’s executive power. And not just any part—perhaps the most coercive peacetime part of the executive power, the power to prosecute a defendant, and ultimately to take away a defendant’s liberty and, in some cases, life. 76 See generally United States v. Nixon, 418 U.S. 683, 693 (1974) (noting Executive has “exclusive authority and absolute discretion to decide whether to prosecute a case”). There are of course civil actions that involve very forceful uses of sovereign coercive power. E.g., Kansas v. Hendricks, 521 U.S. 346, 356–59 (1997) (describing civil detention). The point, though, is that due process checks personal jurisdiction in even the most garden-variety of civil actions—relatively low-stakes contract disputes, for example, that implicate sovereign coercive power in a way that is plainly less forceful than criminal prosecutions.
Phillips Petroleum Co. v. Shutts 77 472 U.S. 797 (1985). buttresses the logic of this inference. There, the Supreme Court held that due process, which had long been understood to protect civil defendants from exorbitant exercises of personal jurisdiction, also protects some plaintiffs—namely, absent plaintiffs in class action suits. 78 Id. At 810–11. But, Phillips Petroleum held, a lower due process standard is permissible as to class action plaintiffs—in part because “[t]hey are almost never subject to counterclaims or cross-claims, or liability for fees or costs,” while an ordinary civil defendant “is faced with the full powers of the forum State to render judgment against it.” 79 Id. At 808, 810 (emphasis omitted). The more forceful the potential claims (“full powers . . . to render judgment against it”), the more work due process must do (as in the case of ordinary civil defendants). The less forceful the potential claims (no “counterclaims or cross-claims, or liability for fees or costs”), the less work due process must do (as in the case of absent class action plaintiffs). Due process limits on personal jurisdiction are, in short, roughly responsive to the power of the claims that might be brought against the party challenging jurisdiction. And that same logic undergirds the argument developed above, that if less forceful claims justify due process protections, as in a garden-variety civil action, then more forceful claims must do so as well, as in a criminal prosecution.
In sum, if due process limits personal jurisdiction in civil cases because adjudication is an incident of sovereign coercive power, then due process should be understood to limit personal jurisdiction in criminal cases—where sovereign coercive power is also brought to bear, and in more direct and more forceful ways.
History bolsters this conclusion. The Supreme Court has said that, in civil actions, due process limits on jurisdiction flow from the English rule of coram non judice:
The proposition that the judgment of a court lacking jurisdiction is void traces back to the English Year Books . . . . Traditionally that proposition was embodied in the phrase coram non judice, “before a person not a judge”—meaning, in effect, that the proceeding in question was not a judicial proceeding because lawful judicial authority was not present, and could therefore not yield a judgment. American courts invalidated . . . judgments that violated this common-law principle long before the Fourteenth Amendment was adopted . . . . In Pennoyer v. Neff we announced that the judgment of a court lacking personal jurisdiction violated the Due Process Clause of the Fourteenth Amendment as well. 80 Burnham v. Superior Court, 495 U.S. 604, 608–09 (1990) (plurality opinion) (emphasis omitted) (citations omitted).
But if coram non judice was in some sense constitutionalized in Pennoyer, then it is telling that coram non judice was routinely invoked in criminal cases during both the immediate pre- and post-Pennoyer periods. 81 E.g., Pennsylvania v. Franklin, 4 U.S. (4 Dall.) 316, 316 (1804) (holding proceedings in criminal case are coram non judice and thus void); Brumley v. State, 20 Ark. 77, 78 (1859) (same); Rector v. State, 6 Ark. 187, 190 (1845) (same); Dunn v. State, 2 Ark. 229, 258–59 (1840) (same); State v. Odell, 4 Blackf. 156, 156 (Ind. 1836) (same); State v. Briton, 2 Penning. 506, 506 (N.J. 1812) (same); Hester v. Commonwealth, 85 Pa. 139, 155 (1877) (same); see also, e.g., State v. Simone, 149 La. 287, 289 (1921) (reasoning in criminal case if proceeding is coram non judice, court is without authority to go forward); Commonwealth v. Bugbee, 70 Mass. (4 Gray) 206, 207 (1855) (same); State v. McClain, 156 Mo. 99, 101 (1900) (same); In re Patzwald, 5 Okla. 789, 794 (1897) (same). In 1845, for example, California’s Supreme Court cited coram non judice to vacate a murder conviction because the impact of the defendant’s conduct in one county did not permit him to be tried in another. 82 People v. Hodges, 27 Cal. 340, 341–42 (1865). And in 1887, Missouri’s Supreme Court referenced coram non judice to vacate an embezzlement conviction because the defendant’s actions did not allow for trial in St. Louis, where no relevant acts had been taken. 83 State v. Hatch, 91 Mo. 568, 570 (1887). Coram non judice’s use in criminal cases is one more reason to conclude that due process limits personal jurisdiction in both civil and criminal cases.
The Constitution’s text points in the same direction. There are some constitutional obligations that apply just to criminal cases, 84 See, e.g., U.S. Const. amend. V (requiring grand jury in trial for any defendant “held to answer for a capital, or otherwise infamous crime”); id. Amend. VI (extending right to speedy trial to “all criminal prosecutions”). and some that focus on civil ones. 85 See id. Amend. VII (guaranteeing right to jury trial in “[s]uits at common law”). But due process is not either/or. The Constitution requires due process in both civil cases (in which it is “property” that is most typically at stake) and criminal cases (in which “liberty” is usually at stake, and sometimes “life”). 86 Id. Amend. V. And the Supreme Court has never hesitated to apply due process to both civil and criminal cases. 87 See, e.g., Kansas v. Hendricks, 521 U.S. 346, 356–60 (1997) (invoking due process in connection with civil detention); Brady v. Maryland, 373 U.S. 83, 87 (1963) (invoking due process in connection with criminal prosecution). Note that the most direct antecedents of the Fifth Amendment’s Due Process Clause were the Magna Carta and an English statute from 1354. Pac. Mut. Life Ins. Co. v. Haslip, 499 U.S. 1, 28–29 (1991) (Scalia, J., concurring in the judgment) (canvassing history). Each spoke to what are thought of today as criminal matters. The 1354 statute, for example, stipulated that “no Man . . . shall be put . . . to Death, without . . . due Process of Law.” (1354) 28 Edw. III c. 3 (Eng.). During the nineteenth century, it was taken for granted by leading commentators that due process applied in criminal cases—indeed, it was sometimes suggested that due process might apply only in criminal cases. See 2 James Kent, Commentaries on American Law 12–16 (John M. Gould ed., 14th ed. 1896); 2 Joseph Story, Commentaries on the Constitution of the United States §§ 1940–1950 (Melville M. Bigelow ed., 5th ed. 1891). There is no reason to think that due process might limit jurisdiction in one of the two classes of cases alluded to in the Constitution (civil cases)—but not in the other (criminal cases).
And this conclusion is only strengthened by the basic structure of modern due process jurisprudence. Due process limits legislative jurisdiction in both civil cases 88 See Allstate Ins. Co. v. Hague, 449 U.S. 302, 307–20 (1981) (plurality opinion) (considering whether state court’s “choice of its own substantive law . . . exceeded federal constitutional limitations”). and criminal cases. 89 See Farbiarz, supra note 15, at 516–17 (describing case law). Why, then, should due process not limit personal jurisdiction in both civil cases 90 See Int’l Shoe Co. v. Washington, 326 U.S. 310, 316–19 (1945) (establishing “minimum contacts” inquiry). and criminal cases? 91 Due process limits the exercise of personal jurisdiction in all civil cases—including those cases in which the Constitution requires heightened procedural or substantive protections for the defendant in light of the gravity of the relief sought against that defendant. E.g., Hendricks, 521 U.S. at 356–59 (discussing due process in context of civil confinement); BMW of N. Am., Inc. v. Gore, 517 U.S. 559, 574–86 (1996) (discussing due process in context of punitive damages). Accordingly, there is little reason to think that due process need not check criminal personal jurisdiction because of the heightened protections the Constitution affords criminal defendants. More fundamentally, it might be thought that it makes little sense to analogize from civil cases to criminal ones, with the effect (as here) that criminal defendants are ultimately vested with additional rights; after all, it might be said, criminal defendants need not be protected with certain rights (like personal jurisdiction protections) because they are already given such a powerful package of other rights, rights that civil defendants do not have (like the right to a speedy trial, or to confront witnesses). But whatever the general merits of this argument, it is not persuasive in the particular context of the federal extraterritorial prosecutions that are the focus of this Article. Defendants in federal extraterritorial prosecutions are, as a practical matter, forced to do without one of the most elemental of criminal procedure rights—the right to compulsory process. See infra Part V (developing this point). Due process, on the argument developed in this Article, works to restore this fundamental compulsory process right to extraterritorial defendants; in the context of extraterritorial prosecutions, due process replaces a stick that has gone missing from the ordinary bundle of core constitutional criminal procedure rights. It does not add a new one. Part V.
B. Scope and Method
The argument for due process limits on personal jurisdiction in criminal cases applies to federal extraterritorial prosecutions, as when the United States mounts a criminal case based on conduct that took place in France. The same argument applies to state extraterritorial prosecutions, as when New York mounts a prosecution based on conduct in Massachusetts or Spain. 92 N.Y. Crim. Proc. Law § 20.20(2)–(3) (McKinney 2003) (providing for extraterritorial criminal jurisdiction); see, e.g., Chad Bray & Maya Jackson Randall, China Firm Charged With Aiding Iran Nuclear Effort, Wall St. J. (Apr. 8, 2009, 12:01 am), http://www.wsj.com/articles/SB123911892904997247 (on file with the Columbia Law Review) (describing extraterritorial charges filed by state prosecutors in New York).
That said, this Article focuses hereinafter on federal extraterritorial prosecutions. There are two reasons for this. First, federal extraterritorial prosecutions have been extraordinarily important nationally and internationally in a way that state extraterritorial prosecutions have not. Second, there are deep and quite relevant legal differences between federal and state prosecutions in this area, such that it is not clear that they should be analyzed together. For example, while the sovereign’s discretion not to extradite looms large when federal prosecutors seek a defendant from France (as will be discussed below), 93 See infra text accompanying notes 175–177 (discussing discretion permitted by extradition treaties). there is very little discretion when a New York state prosecutor seeks to extradite a defendant from Minnesota. 94 See, e.g., Michigan v. Doran, 439 U.S. 282, 288 (1978) (noting Article IV, section 2 of Constitution renders interstate extradition mandatory); see also Biddinger v. Comm’r of Police, 245 U.S. 128, 132–33 (1917) (contrasting interstate extradition regime with international extradition regime). For further discussion of relevant distinctions between state and federal extraterritorial prosecutions, see Farbiarz, supra note 15, at 524–31.
Against this backdrop, the question is how to pivot from the more abstract to the less, from the general idea that due process limits criminal personal jurisdiction to a fine-grained sense of what those limits should be in the specific context of federal extraterritorial prosecutions. As to method, there are, as always, any number of ways to proceed. But here, some of the usual suspects are not promising. The language of the Fifth Amendment’s Due Process Clause is too open ended to be of much use. 95 See generally supra note 32 for a discussion of how federal actions, such as federal extraterritorial prosecutions, are checked by the Fifth Amendment’s Due Process Clause. And it has virtually no drafting history. 96 See generally Frank H. Easterbrook, Substance and Due Process, 1982 Sup. Ct. Rev. 85, 95–96. History, more generally, is also not especially clarifying. There is no strong constitutional tradition of how to handle extraterritorial prosecutions because there is no real tradition of such prosecutions. The classic view all but precluded extraterritorial prosecutions; crimes were to be prosecuted “exclusively” where they were committed. 97 See section I.A (describing ideal of local criminal jurisdiction). The key exceptions, extraterritorial prosecutions of pirates or U.S. citizens, are not considered here. See supra note 30. This was, until relatively recently, the view of all three branches. The Supreme Court held that “[t]he laws of no nation can justly extend beyond its own territories.” 98 The Apollon, 22 U.S. (9 Wheat.) 362, 370 (1824); accord, e.g., Am. Banana Co. v. United Fruit Co., 213 U.S. 347, 356–57 (1909) (“[T]he general . . . rule is that the character of an act as lawful or unlawful must be determined wholly by the law of the country where the act is done.”); The Schooner Exch. v. McFaddon, 11 U.S. (7 Cranch) 116, 136 (1812) (referring to “absolute and complete jurisdiction within [all sovereigns’] respective territories which sovereignty confers”); Rose v. Himely, 8 U.S. (4 Cranch) 241, 279 (1808) (“[T]he legislation of every country is territorial . . . beyond its own territory, [a country] can only affect its own subjects or citizens.”); Chisholm v. Georgia, 2 U.S. (2 Dall.) 419, 432 (1793) (noting criminal proceedings are “local” in nature). The Executive Branch hewed to Secretary of State John C. Calhoun: “[C]riminal jurisdiction . . . can not extend . . . to acts committed within the dominion of another . . . .” 99 2 John Bassett Moore, A Digest of International Law § 200, at 225 (1906). During the late nineteenth century, Secretary of State Thomas Bayard famously objected to Mexico’s prosecution of a U.S. citizen for an action he took in Texas, stating that “[t]here is no principle better settled than that the penal laws of a country have no extraterritorial force.” Id. § 201, at 236; accord, e.g., Breach of Neutrality, 1 Op. Att’y Gen. 57, 57–58 (1852) (“The offence in question being committed out of the territories of the United States, cannot be noticed by our courts; the offenders must be dealt with abroad . . . .”); see also Manuel R. García-Mora, Criminal Jurisdiction over Foreigners for Treason and Offenses Against the Safety of the State Committed Upon Foreign Territory, 19 U. Pitt. L. Rev. 567, 576 (1958) (describing Secretary of State Webster’s views on extraterritorial scope of federal treason laws). And the Legislative Branch did not push the envelope—Congress passed all of the extraterritorial statutes referenced in this Article during the late twentieth century. 100 Indeed, when Congress did push the envelope, in an eighteenth-century statute relating to extraterritorial felonies, the Supreme Court pushed back—and much harder. See Eugene Kontorovich, Beyond the Article I Horizon: Congress’s Enumerated Powers and Universal Jurisdiction over Drug Crimes, 93 Minn. L. Rev. 1191, 1210–14 (2009) (describing dramatic “narrowing construction” of extraterritorial criminal statute).
Text and history are, in short, of limited use here. But that does not leave the analysis rudderless. Reasoning by analogy is the dominant form of legal reasoning, 101 See, e.g., Cass R. Sunstein, On Analogical Reasoning, 106 Harv. L. Rev. 741, 741–49 (1993) (noting this fact and offering extensive defense of analogical reasoning). and proceeding that way is particularly sensible here. With respect to due process limits on personal jurisdiction, analogical reasoning has been the Supreme Court’s dominant method; indeed, it has typically been the Court’s exclusive method. 102 See, e.g., Daimler AG v. Bauman, 134 S. Ct. 746, 753–58 (2014) (reasoning solely from precedent); Int’l Shoe Co. v. Washington, 326 U.S. 310, 316–20 (1945) (same). Moreover, if due process checks personal jurisdiction in criminal cases for the same basic reasons it checks personal jurisdiction in civil cases, 103 See supra section II.A (advancing this argument). then it makes sense to follow through with the analogy—and to look to due process doctrine, as developed in civil cases, as a basis for due process doctrine in criminal ones. That said, there are large differences between civil and criminal cases. And there is every reason to think those differences should be reflected in the law. 104 Indeed, there is no unitary standard of personal jurisdiction even as between different classes of civil cases. For an example, see the discussion above of the different protections available to defendants and absent class action plaintiffs. Supra text accompanying notes 78–80; see also Russell J. Weintraub, Commentary on the Conflict of Laws 135–37 (6th ed. 2010) [hereinafter Weintraub, Commentary] (supplying other examples). As always, the problem is to analogize—but not too much.
In short, this Article proposes to derive a specific understanding of what, precisely, the limits on criminal personal jurisdiction should look like by drawing an analogy to long-established limits on civil personal jurisdiction. Proposing to articulate criminal doctrine based on an analogy to extant civil doctrine requires, at the outset, a rough sense of the underlying concerns that have shaped the established civil jurisprudence. There are two principal ones. 105 See World-Wide Volkswagen Corp. v. Woodson, 444 U.S. 286, 291–92 (1980) (describing “two related, but distinguishable” concerns). The first is a structural concern for the allocation of government power. And the second is a concern for the individual defendant. Structure is considered in Part III, and concern for the individual defendant in Part IV. 106 As noted above, see supra note 15, and as elaborated upon below, see infra text accompanying notes 328–338, this Article is part of a larger project to rethink due process limits on criminal jurisdiction. Limits on criminal legislative jurisdiction (the subject of another piece by this author, see supra note 15) are motivated by some of the same underlying concerns as limits on criminal personal jurisdiction (the subject of this Article). For that reason, and also to make it crystal clear how these two jurisdictional inquiries ultimately fit together as part of a single integrated whole, the prior piece and this Article are structured in similar ways. Ultimately, though, there are deep differences between the two jurisdictional inquiries. For example, while fairness in the legislative jurisdiction context is largely a matter of whether the defendant is surprised by the content of the substantive law by which he or she is judged, see generally Farbiarz, supra note 15, at 531–45, fairness in the personal jurisdiction context is largely a matter of the integrity of the adjudicative process by which the defendant is judged. See infra Parts IV–V (proposing due process standard for criminal cases). These are very far-reaching differences. And in the end, they mean that while due process limits on criminal legislative jurisdiction should be quite loose, due process limits on criminal personal jurisdiction should have a much stronger practical impact. See infra notes 328–338 (discussing overall project).
III. Structure
The Supreme Court has consistently held that structural concerns (along with concerns for individual fairness) undergird due process personal jurisdiction doctrine. 107 See, e.g., Daimler AG, 134 S. Ct. at 763 (holding exercise of personal jurisdiction violated due process based, in part, on its impact on another sovereign); Asahi Metal Indus. Co. v. Superior Court, 480 U.S. 102, 115–16 (1987) (same); see also J. McIntyre Mach., Ltd. v. Nicastro, 131 S. Ct. 2780, 2789 (2011) (plurality opinion) (“[J]urisdiction is in the first instance a question of [sovereign] authority rather than fairness . . . .”). The statement in Ins. Corp. of Ir. v. Compagnie des Bauxites de Guinee, 456 U.S. 694, 702–03 (1982), that due process limits personal jurisdiction “not as a matter of sovereignty, but as a matter of individual liberty,” seems to be a dead letter. The Court, though, has failed to explain why. 108 E.g., Jay Conison, What Does Due Process Have to Do with Jurisdiction?, 46 Rutgers L. Rev. 1071, 1188–89 (1994) (noting this). Scholars have filled the void, and there are two dominant schools of thought as to the reason structural concerns should shape personal jurisdiction limits. The first is an “internal” approach. It considers the outer limit of a sovereign’s power to resolve a case in isolation, without reference to the impact that resolving the case might have on other sovereigns. The second school of thought is an “external” approach. It maintains that the limit on a sovereign’s power to resolve a case is a product of the impact that its doing so may have on other sovereigns.
Though not mutually exclusive, these two approaches differ from one another. But sections I.A and I.B, respectively, argue that on neither the internal approach nor the external approach should structural concerns play a role in shaping due process doctrine with respect to personal jurisdiction limits on extraterritorial prosecutions.
A. The Internal Approach
The leading exponent of the internal approach is Professor Lea Brilmayer, who has argued that issues of jurisdictional authority “should be analyzed in terms of a state’s right to exercise coercive power over the individual or dispute” 109 Lea Brilmayer, Jurisdictional Due Process and Political Theory, 39 U. Fla. L. Rev. 293, 294 (1987) [hereinafter Brilmayer, Jurisdictional Due Process]. —not principally in terms of the follow-on consequences of the sovereign’s exercise of power. On this account, because government coercion is government coercion no matter where it takes place, the sovereign’s right to exercise power abroad “must be analyzed by reference to the constituting political theory that grants it authority to act domestically,” 110 Lea Brilmayer, Justifying International Acts 22 (1989) [hereinafter Brilmayer, Justifying]. the “political justification” 111 Id. at 2; see also Lea Brilmayer, American Hegemony: Political Morality in a One-Superpower World 11–32, 61–175 (1994) (developing related argument with respect to international arena); cf. Lea Brilmayer, Liberalism, Community, and State Borders, 41 Duke L.J. 1, 4–6 (1991) (discussing justification for reach of state power in context of general and specific personal jurisdiction). that vests the sovereign with authority at home. If extraterritorial action is not justified in this way, it is simply ultra vires and, accordingly, is impermissible, apart from how it might impact other sovereigns. 112 Cf. Nicastro, 131 S. Ct. at 2787, 2789 (plurality opinion) (“[W]hether a judicial judgment is lawful depends on whether the sovereign has authority to render it.”).
On this approach, structural concerns would seem to loom large in the context of extraterritorial prosecutions. When it prosecutes an Algerian citizen for his conduct in Algeria, 113 See supra note 25 (discussing this scenario). there would appear to be at least the possibility that the United States is acting without the requisite “justification.” 114 Brilmayer, Justifying, supra note 110, at 2.
This, though, is not ultimately persuasive. To see why, imagine a car accident in Connecticut, in which the driver and the injured pedestrian are Connecticut residents. And imagine further that there were no constitutional limits on personal jurisdiction. If all that were the case, longstanding background rules would permit a lawsuit growing out of the Connecticut accident to be brought anywhere; in California, for example. 115 The referenced background rules concern “transitory” actions. These are civil suits that can “be brought in any place,” regardless of where the underlying facts of the suit took place. Story, Conflict of Laws, supra note 13, at 771. Transitory actions have long been a part of Anglo-American law. See, e.g., Livingston v. Jefferson, 15 F. Cas. 660, 663–64 (C.C.D. Va. 1811) (No. 8,411) (discussing transitory actions); 3 William Blackstone, Commentaries on the Laws of England *294 (same). And actions closely akin to our hypothetical car accident—tort actions that do not involve real property—have generally been regarded as transitory. See McKenna v. Fisk, 42 U.S. 241, 249 (1843) (characterizing “cases of trespass other than trespass upon real property” as “transitory” such that trial location may be set “without stating where the trespass was in fact committed”); see also, e.g., 21 C.J.S. Courts § 29 (2015) (“An action for tort is, as a rule, transitory.”). But if California were to take jurisdiction over the Connecticut car accident case, it would be exercising power without any legitimate basis for doing so. This is because the principal “justification” that gives a state the “right to operate” 116 Brilmayer, Justifying, supra note 110, at 2. is territorial. “[A] state’s claim to . . . govern a dispute must be based on some thing or event within its territory.” 117 Douglas Laycock, Equal Citizens of Equal and Territorial States: The Constitutional Foundations of Choice of Law, 92 Colum. L. Rev. 249, 251 (1992); accord Hanson v. Denckla, 357 U.S. 235, 251 (1958) (describing “territorial limitations on the power of the respective States”); Prigg v. Pennsylvania, 41 U.S. (16 Pet.) 539, 625 (1842) (describing states’ power as “extend[ing] over all subjects within territorial limits of the states”); Gibbons v. Ogden, 22 U.S. (9 Wheat.) 1, 203 (1824) (describing “immense mass” of legislative power that a states possesses over “every thing within the territory of a State”). But California would not be justified in taking jurisdiction on that basis because our hypothetical car accident happened in Connecticut. States are also justified in exerting power over events that have a sufficient impact within their territorial borders, regardless of where those events themselves occurred. 118 See Strassheim v. Daily, 221 U.S. 280, 285 (1911) (“Acts done outside a jurisdiction, but intended to produce and producing detrimental effects within it, justify a State in punishing the cause of the harm as if he had been present at the effect . . . .”); see also, e.g., Calder v. Jones, 465 U.S. 783, 788–89 (1984) (finding jurisdiction in California proper based on “‘effects’ of . . . [defendant’s] Florida conduct in California”); cf. Allstate Ins. Co. v. Hague, 449 U.S. 302, 315 (1981) (plurality opinion) (affirming application of Minnesota law to fatal Wisconsin motorcycle accident in light of accident’s impact in Minnesota). Stipulate, though, that the Connecticut car accident had no effect on California. And, finally, states are justified in sometimes controlling the conduct of their citizens, even outside of the territory of the state. 119 See Skiriotes v. Florida, 313 U.S. 69, 77 (1941) (affirming application of Florida substantive law to Florida citizen, who acted outside of Florida, based on Florida citizenship). But, again, this justification would be unavailable in the hypothetical Connecticut car accident case. The driver and pedestrian were from Connecticut; no California citizen was involved.
The Connecticut car accident example shows why, on Professor Brilmayer’s theory, it will often make sense to allow structural concerns to play a role in shaping limits on personal jurisdiction. Indeed, if that were not the case, a California court could take jurisdiction over the Connecticut accident even though California would, as noted above, entirely lack the sort of links to its territory or to its citizens that, in our constitutional system, justify a state in acting.
But criminal cases are different. In a civil case, a court can take jurisdiction over a suit alleging a breach of another sovereign’s law. A Texas case may allege a breach of Florida contract law or French tort law; if New Hampshire’s libel law has been violated, the case might go forward in New Hampshire or, for that matter, in Massachusetts. 120 See, e.g., Phillips Petroleum Co. v. Shutts, 472 U.S. 797, 823 (1985) (affirming Kansas court’s determination to take personal jurisdiction over class action but vacating court’s decision that Kansas substantive law applies to all class members); see also Day & Zimmermann, Inc. v. Challoner, 423 U.S. 3, 3–5 (1975) (remanding case to Texas court for application of Texas choice of law rules said to point to application of Cambodian substantive law). Personal jurisdiction (where a case is tried) and legislative jurisdiction (what substantive law is applied to resolve the case) can point in different directions. 121 See J. McIntyre Mach., Ltd. v. Nicastro, 131 S. Ct. 2780, 2790 (2011) (plurality opinion) (“A sovereign’s legislative authority to regulate conduct may present considerations different from those presented by its authority to subject a defendant to judgment in its courts.”); Keeton v. Hustler Magazine, Inc., 465 U.S. 770, 778 (1984) (distinguishing questions of choice of law and personal jurisdiction); Kulko v. Superior Court of Cal., 436 U.S. 84, 98 (1978) (explaining even if application of California law is appropriate, California courts may not have jurisdiction to hear case); Shaffer v. Heitner, 433 U.S. 186, 215 (1977) (rejecting notion that applying a certain state’s law means that state’s courts necessarily have jurisdiction to hear case); Hanson, 357 U.S. at 254 (distinguishing between inquiries into personal jurisdiction and choice of law). But not in criminal cases. In criminal cases, one sovereign’s court can take jurisdiction only when that sovereign’s criminal laws have allegedly been violated. 122 See supra note 75 and accompanying text. And this ultimately means that, in federal extraterritorial prosecutions, the sovereign (the United States) is always acting with constitutional justification—even if no due process limits are imposed on personal jurisdiction.
To begin unpacking the argument, consider the Maritime Drug Law Enforcement Act (MDLEA). The MDLEA authorizes federal narcotics prosecutions of people on non-U.S. ships even if the ships are apprehended on the high seas, far away from U.S. territorial waters. 123 46 U.S.C. § 1903(c)(1)(E) (2012). But in 2012, in United States v. Bellaizac-Hurtado, 124 700 F.3d 1245 (11th Cir. 2012). the Eleventh Circuit vacated a series of MDLEA convictions, reasoning that the MDLEA exceeded Congress’s Article I power under the Law of Nations Clause. 125 Id. at 1249–58.
Or consider the case of Michael Clark. 126 United States v. Clark, 435 F.3d 1100 (9th Cir. 2006). Clark sexually abused scores of children in Phnom Penh and, for this, he was arrested in Cambodia, extradited to the United States, and prosecuted under the so-called Protect Act. 127 Id. at 1103–05. Clark argued that the prosecution was unconstitutional because, in passing the Protect Act, Congress had pushed out beyond the limits of its Article I power to regulate commerce. 128 Id. at 1104. The Ninth Circuit split 2-1. But it ultimately rejected Clark’s argument, holding that the bounds of Article I had not been exceeded. 129 See id. at 1109–17 (holding statute constitutional “[i]n light of Congress’s sweeping powers over foreign commerce”).
Bellaizac-Hurtado and the Michael Clark case are typical. When it comes to federal prosecutions, the extraterritorial application of criminal law is tested against the limits set out in Article I. 130 See, e.g., United States v. Campbell, 743 F.3d 802, 810 (11th Cir. 2014) (measuring reach of extraterritorial criminal statute against scope of Article I power under Piracies and Felonies Clause); United States v. Brehm, 691 F.3d 547, 551 & n.4 (4th Cir. 2012) (conducting this inquiry with respect to Raise and Support Armies Clause); United States v. Pendleton, 658 F.3d 299, 308 (3d Cir. 2011) (conducting this inquiry with respect to Foreign Commerce Clause); United States v. Belfast, 611 F.3d 783, 804–09 (11th Cir. 2010) (conducting this inquiry with respect to Necessary and Proper Clause); United States v. Bredimus, 352 F.3d 200, 204–08 (5th Cir. 2003) (conducting this inquiry with respect to Foreign Commerce Clause). The Executive Branch has not attempted to argue that extraterritorial criminal statutes can be grounded in a nonenumerated foreign affairs power. See Farbiarz, supra note 15, at 524 n.100 (discussing United States v. Curtiss-Wright Export Corp., 299 U.S. 304 (1936)). If those limits are exceeded (as in Bellaizac-Hurtado) the prosecution cannot go forward. If those limits are not exceeded (as in Clark) it can. In our constitutional order, compliance with Article I justifies the federal government’s right to “exercise coercive power,” 131 Brilmayer, Jurisdictional Due Process, supra note 109, at 294. just as a connection to its territory or to its citizens justifies state governments in exercising the same right. 132 See supra text accompanying notes 117–119 (describing bases of state power).
This is a crucial point. If, from the perspective of structural concerns, jurisdiction must be limited to ensure that there is “justification” for the exercise of a sovereign’s power, then there is no need to impose limits on personal jurisdiction in federal criminal prosecutions. This is because justification is already required in such prosecutions—to satisfy Article I’s standards—without any need to superimpose limits on personal jurisdiction. 133 While Article I gives the federal government constitutional justification to act internationally in criminal cases, as in Clark and Bellaizac-Hurtado, it performs much the same function domestically. For example, in United States v. Lopez, 514 U.S. 549, 561–68 (1995) and in Gonzales v. Raich, 545 U.S. 1, 15–33 (2005), federal criminal laws that applied domestically were tested for their compliance with Article I. When Article I limits were exceeded—as in Lopez domestically, 514 U.S. at 567–68, or as in Bellaizac-Hurtado internationally, 700 F.3d 1245, 1249–58 (11th Cir. 2012)—the prosecution could not go forward. And when Article I limits were not exceeded—as in Raich domestically, 545 U.S. at 23–33, and Clark internationally, 435 F.3d at 1109–17—the prosecution could go forward. Accordingly, Article I is not just a justification that must always be present in federal criminal cases, regardless of whether personal jurisdiction limits are layered on. It is also a justification that would arguably seem to pass muster under Professor Brilmayer’s “act[s] domestically” test, at least on the assumption that the Constitution (or, more accurately, certain relevant parts of the Constitution) can be characterized as America’s “constituting political theory” for these purposes. See Brilmayer, Justifying, supra note 110, at 22 (arguing sovereign’s authority to act abroad “must be analyzed by reference to the constituting political theory that grants it authority to act domestically”).
To summarize the point, and to put it in slightly different terms: Because, in extraterritorial federal prosecutions, criminal legislative jurisdiction must always itself be “justified,” by reference to Article I, there is no need separately to justify criminal personal jurisdiction. 134 As noted above, see supra note 15, legislative jurisdiction is the power of a sovereign to prescribe substantive rules; criminal law, for example. In turn, the logic of this point depends on two premises, both of which are valid.
The first premise is qualitative. The justification for criminal legislative jurisdiction (under Article I) can “count” as justification for criminal personal jurisdiction only if, with respect to structural concerns, they require the same kind of justification. Such dovetailing makes sense. There is a close overlap in civil cases between the sorts of facts that give rise to legislative jurisdiction and the sorts of facts that give rise to personal jurisdiction. This has been noted for decades by commentators 135 See, e.g., Eugene F. Scoles et al., Conflict of Laws § 3.26, at 165–66 (4th ed. 2004) (analyzing overlapping considerations relevant to both choice of law and personal jurisdiction inquiries); see also Harold S. Lewis, Jr., The Three Deaths of “State Sovereignty” and the Curse of Abstraction in the Jurisprudence of Personal Jurisdiction, 58 Notre Dame L. Rev. 699, 737 (1983) (describing significant overlap among factors taken into consideration for choice of law and personal jurisdiction inquiries); Courtland H. Peterson, Jurisdiction and Choice of Law Revisited, 59 U. Colo. L. Rev. 37, 37–38 (1988) (describing factors for two inquires as “obviously similar”); A. Benjamin Spencer, Jurisdiction to Adjudicate: A Revised Analysis, 73 U. Chi. L. Rev. 617, 659 (2006) (“[S]ignificant differences between a state’s authority to enact legislation applicable to a dispute and its authority to adjudicate that dispute make little sense.”); James Weinstein, The Federal Common Law Origins of Judicial Jurisdiction: Implications for Modern Doctrine, 90 Va. L. Rev. 169, 240–42 (2004) [hereinafter Weinstein, Federal Common Law Origins] (describing judicial jurisdiction and legislative jurisdiction as “close cousins”). and, occasionally, by individual justices. 136 See, e.g., Allstate Ins. Co. v. Hague, 449 U.S. 302, 320 n.3 (1981) (Stevens, J., concurring in the judgment) (describing two inquiries as “similar,” though “not identical”); Shaffer v. Heitner, 433 U.S. 186, 224–25 (1977) (Brennan, J., concurring in part and dissenting in part) (noting choice of law and personal jurisdiction inquiries “‘depend upon similar considerations’” (quoting Hanson v. Denckla, 357 U.S. 235, 258 (1958) (Black, J., dissenting))); cf. J. McIntyre Mach., Ltd. v. Nicastro, 131 S. Ct. 2780, 2800–01 (2011) (Ginsburg, J., dissenting) (arguing personal jurisdiction is proper in state when “choice-of-law considerations” point in direction of that state’s laws). Moreover, this overlap has only increased. 137 One area of some daylight between them has been that while personal jurisdiction classically focuses on contacts between the sovereign and the defendant, legislative jurisdiction typically considers contacts between the sovereign, the defendant—and also the plaintiff. Walter W. Heiser, A “Minimum Interest” Approach to Personal Jurisdiction, 35 Wake Forest L. Rev. 915, 951 (2000). This is a potentially significant difference, but it has become a bit less sharp. See, e.g., Asahi Metal Indus. Co. v. Superior Court, 480 U.S. 102, 113 (1987) (holding “plaintiff’s interest in obtaining relief” is relevant factor in personal jurisdiction inquiry). And given the much tighter bundling of legislative and personal jurisdiction in criminal cases than in civil ones, 138 Cf. Brilmayer, Introduction to Jurisdiction, supra note 34, at 321 (“[C]riminal law conflates the two inquiries of legislative and adjudicative jurisdiction into one.”). there is that much more reason to think that, in criminal cases, legislative and personal jurisdiction should be predicated on qualitatively similar kinds of facts.
There is a necessary quantitative premise, too. The justification for criminal legislative jurisdiction (under Article I) can suffice as a justification for criminal personal jurisdiction only if, with respect to structural concerns, the standard for the exercise of legislative jurisdiction is set as high as (or higher than) the standard for personal jurisdiction. It should be. In the civil context, scholars have persuasively argued that the bar for legislative jurisdiction should be higher than the bar for personal jurisdiction. 139 For the canonical statement, see Linda J. Silberman, Shaffer v. Heitner: The End of an Era, 53 N.Y.U. L. Rev. 33, 88 (1978) (“To believe that a defendant’s contacts with the forum state should be stronger under the due process clause for jurisdictional purposes than for choice of law is to believe that an accused is more concerned with where he will be hanged than whether.”). For a more recent statement along similar lines, see Weinstein, Federal Common Law Origins, supra note 135, at 240–41 (“[D]ue process would seem to demand a greater connection between the state and an individual before it may legitimately apply its [substantive] law to a controversy than it demands for a state merely to decide a controversy in its courts.”). And for two reasons, this argument applies with greater force in the context of extraterritorial prosecutions. The first reason is the tighter link between personal jurisdiction and legislative jurisdiction in criminal cases. 140 See supra note 138 and accompanying text (describing “tighter bundling” between two).
The second reason relates to the nature of criminal litigation. When federal criminal law is applied to a non-U.S. defendant who acted outside of the United States, the law is being applied to someone who played no direct part in helping to formulate it—who could not have voted for the legislators who wrote the law or for the President charged with enforcing it. 141 See 18 U.S.C. § 611(a) (2012) (criminalizing voting by noncitizens for members of Congress and President); cf. 52 U.S.C.A. § 30121(a)(1) (West 2015) (prohibiting financial contributions by foreign nationals to congressional or presidential elections). This at least potentially presents a problem of justification for a nation generally committed to the idea that government powers rest on “the consent of the governed.” 142 The Declaration of Independence para. 2 (U.S. 1776). But for better or for worse, every time an extraterritorial prosecution goes forward (and is not struck down under Article I), that hurdle of justification-without-representation has, by definition, been cleared. And when Article I has been satisfied, and legislative jurisdiction is accordingly deemed justified, then it should follow a fortiori that, with respect to structural concerns, personal jurisdiction is permissible, too. Establishing legislative jurisdiction in a given extraterritorial prosecution means concluding that the defendant can be subject to the coercive output (criminal laws) of a government body (the legislature) in which the defendant was not permitted to participate. Personal jurisdiction is similar in one respect: The defendant is subject to the coercive output (a criminal judgment) of a government body (the court). But in the personal jurisdiction context, the defendant is emphatically permitted to participate in the activities of the relevant government body, the court. The defendant has the constitutional right to be present in court; 143 See supra notes 18–21 and accompanying text. to call witnesses 144 Washington v. Texas, 388 U.S. 14, 19 (1967). and to cross-examine the government’s witnesses; 145 Crawford v. Washington, 541 U.S. 36, 68–69 (2004). to testify 146 Rock v. Arkansas, 483 U.S. 44, 52 (1987). or not to testify; 147 Griffin v. California, 380 U.S. 609, 615 (1965). and to use an appointed lawyer 148 Gideon v. Wainwright, 372 U.S. 335, 344–45 (1963). or to use no lawyer at all. 149 Faretta v. California, 422 U.S. 806, 820–21 (1975). In light of all the constitutional guarantees set out in the text, it is clear that a criminal defendant does not participate in his or her trial in some merely ancillary or symbolic way, but rather at the heart of the venture. “[T]he distinguishing characteristic of adjudication lies in . . . a peculiar form of participation in the decision, that of presenting proofs and reasoned arguments for a decision in his favor.” Lon L. Fuller, The Forms and Limits of Adjudication, 92 Harv. L. Rev. 353, 364 (1978). That is how a criminal defendant is empowered to participate in trial. If criminal legislative jurisdiction can be justified under Article I—even in the absence of any meaningful (electoral) participation—then, with this much meaningful (courtroom) participation, criminal personal jurisdiction should necessarily pass muster, too. 150 Cf. Daniel Markovits, Adversary Advocacy and the Authority of Adjudication, 75 Fordham L. Rev. 1367, 1371–76 (2006) (arguing legitimacy of government institutions depends on legitimacy with respect to both underlying “theoretical model,” with “emphasis on abstract propositions about justified political power,” and “practical approach,” with emphasis on “consequences of actual engagement” and impact of “participation . . . on the . . . attitudes of the participants”).
* * *
To summarize: If structural concerns are understood from an internal perspective, in terms of the need to justify U.S. actions, then such concerns should play no role in shaping due process limits on federal criminal personal jurisdiction. This is because Article I already supplies the requisite justification. And that justification should suffice to address any structural concerns. Criminal legislative jurisdiction, which Article I controls, should be understood as qualitatively similar to criminal personal jurisdiction, and, indeed, to set a higher bar for U.S. actions than criminal personal jurisdiction does. 151 Especially when a civil defendant is afforded robust rights with respect to participating in the litigation against her, see supra text accompanying notes 143–149 (discussing participation rights afforded to criminal defendants), the argument developed in this section could potentially apply not only to federal criminal prosecutions but also to federal civil actions. In both sorts of cases, the reach of the federal substantive law applied to the defendant is subject to Article I limits.
B. The External Approach
Structural concerns can also be analyzed from an external perspective, which emphasizes that the reach of a sovereign’s power to resolve a case should be limited in light of the impact its doing so might have on other sovereigns. 152 For leading scholarly statements that partake of this external approach, see Allan R. Stein, Styles of Argument and Interstate Federalism in the Law of Personal Jurisdiction, 65 Tex. L. Rev. 689, 689–90, 707–09, 725–26, 728–29 (1987); Stewart E. Sterk, Personal Jurisdiction and Choice of Law, 98 Iowa L. Rev. 1163, 1164–65, 1174 (2013). For an example of the Supreme Court alluding to structural concerns in an “external” way, see World-Wide Volkswagen Corp. v. Woodson, 444 U.S. 286, 293 (1980) (“The sovereignty of each State . . . implie[s] a limitation on the sovereignty of all of its sister States . . . .”); see also Daimler AG v. Bauman, 134 S. Ct. 746, 763 (2014) (holding personal jurisdiction did not satisfy due process based in part on “foreign governments’ objections to some domestic courts’ expansive views of . . . jurisdiction” and “[c]onsiderations of international rapport” (internal quotation marks omitted)) (quoting Brief of United States as Amicus Curiae Supporting Petitioner, Daimler, 134 S. Ct. 746 (No. 11-965), 2013 WL 3377321, at *2); Asahi Metal Indus. Co. v. Superior Court, 480 U.S. 102, 115 (1987) (holding personal jurisdiction did not satisfy due process based in part on concern for “other nations whose interests are affected by the assertion of jurisdiction by the [domestic] court”). When it comes to evaluating the negative impacts that one sovereign may visit upon another, not all sovereigns may be created equal. For example, the fact that an action by a state government ruffles feathers in France may be a constitutional cause for concern; but that does not imply that there is a similar constitutional concern when the federal government takes the same action (and ruffles the same feathers). See Farbiarz, supra note 15, at 524–31. On an external approach, what role should structural concerns play in the particular context of federal extraterritorial prosecutions? “None,” some might answer. An external approach may make sense domestically. States owe fidelity to one another as members of a common union, and due process works to keep them off each other’s toes. But, some would argue, the United States owes other nations nothing as a constitutional matter. If a U.S. action visits negative consequences on another country, there may be a political problem or a policy problem. But not a constitutional one.
This is a strong argument, to which there are strong counterarguments. But there is no need, now, to enter this debate. This is because, as argued below, even if one assumes that it can generally make sense for federal jurisdiction to be constitutionally curbed in light of its impact on foreign sovereigns, there is no reason to be concerned about such impacts in the specific context of extraterritorial prosecutions.
1. Ouster and Commandeering. — To begin the argument, focus on an ordinary domestic civil action, as when a court (say, in California) takes jurisdiction and resolves a tort suit. Under long-established res judicata rules, the California court’s judgment will all but preclude another court (say, in New York) from resolving the same case de novo—on a clean slate, by its own lights. 153 See generally Taylor v. Sturgell, 553 U.S. 880, 892 (2008) (describing res judicata principles); Restatement (Second) of Judgments § 27 (Am. Law Inst. 1982) (same). Because California’s court has considered the case, New York’s court will be prevented from doing so for itself. California’s exercise of jurisdiction will have ousted New York’s. Moreover, if asked to do so, New York’s court may be required affirmatively to get involved by helping to enforce California’s money judgment. 154 Weintraub, Commentary, supra note 104, at 814–16. If New York very strenuously objects to the California judgment, there may be room not to enforce it. But this “public policy” exception is narrow. Baker v. General Motors Corp., 522 U.S. 222, 234 (1998). By exercising jurisdiction today, California’s court will, tomorrow, be able to press New York’s judiciary into California’s service. This is a kind of (mild) commandeering. 155 “Commandeering” is borrowed from New York v. United States, 505 U.S. 144, 161 (1992) (quoting Hodel v. Va. Surface Mining & Reclamation Ass’n, 452 U.S. 264, 288 (1981)), which held that “Congress may not simply ‘commandee[r] the legislative processes of the States by directly compelling them to enact and enforce a federal regulatory program.’”
In sum, by taking jurisdiction, the court of one state can all but force the court of another state not to act (as in ouster) or to act (as in commandeering). This is in some tension with the idea of states as “coequal[s]” 156 World-Wide Volkswagen, 444 U.S. at 292. that enjoy autonomy with respect to one another. Due process checks on personal jurisdiction can help to dissipate that tension. If California’s exercise of jurisdiction was exorbitant—say, if the underlying case involved conduct by New Yorkers in New York—then California’s ousting of New York is perhaps more worrisome. And if California’s exercise of jurisdiction was modest, it will be of less moment that a subsequent exercise of jurisdiction by New York has been displaced. On an external approach, due process curbs one sovereign’s personal jurisdiction on the front end with an eye to the impact on other sovereigns on the back end. 157 For commentators making at least somewhat similar points, see Lea Brilmayer, Conflict of Laws 133–34 (1995) [hereinafter Brilmayer, Conflict of Laws]; Alfred Hill, Choice of Law and Jurisdiction in the Supreme Court, 81 Colum. L. Rev. 960, 977–78 (1981); Jacob Kreutzer, Incorporating Personal Jurisdiction, 119 Penn St. L. Rev. 211, 217–18 (2014); Stein, supra note 152, at 710–11.
But none of this looms large in domestic criminal cases. As a matter of constitutional law, California can take jurisdiction over a criminal case and resolve it on the merits—and New York is then free to take jurisdiction for itself over the same criminal case. 158 See Heath v. Alabama, 474 U.S. 82, 88–91 (1985) (affirming existence of intersovereign exception to double jeopardy protections); see also Abbate v. United States, 359 U.S. 187, 195–96 (1959) (same); Bartkus v. Illinois, 359 U.S. 121, 138–39 (1959) (same). New York is not ousted. Moreover, in its own criminal case, New York is free to resolve disputed factual issues that arose during the California criminal case, regardless of how they were resolved in California. 159 See, e.g., United States v. Charles, 213 F.3d 10, 21 (1st Cir. 2000) (holding federal prosecutors not collaterally estopped by state court ruling on admissibility of evidence); United States v. Davis, 906 F.2d 829, 832–33 (2d Cir. 1990) (“[I]t has long been the rule in this Circuit that ‘collateral estoppel never bars the United States from using evidence previously suppressed in a state proceeding in which the United States was not a party.’” (quoting United States v. Panebianco, 543 F.2d 447, 456 (2d Cir. 1976)); United States v. Safari, 849 F.2d 891, 893 (4th Cir. 1988) (“[C]ollateral estoppel does not apply here because the federal government was not a party in the state court action.”). In addition, the way the criminal case was handled in California will not later tie hands in New York. For example, a prosecutor’s factual concession at the California trial would not be admissible over government objection at a subsequent criminal trial in New York, 160 See generally 5 Jack B. Weinstein & Margaret A. Berger, Weinstein’s Federal Evidence § 801.30[1] (Mark S. Brodin & Matthew Bender eds., 2d ed. 2015) [hereinafter Weinstein & Berger, Federal Evidence] (discussing admissibility of outside statements). The California prosecutor’s statement would potentially be admissible at a subsequent trial in California. See, e.g., United States v. James, 712 F.3d 79, 102 (2d Cir. 2013) (describing conditions under which prosecutor’s statement at prior trial could be admissible). and neither would a transcript of witness testimony from the California trial. 161 See generally 30C Michael H. Graham, Federal Practice and Procedure § 7073 n.6, at 382–83 (interim ed. 2011) (explaining relevant rules and rationale behind preference for live testimony); cf. United States v. North, 910 F.2d 843, 906–07 (D.C. Cir. 1990) (affirming lower court decision to exclude prior videotape testimony); United States v. Martoma, No. 12 CR. 973 (PGG), 2014 WL 5361977, at *4–5 (S.D.N.Y. Jan. 8, 2014) (holding prior deposition inadmissible). The California transcript would very likely be admissible at a subsequent California trial. Weinstein & Berger, Federal Evidence, supra note 160, at § 804.04[2] (noting transcript is acceptable method to prove former testimony).
Commandeering concerns are also largely immaterial in the domestic criminal context. In civil cases, one state must generally enforce another’s judgments. 162 See supra text accompanying notes 154–155 (discussing commandeering). In criminal cases, the rule is the opposite: States are not expected to enforce other states’ judgments. 163 For the classic state court discussion of this point, see Loucks v. Standard Oil Co. of N.Y., 120 N.E. 198, 199–202 (N.Y. 1918) (Cardozo, J.). A California criminal judgment may require the defendant to go to prison. But no one expects New York to enforce that judgment by incarcerating the defendant for California.
Ouster and commandeering are also generally irrelevant in the context of extraterritorial prosecutions. If the United States prosecutes a French resident for conduct that he or she undertook in France, a U.S. prosecution will not displace a subsequent French prosecution. 164 See Christopher L. Blakesley, Extradition Between France and the United States: An Exercise in Comparative and International Law, 13 Vand. J. Transnat’l L. 653, 695 (1980) (“[I]n both France and the United States, foreign judgments are not a bar to prosecution . . . .”); cf. Anthony J. Colangelo, Double Jeopardy and Multiple Sovereigns: A Jurisdictional Theory, 86 Wash. U. L. Rev. 769, 817–18 (2009) (indicating nearly all major civil law countries do not permit double jeopardy claims when first relevant judgment has been handed down by foreign court). Of course, France may not be concerned with losing a chance to prosecute the French defendant. It may be concerned that the defendant is being prosecuted at all. As to which, see infra section III.B.2. Similarly, nations do not enforce each other’s criminal judgments any more than states do. 165 See Wisconsin v. Pelican Ins., 127 U.S. 265, 290 (1888) (“‘The Courts of no country execute the penal laws of another . . . .’” (quoting The Antelope, 23 U.S. (1 Wheat.) 66, 123 (1825))); Restatement (Third) of Foreign Relations Law of the United States § 483 note 3 (Am. Law Inst. 1987) (“Unless required to do so by treaty, no state enforces the penal judgments of other states.”).
To be sure, treaties may alter the background rules. For example, American extradition treaties always include double jeopardy-type protections, 166 Colangelo, supra note 164, at 809–10; see generally Sindona v. Grant, 619 F.2d 167, 176–79 (2d Cir. 1980) (describing history of double jeopardy clauses in extradition treaties). such that if France extradites its citizen to the United States, France will not later be able to extradite that citizen back to France for a follow-on prosecution. 167 See Extradition Treaty Between the United States of America and France, Fr.-U.S., art. 8(1), Apr. 23, 1996, T.I.A.S. No. 02-201 (“Extradition shall not be granted when the person sought has been finally convicted or acquitted in the Requested State for the offense for which extradition is requested.”). This is a kind of ouster. But not the sort that should give us pause. If France goes forward with an extradition, it will be affirmatively choosing to oust its own jurisdiction, by opting to extradite the defendant, a choice that cloaks the defendant with treaty-based double jeopardy protections and precludes a later French prosecution. 168 Extradition treaty protections, including with respect to double jeopardy, only become applicable if the defendant arrives in the United States by means of extradition. Ker v. Illinois, 119 U.S. 436, 443 (1886) (stating nonextradited defendant arrived from Peru “clothed with no rights which a proceeding under the treaty could have given him”); United States v. Valencia-Trujillo, 573 F.3d 1171, 1177–79 (11th Cir. 2009) (noting rule of specialty would only apply to defendant extradited pursuant to treaty); United States v. Valot, 625 F.2d 308, 310 (9th Cir. 1980) (holding when no extradition demand is made, “no ‘extradition’ has occurred and failure to comply with the extradition treaty does not bar prosecution”). As to a country’s discretion to deny an extradition request, see infra notes 176–177 and accompanying text. It is hard to see why U.S. constitutional law should protect France from a consequence it embraces. 169 Cf. Sterk, supra note 152, at 1164–65 (“If the United Kingdom were prepared to require its corporations to submit to worldwide jurisdiction . . . there would be no sovereignty-based reason . . . to limit New Jersey’s power to assert jurisdiction over an entity incorporated in the United Kingdom.”).
In a similar vein, so-called “prisoner transfer” treaties can allow a foreign citizen convicted of a crime in the United States to serve out their prison sentence in the country of his or her citizenship. Such treaties put countries into the business of enforcing each other’s criminal judgments, as when France, for example, keeps a U.S.-convicted defendant behind French bars.
170
The United States and France are parties to the so-called Strasbourg Convention, a multilateral prisoner-transfer treaty that has been ratified by over sixty countries. Convention on the Transfer of Sentenced Persons, Mar. 21, 1983, 35 U.S.T. 2867, 1496 U.N.T.S. 91.
But in the very large majority of cases, it is the receiving sovereign (France, in our example) that asks to enforce the judgment of the sentencing sovereign (the United States), so that its citizen can come home.
171
For the most prominent recent example, see Basil Katz, U.S. Denies Russia Request for Convicted Arms Dealer, Reuters (Nov. 10, 2012, 11:16 am), http://
www.reuters.com/article/2012/11/10/us-russia-usa-bout-idUSBRE8A90BC20121110 [http://
perma.cc/F2KC-XUEA] (describing U.S. denial of Russian request that arms dealer be returned to Russia to serve out remainder of sentence).
An obligation to enforce a U.S. judgment may raise commandeering concerns,
172
See supra notes 154–155 and accompanying text.
but a request to enforce a U.S. judgment does not.
173
Even if it is the sentencing sovereign (the United States) that makes the prisoner-transfer request, the receiving sovereign (France) cannot be forced to accede to it. Convention on the Transfer of Sentenced Persons art. 3(1)(f), Mar. 21, 1983, 35 U.S.T. 2867, 1496 U.N.T.S. 91 (stating both countries must agree to transfer).
2. Control. — Ouster and commandeering, as discussed above, are largely inapposite with respect to criminal prosecutions, domestic and extraterritorial. But consider another set of consequences that one sovereign can visit on another by exercising jurisdiction over a case. Imagine, for example, that the United States prosecutes a French citizen for violating U.S. securities fraud laws, based on the French citizen’s conduct at a Paris investment bank. And imagine further that securities fraud is punishable by a stiffer sentence under U.S. law than French; a maximum of life in prison, say, versus a five-year maximum. Seeing that their colleague is facing up to life in prison, the defendant’s coworkers at the Paris bank may well change their behavior. An action that seemed reasonable when its potential “cost” was five years in prison (when France was presumed to be doing the prosecuting) may seem foolish if its potential cost is a life sentence (now that U.S. prosecutors appear to be reaching into France). 174 The assumption here is that general deterrence works to some extent. That proposition is the subject of a large body of empirical literature, kicked off in the modern era by Jack P. Gibbs, Crime, Punishment, and Deterrence, 48 Sw. Soc. Sci. Q. 515 (1968). In short, as a consequence of exercising jurisdiction over the securities fraud case, the United States will be implicitly controlling French residents’ conduct inside of France. Cross-border control of this sort can plainly cause deep friction between sovereigns. And where cross-border control is a foreseeable follow-on consequence of one sovereign taking jurisdiction of a case, it may seem sensible to limit that sovereign’s jurisdiction up front—so that, say, the United States cannot reach the hypothesized securities fraud case in the first place.
Ultimately, though, that is not persuasive. If France does not want the United States to undertake the prosecution, there is a simple solution: France can decline to extradite the defendant. There will be no U.S. prosecution, and none of the cross-border control that would follow from one. There is little reason for due process to protect a foreign sovereign from a potentially negative consequence that, given the chance, it did not choose to protect itself from. France opted to enter into the extradition treaty. France has opted to stay in the treaty. 175 Every United States extradition treaty permits unilateral withdrawal. See, e.g., Extradition Treaty Between the United States of America and France, Fr.-U.S., art. 26, Apr. 23, 1996, T.I.A.S. No. 02-201 (“Either Contracting State may terminate this Treaty at any time. . . .”). And if France affirmatively decides to send the defendant to the United States for prosecution, why should due process intervene in the name of protecting France?
This argument of course depends on the idea that one country can deny another’s extradition request. It can. Countries generally have discretion to comply or not to comply with extradition requests. 176 See generally Restatement (Third) of Foreign Relations Law of the United States § 475 cmt. g (Am. Law Inst. 1987) (stating “requested state” has “discretion to decline to surrender a particular person to a given state”). Thus, for example, U.S. law vests the Secretary of State with discretion to decline to extradite a defendant—even if the legal requisites of the extradition treaty have been satisfied. 177 See 18 U.S.C. § 3186 (2012) (indicating Secretary of State “may” order extradition (emphasis added)); Escobedo v. United States, 623 F.2d 1098, 1105 n.20 (5th Cir. 1980) (“The Secretary of State always has discretion to refuse to extradite . . . .”); see also 1 John Bassett Moore, A Treatise on Extradition and Interstate Rendition §§ 361–365 (1891) (recounting history of discretionary approach to extradition); Note, Executive Discretion in Extradition, 62 Colum. L. Rev. 1313, 1313–15 (1962) (same). And French domestic law, to continue the analogy, is closely similar. 178 Blakesley, supra note 164, at 715.
Moreover, there is no reason to think that this is, in reality, an untenable Hobson’s choice—between letting an alleged criminal walk free domestically or extraditing him or her abroad. Under the rule of dual criminality, “an accused person can be extradited only if the conduct complained of is considered criminal . . . under the laws of both the requesting and requested nations.” 179 Quinn v. Robinson, 783 F.2d 776, 783 (9th Cir. 1986); accord Collins v. Loisel, 259 U.S. 309, 311 (1922) (“[A]n offense is extraditable only if the acts charged are criminal by the laws of both countries.”). As to the ubiquity of dual criminality requirements in American extradition treaties, see John G. Kester, Some Myths of United States Extradition Law, 76 Geo. L.J. 1441, 1459 (1988) (“[A] standard provision in nearly every United States extradition treaty, is that extradition will not take place unless the offense charged is a crime in both the demanding and the requested country.”). Accordingly, France can extradite to the United States a French citizen for criminal conduct in France only when the underlying conduct violated U.S. criminal law—and also French criminal law. France’s choice will therefore not generally be between extraditing a (potentially dangerous) defendant for prosecution in the United States or leaving the suspect at liberty in France. Rather, France’s choice will be between extraditing the defendant for prosecution in the United States, or denying the extradition request and, if it wants to, proceeding with a prosecution in France. 180 Cf. Extradition Treaty, Mex.-U.S., art. 9(2), May 4, 1978, 31 U.S.T. 5059 (noting if request for extradition is not granted, country that received and then denied extradition request “shall submit the case to its [domestic] authorities for the purpose of prosecution”).
And the proof is in the pudding. Countries not only have formal legal discretion with respect to complying with an extradition request—they make real use of it. Systematic statistics are hard to come by. But there are numerous examples of countries declining U.S. extradition requests. This is true in lower-profile contexts,
181
See, e.g., Rodrigo Labardini, Life Imprisonment and Extradition: Historical Development, International Context, and the Current Situation in Mexico and the United States, 11 Sw. J.L. & Trade Am. 1, 17–18 (2005) (noting Mexico extradited twenty-five people to United States in 2002 but denied twenty-five U.S. extradition requests in same year); Steven Y. Otera, International Extradition and the Medellín Cocaine Cartel: Surgical Removal of Colombian Cocaine Traffickers for Trial in the United States, 13 Loy. L.A. Int’l & Comp. L.J. 955, 969–70 (1991) (noting four-year period during which U.S. made “dozens” of extradition requests and Colombian executive branch officials declined to authorize any of them); see also Bruce Zagaris & Julia Padierna Peralta, Mexico-United States Extradition and Alternatives: From Fugitive Slaves to Drug Traffickers—150 Years and Beyond the Rio Grande’s Winding Courses, 12 Am. U. J. Int’l L. & Pol’y 519, 532 (1997) (describing Mexico’s denial of numerous U.S. extradition requests during 1990s).
and in extremely high-profile ones too.
182
See, e.g., Jeremy M. Sharp, Cong. Research Serv., RL34170, Yemen Background and U.S. Relations 9 (2010), http://fpc.state.gov/documents/organization/152043.pdf [http://perma.cc/PG9C-MSD5] (describing Yemen’s refusal to send to United States Jamal al Badawi, “mastermind” of al Qaeda attack on USS Cole that killed seventeen American sailors); Azam Ahmed & Randal C. Archibold, Mexican Drug Kingpin, El Chapo, Escapes Prison Through Tunnel, N.Y. Times (July 12, 2015), http://www.nytimes.com/2015/07/13/
world/americas/joaquin-guzman-loera-el-chapo-mexican-drug-kingpin-prison-escape.html (on file with the Columbia Law Review) (describing Mexico’s rejection of U.S. extradition request for world’s most prominent drug lord, who later escaped from Mexican prison); Nick Cumming-Bruce & Michael Cieply, Swiss Reject U.S. Request to Extradite Polanski, N.Y. Times (July 12, 2010), http://www.nytimes.com/2010/07/13/movies/13polanski.html (on file with the Columbia Law Review) (describing denial of U.S. extradition request for prominent film director); cf. Oren Dorell, Snowden Extradition Depends on Politics, USAToday (June 25, 2013, 6:32 pm), http://www.usatoday.com/story/news/world/2013/06/25/snowden-extradition/2456719/ [http://perma.cc/JX5V-3EMS] (“National Security Agency leaker Edward Snowden is evading the grasp of U.S. law enforcement because extradition requests by their nature are based not always on legal obligations but political whims.”).
In short, if cross-border control by the United States is the problem, denying extradition to the United States is very often the solution—and a viable one. 183 There are some instances in which U.S. soldiers unilaterally enter a country, arrest a defendant, and forcibly carry the defendant back to the United States for prosecution. Such unilateral abductions raise issues that are not answered by the argument developed in the text. But they need not detain us here. Forcible abductions of this kind are very rare. Cf. Abraham D. Sofaer, The Sixth Annual Waldemar A. Solf Lecture in International Law: Terrorism, the Law, and the National Defense, 126 Mil. L. Rev. 89, 110–11 (1989) (“The only abductions carried out during the Reagan Administration were in international airspace and in international waters.”). And even when they happen, it seems that host-country consent is often given to the United States, though quietly. Id. at 111–12; see also Michael S. Schmidt & Eric Schmitt, U.S. Officials Say Libya Approved Commando Raids, N.Y. Times (Oct. 9, 2013), http://www.nytimes.com/2013/10/09/world/africa/us-officials-say-libya-approved-commando-raids.html (on file with the Columbia Law Review) (describing Libya’s tacit approval of two American commando operations).
C. Conclusion—and Comity
This Part has argued that when it comes to extraterritorial prosecutions, structural concerns should not play a role in shaping due process limits on personal jurisdiction. This is true on an internal approach to structural concerns because Article I already supplies the requisite justification. 184 See supra section III.A. And this is true on an external approach, because some consequences for other sovereigns of the United States taking jurisdiction, like ouster and commandeering, are largely immaterial in criminal cases. 185 See supra section III.B.1. And because other consequences, like cross-border control, can be avoided simply by not extraditing the defendant in the first place. 186 See supra section III.B.2. There may of course be additional theories of why structural concerns might matter in this context. See, e.g., J. McIntyre Mach., Ltd. v. Nicastro, 131 S. Ct. 2780, 2798–99 (2011) (Ginsburg, J., dissenting) (suggesting some justices may be returning to long-dormant “consent”-based approach to personal jurisdiction). Sections III.A and III.B purport to discuss only the two most prominent such theories.
All this said, a final point about structural concerns should be noted. Namely, it might be argued that comity should have a role to play in shaping due process doctrine. France may be willing to extradite a defendant to the United States. But it may begrudge having been asked, and this sentiment can have negative consequences for the whole relationship between France and the United States. 187 See Ralf Michaels, Two Paradigms of Jurisdiction, 27 Mich. J. Int’l L. 1003, 1006 (2006) (describing deep European frustration with respect to U.S. exercises of extraterritorial jurisdiction). But as this author has argued elsewhere, such concerns are best left to the political branches for their consideration—to the Congress that passed the extraterritorial criminal law, and to the Executive Branch that opted to prosecute a given defendant for its breach. 188 See Farbiarz, supra note 15, at 524–31. The question of when, in the name of averting international friction, the United States should stand down from a potential prosecution is shot through with judgments of politics and diplomacy and strategy. It should be left to the political branches, not transmuted into due process doctrine and, in that way, handed over to the courts. 189 Id.
IV. “Burden”
With respect to personal jurisdiction, this Article has argued that the structural concerns that shape due process doctrine in civil cases should not shape due process doctrine in the context of extraterritorial prosecutions. 190 See supra Part III. But a second underlying concern also molds civil due process doctrine—to protect the defendant from “the burdens of litigating in a distant or inconvenient forum.” 191 World-Wide Volkswagen Corp. v. Woodson, 444 U.S. 286, 292 (1980); see also, e.g., Phillips Petroleum Co. v. Shutts, 472 U.S. 797, 807 (1985) (“The purpose of this [minimum-contacts] test, of course, is to protect a defendant from the travail of defending in a distant forum . . . .”). This Part argues that this concern, properly understood, 192 See infra section IV.A. should indeed shape due process doctrine with respect to extraterritorial prosecutions. 193 See infra sections IV.B–C.
A. The Adjudicative Accuracy Theory
A civil action imposes burdens. There are lawyers to retain, depositions to defend, court appearances to make. And all of it is more costly and inconvenient for a defendant sued in a “distant” location. But why should curbing litigation burdens be a matter for due process? The Supreme Court has emphasized that “the individual interest protected” by due process is in “‘not being subject to the binding judgment’” of a court. 194 Van Cauwenberghe v. Biard, 486 U.S. 517, 526 (1988) (quoting Burger King Corp. v. Rudzewicz, 471 U.S. 462, 471–72 (1985)). And this singular focus generally makes sense. Due process checks the exercise of judicial power, 195 See supra text accompanying notes 59–60. and “the most prototypical exercise of judicial power” is “the entry of a final, binding judgment by a court.” 196 Stern v. Marshall, 131 S. Ct. 2594, 2615 (2011). Moreover, due process can be brought to bear only when government deprives a person of property (or life, or liberty). 197 Am. Mfrs. Mut. Ins. Co. v. Sullivan, 526 U.S. 40, 59 (1999). And that is what government does when, through its courts, it enters a binding judgment against the defendant. But why should the due process concern for what happens at the end of a lawsuit (when the binding judgment is entered) imply a concern for what happens in the months or years before the lawsuit ends (when the “travail[s]” 198 Phillips Petroleum Co. v. Shutts, 472 U.S. 797, 807 (1985). of litigation are undergone)? If the individual interest protected by due process is that the defendant is subjected to a judgment, why does another individual interest, the litigation burdens imposed on the defendant, shape due process?
Doctrine sharpens these questions. Consider an individual defendant who moves to dismiss a suit for lack of personal jurisdiction. If the trial court denies the motion, the denial is not immediately appealable. Rather, the defendant must wait until the proceedings have run their course to judgment; at that point, the defendant may file an appeal and, then, renew the jurisdictional argument. 199 Van Cauwenberghe, 486 U.S. at 526–27; see also Catlin v. United States, 324 U.S. 229, 236 (1945) (“[D]enial of a motion to dismiss, even when the motion is based upon jurisdictional grounds, is not immediately reviewable.”). This rule, the Supreme Court has explained, is based in part on the idea that due process protects against the entry of a binding judgment, an interest that can be “effectively vindicated” on appeal with a reversal of the judgment. 200 Van Cauwenberghe, 486 U.S. at 526–27. But, the Court has said, due process does not protect from the difficulties of “stand[ing] trial.” 201 Id. Indeed, that interest could be vindicated only by allowing what the law forbids: an immediate appeal, taken after the motion to dismiss is denied and before the defendant has to stand trial. A concern for litigation burdens is said to shape doctrine. 202 See supra note 191 and accompanying text (describing case law). But then, with respect to the most pointed litigation burden of all, “stand[ing] trial”—it does not. 203 Van Cauwenberghe, 486 U.S. at 526–27.
Inconsistency, though, is not all. Imagine a Connecticut businesswoman sued in Chicago. The defendant may well want to be present for, say, the most important depositions in the case against her. That will be more expensive for the defendant in Illinois than it would have been in Connecticut, in terms of hotels, airfare, and the opportunity cost of missing work. Similarly, while the defendant will surely hire an Illinois lawyer, someone familiar with Chicago judges and juries, the defendant will also, perhaps, choose to retain a second lawyer—from Connecticut, whom the defendant has consulted over the years and who knows the defendant’s business well. In short, for the Connecticut businesswoman, the marginal burdens of litigating in a distant forum are real enough; they include the difference between one lawyer and two, and between flying to depositions in Illinois and driving to them in Connecticut.
But how can this amount to a due process concern? Due process limits only government action, as when a person is required to do something by a statute or by a government official. 204 See Civil Rights Cases, 109 U.S. 3, 10–11 (1883) (holding due process prohibits “State action of a particular character”). But no Illinois statute or court rule compels the defendant to attend a particular deposition 205 See Ill. Sup. Ct. R. & P. 209(a)–(b) (assuming party’s attorney may attend deposition without party present); see also Fed. R. Civ. P. 30(g) (same). or to opt for two lawyers. 206 See Ill. Sup. Ct. R. passim (illustrating absence of rule). And it would be exceedingly unusual for a judge to direct a defendant to hire a second lawyer, or to require certain people to attend a deposition as spectators. The defendant’s choices to hire an extra lawyer or to attend a far-off deposition are not likely to be a direct product of governmental action—a court rule or a judicial order. As such, it is not clear why they are a matter for due process.
It might be argued, though, that the defendant’s choices are an indirect product of government action. The defendant is hiring lawyers and attending depositions only because, by suing her, the plaintiff has invoked the power of the government. And, in that sense, due process might be in play—to limit the choices that the plaintiff, cloaked in government power, can induce the defendant to make, by initiating a suit against the defendant in a distant forum. But this argument from indirect government action is doubly flawed. First, “the due process provision of the Fifth Amendment does not apply to the indirect adverse effects of governmental action.” 207 O’Bannon v. Town Court Nursing Ctr., 447 U.S. 773, 789 (1980). And second, if this argument were accepted, virtually every action taken by a plaintiff in a civil lawsuit would be subject to due process standards. But nothing like that has ever been the law. 208 See 1A Martin A. Schwartz, Section 1983 Litigation: Claims and Defenses § 5.18 n.1031, at 5-225 (4th ed. 2016) (collecting large numbers of federal and state cases); see also, e.g., Harvey v. Harvey, 949 F.2d 1127, 1133 (11th Cir. 1992) (“Use of the courts by private parties does not constitute [government] act[ion] . . . .”). Indeed, there is, for example, a “vast weight of authority” 209 Barnard v. Young, 720 F.2d 1188, 1189 (10th Cir. 1983) (collecting cases). that “an attorney does not become a state actor simply by employing the state’s subpoena laws.” 210 Angelico v. Lehigh Valley Hosp., Inc., 184 F.3d 268, 278 (3d Cir. 1999); accord, e.g., Hahn v. Star Bank, 190 F.3d 708, 717 (6th Cir. 1999) (holding “private attorney issuing a subpoena does not become a state actor for the purposes of § 1983”). And that fundamental principle does not depend on whether the subpoena is for a deposition in a distant locale or a nearby one.
There are, in sum, serious difficulties with the often-invoked idea that due process limits personal jurisdiction to spare defendants from the burdens of litigating in a far-off location. But these difficulties can be resolved by reconceptualizing why litigation burdens matter in the first place. Consider Honda Motor Co. v. Oberg. 211 512 U.S. 415 (1994). In that case, the Supreme Court canvassed its precedents with respect to the issues that arise when a long-existing common law procedure is eliminated: “When the absent [common law] procedures would have provided protection against arbitrary and inaccurate adjudication, this Court has not hesitated to find the proceedings violative of due process.” 212 Id. at 430. As an example, Oberg cited In re Winship, 213 Id. (citing 397 U.S. 358, 361 (1970)). in which the Court had refused to permit New York to jettison the common law “beyond a reasonable doubt” standard, a familiar bulwark against inaccurate adjudication. 214 Winship, 397 U.S. at 361–64 (“The reasonable-doubt standard . . . is a prime instrument for reducing the risk of convictions resting on factual error.”).
But “arbitrary and inaccurate adjudication” does not always result when a particular common law procedure is done away with—and when it does not, Oberg suggested, due process is not necessarily violated. As an example, Oberg cited due process checks on personal jurisdiction. 215 Oberg, 512 U.S. at 431 (indicating “extension of state-court jurisdiction over persons not physically present” was “necessitated by the growth of a new business entity, the corporation” (citing Int’l Shoe Co. v. Washington, 326 U.S. 310 (1945))). The “well-established prior practice,” 216 Id. reflected in Pennoyer v. Neff, 217 95 U.S. 714 (1877). had required a defendant to be physically present in a state before its courts could exercise jurisdiction over her. 218 See id. at 723. And that rule had been based in part on a concern for accurate adjudication. 219 See id. at 726 (explaining, in absence of rule, there would be “fraud,” decisions based on evidence that had “perished,” and judgments for “pretended” claims). But states began to whittle back this requirement. 220 As the Oberg Court noted, see Oberg, 512 U.S. at 431, the history of this process was set out in Burnham v. Superior Court, 495 U.S. 604, 617 (1990) (plurality opinion). And, the Oberg Court noted, the Supreme Court approved of some of this change in International Shoe, 221 Oberg, 512 U.S. at 431. holding that even out-of-state defendants could be subjected to personal jurisdiction if they had “minimum contacts” with the state. 222 Int’l Shoe Co. v. Washington, 326 U.S. 310, 316 (1945). On Oberg’s account, the switch from Pennoyer to International Shoe, from physical presence as the necessary requisite of personal jurisdiction to a more expansive “minimum contacts” approach, seems to have been permissible as a matter of due process because, it was thought, the switch from one legal regime to the other would not lead to “arbitrary and inaccurate adjudication.” 223 Oberg, 512 U.S. at 430–31.
But why not? Oberg does not say. The link, though, between the geographical breadth of personal jurisdiction and the accuracy of adjudication is not hard to see. Accurate adjudication, as used here, is adjudication that gets at the truth, that yields a result that reflects what happened in the world given background legal constraints, like the rules of evidence. 224 The focus here, and throughout this Article, is on the subset of adjudicative inaccuracies that harm the defendant—false positives, for example—with respect to liability. This exclusive focus on the defendant makes sense because the defendant is the holder of the relevant due process right. See infra note 230 (offering less formalistic explanation for focus on defendant). Adjudication can be accurate even if it is arbitrary; a jury, deciding the matter by coin toss, might correctly determine whether the defendant ran a red light. But “[non-]arbitrary and []accurate adjudication” 225 Oberg, 512 U.S. at 430. is possible only when the decision is reasonably information-rich, when the decisionmaker has access to the significant relevant evidence—the key admissible documents, the major eyewitnesses.
When communication and transportation technologies were relatively primitive, it made some intuitive sense to keep the physical scope of personal jurisdiction narrow. A Massachusetts witness might balk at a ten-day trip to a courthouse in Illinois, in which case the jury might never hear the witness’s testimony. And because of that gap in the evidence, the jury might reach an “inaccurate” decision, one that hewed to the facts presented in court but not to what actually happened out in the world. But that changed, as Oberg suggested: “[D]ramatic improvements in communication and transportation made litigation in a distant forum less onerous.” 226 Id. at 431. The same Massachusetts witness could, today, fly to the Illinois trial over the course of an afternoon, or appear virtually, by videotaped deposition or Skype. Bank records that existed in handwritten ledgers in the 1870s, when Pennoyer was decided, are now bits and bytes that can be sent from here to there instantaneously. Given these sorts of sweeping changes, jurisdictional rules could be greatly loosened, as in International Shoe, without much of a decline in “[]accurate adjudication.” 227 Id. at 430.
The same “dramatic improvements” 228 Id. at 431. that made it easier for evidence to get from Massachusetts to Illinois have made it easier and cheaper for the defendant to get to court as well. That matters, too. A defendant may opt to default in the face of anticipated litigation costs, 229 See Daniel Klerman, Rethinking Personal Jurisdiction, 6 J. Legal Analysis 245, 257 (2014) (“[I]f personal jurisdiction rules allowed the plaintiff to choose a forum that was very inconvenient for the defendant, the defendant might choose to allow default judgment to be entered against it rather than to litigate the case.”). and those costs include the expense of the defendant hauling back and forth to Illinois or wherever the trial is to be—an expense that was off the charts in the nineteenth century, but is much less so now. On Oberg’s logic, the reduction in the cost of travel was an argument for tighter jurisdictional rules then (during the Pennoyer era) and for looser jurisdictional rules now (in the post-International Shoe era). This is because a default induced by the “travail” of litigating in a far-off forum can lead to adjudicative inaccuracy; the defendant who defaults to avoid extravagant costs may have had a winning defense on the merits. 230 The plaintiff does not need to look to the Constitution for protection from litigating in an inconvenient forum and the inaccuracies that might cause. Plaintiffs can protect themselves, simply by choosing to bring suit in a forum they find convenient.
To be sure, Oberg is a good deal less explicit in its reasoning than the reconstruction above might suggest. And there are real difficulties with some of the logic sketched out above. Tight jurisdictional rules, for example, may sometimes make it easier for the defendant to get evidence to court. Not always, though. A Connecticut witness might struggle to make it to Illinois. But what if the witness is from Illinois?
But for present purposes, there is no need to go beyond the core point: As the Supreme Court suggested in Oberg, if only elliptically, the oft-cited concern for the burdens of defending in a “distant forum” is actually bottomed on a deeper concern—a concern that such litigation burdens are associated with “arbitrary and inaccurate adjudication.” 231 Oberg, 512 U.S. at 430. To put it differently, litigation’s “direct costs” (like lawyers’ fees and defendants’ time) are principally relevant not in their own right but because they serve as a rough proxy for “error costs.” See Richard A. Posner, An Economic Approach to Legal Procedure and Judicial Administration, 2 J. Legal Stud. 399, 400, 441–42 (1973) (defining “error costs” and “direct costs”).
Stated at this level of abstraction, an accuracy-focused approach to litigation burdens is somewhat familiar, in terms of what the Court 232 See Burger King Corp. v. Rudzewicz, 471 U.S. 462, 478 (1985) (“[J]urisdictional rules may not be employed in such a way as to make litigation so gravely difficult and inconvenient that a party unfairly is at a severe disadvantage in comparison to his opponent.” (internal quotation marks omitted)) (quoting Breman v. Zapata Off-Shore Co., 407 U.S. 1, 18 (1972)); id at 486 (indicating due process “prevent[s] [jurisdictional] rules that would unfairly enable [commercial actors] to obtain default judgments against unwitting customers”); see also World-Wide Volkswagen Corp. v. Woodson, 444 U.S. 286, 301 (1980) (Brennan, J., dissenting) (suggesting connection between burden of litigation and “mobility of defendant’s defense”); Nat’l Equip. Rental, Ltd. v. Szukhent, 375 U.S. 311, 328–29 (1964) (Black, J., dissenting) (warning expansive jurisdiction risks “crippling” defense); cf. Shaffer v. Heitner, 433 U.S. 186, 208 (1977) (suggesting likely location of “important records and witnesses” supports jurisdiction); McGee v. Int’l Life Ins. Co., 355 U.S. 220, 223–24 (1957) (holding, in insurance dispute, personal jurisdiction is proper in state where insured lived because, among other things, “[o]ften the crucial witnesses . . . will be found [there]”); Travelers Health Ass’n v. Virginia ex rel. State Corp. Comm’n, 339 U.S. 643, 649 (1950) (holding jurisdiction proper in Virginia because, among other things, that is “where witnesses would most likely live and where claims . . . would presumably be investigated”). and commentators 233 See Brilmayer, Conflict of Laws, supra note 157, at 269 (stating litigating in distant forum risks undermining “integrity of the adjudicatory process” because “severe inconvenience leads to incomplete factual presentations”); cf. Stephen E. Gottlieb, In Search of the Link Between Due Process and Jurisdiction, 60 Wash. U. L.Q. 1291, 1325 (1983) (“An important justification for minimizing expense is that the cost of suit can place the risk of error on the weaker party.”); Arthur Taylor von Mehren, Adjudicatory Jurisdiction: General Theories Compared and Evaluated, 63 B.U. L. Rev. 279, 289–90 (1983) (discussing potential for “severe inconvenience” to witnesses and parties). have on occasion said. And it makes sense.
First, it solves the jurisprudential problems identified above. Due process is tightly focused on what happens at the end of litigation, when the final judgment is entered. But it is also concerned with what happens before then, when the “travail[s]” 234 Phillips Petroleum Co. v. Shutts, 472 U.S. 797, 807 (1985). of defending in a distant forum are borne. This initially seems like a contradiction. 235 See supra notes 194–203 and accompanying text (outlining apparent incongruity). But ultimately it is not, because bearing the difficult burdens of being forced to litigate in a far-off spot can undermine the accuracy of the final judgment.
Similarly, the “government action” when we speak of litigation burdens is not somehow the plaintiff’s actions, which induce the defendant to bear more litigation burdens, like an extra lawyer or added airfare costs. 236 See supra notes 204–210 and accompanying text (comparing government and private action in due process context). The “government action” is the judgment that the court enters, which may itself be rendered inaccurate because of those litigation burdens.
An accuracy approach to litigation burdens also makes sense of another jurisprudential puzzle. Due process is the “only” part of the Constitution that limits state court personal jurisdiction, 237 Ins. Corp. of Ir. v. Compagnie des Bauxites de Guinee, 456 U.S. 694, 702 n.10 (1982). even as, some argue, other constitutional clauses are a more natural fit. 238 See, e.g., Robert H. Abrams & Paul R. Dimond, Toward a Constitutional Framework for the Control of State Court Jurisdiction, 69 Minn. L. Rev. 75, 87–89 (1984) (proposing “restructuring of jurisdictional dispute resolution” under Full Faith and Credit Clause). But when the scope of personal jurisdiction gets very broad, and a defendant can accordingly be burdened with litigating in a distant forum, the fundamental concern is not litigation burdens qua litigation burdens. It is, rather, a concern for adjudicative accuracy. And adjudicative accuracy is a paradigmatic due process concern. 239 See Greenholtz v. Inmates of Neb. Penal & Corr. Complex, 442 U.S. 1, 13 (1979) (“The function of legal process, as that concept is embodied in the Constitution, and in the realm of factfinding, is to minimize the risk of erroneous decisions.”); see also Jerry L. Mashaw, The Supreme Court’s Due Process Calculus for Administrative Adjudication in Mathews v. Eldridge: Three Factors in Search of a Theory of Value, 44 U. Chi. L. Rev. 28, 48 (1976) (describing leading Supreme Court case as “view[ing] the sole purpose of [due process] procedural protections as enhancing accuracy”).
In addition, an accuracy-focused approach helps to rationalize some of the main lines of personal jurisdiction doctrine. Why are due process standards lower for class action plaintiffs? 240 See Phillips Petroleum Co. v. Shutts, 472 U.S. 797, 811–12 (1985) (holding as such). In part because, given various procedural protections, the accuracy of an adjudication is unlikely to be impacted depending on whether a given class action plaintiff actively participates in the case. 241 Cf. id. at 808–10 (describing limited participation of class action plaintiffs). Personal jurisdiction in a tort suit is often proper (though not always proper 242 See, e.g., Asahi Metal Indus. Co. v. Superior Court, 480 U.S. 102, 112–16 (1987) (finding California court had no personal jurisdiction over defendant even though relevant accident took place in California). ) where the accident took place in part because that is where important eyewitnesses and documents will likely be. 243 See, e.g., McGee v. Int’l Life Ins. Co., 355 U.S. 220, 223 (1957) (“Often the crucial witnesses—as here on the company’s defense of suicide—will be found in the insured’s locality.”). See generally Arthur T. von Mehren & Donald T. Trautman, Jurisdiction to Adjudicate: A Suggested Analysis, 79 Harv. L. Rev. 1121, 1167 (1966) (“[L]itigational convenience, particularly with respect to the taking of evidence, tend in accident cases to point insistently to the community in which the accident occurred.”). And personal jurisdiction is generally proper where a defendant lives 244 See, e.g., Daimler AG v. Bauman, 134 S. Ct. 746, 749 (2014) (holding personal jurisdiction may be asserted against corporate defendants, which are effectively “at home” in forum state); Goodyear Dunlop Tires Operations, S.A. v. Brown, 131 S. Ct. 2846, 2851 (2011) (same). in part because litigation costs are less likely to induce a defendant to default if he or she is being sued close to home.
Finally, an adjudicative-accuracy approach helps to explain a persistent feature of the major personal jurisdiction cases. For generations, the Supreme Court has been focused in such cases on the proverbial “little guy”—the small proprietor who will be pulled into far-off litigation, 245 See, e.g., J. McIntyre Mach., Ltd. v. Nicastro, 131 S. Ct. 2780, 2790 (2011) (plurality opinion) (invoking hypothetical “owner of a small Florida farm” who “could be sued in Alaska”); id. at 2793 (Breyer, J., concurring in the judgment) (stating “[w]hat might appear fair in the case of a large manufacturer . . . might seem unfair” with respect to hypothetical “Appalachian potter” who sells “cups and saucers” and is sued in Hawaii); id. at 2794 (invoking other hypotheticals, including “small Egyptian shirt maker,” “Brazilian manufacturing cooperative,” and “Kenyan coffee farmer”). or the parent fighting for visitation rights on the other side of the country. 246 See, e.g., Kulko v. Superior Court, 436 U.S. 84, 97 (1978) (describing “financial burden and personal strain” on parent “of litigating a child-support suit in a forum 3,000 miles away”). This is unusual. When the Court asks whether due process permits a state to impose tax burdens on an out-of-state resident, it would be surprising if the Court spoke about the out-of-state resident’s ability to pay the tax. 247 See, e.g., Quill Corp. v. North Dakota ex rel. Heitkamp, 504 U.S. 298, 308 (1992) (focusing instead on extent to which corporation directed its activities at forum state). But when the Court asks whether due process permits a state to impose litigation burdens on an out-of-state resident, the Court very much seems to wonder about whether the resident can bear those burdens. On an adjudicative accuracy account, that is no puzzle. A large corporate defendant can afford to get its witnesses to even a far-off forum, while an individual defendant may not be able to do so. 248 See Nicastro, 131 S. Ct. at 2793 (Breyer, J., concurring in the judgment) (noting differences between large and small manufacturers); Klerman, supra note 229, at 257 (discussing impact of litigation costs on defendants without “sufficient assets”); cf. von Mehren, supra note 233, at 313–22 (discussing significance of parties’ abilities to afford litigation burdens); Russell J. Weintraub, Due Process Limitations on the Personal Jurisdiction of State Courts: Time for Change, 63 Or. L. Rev. 485, 526–27 (1984) (arguing issues such as defendant’s physical handicap or work arrangements may make litigation in different forums unfair). One dollar’s worth of added litigation burdens may impact adjudicative accuracy or not depending on how well off a class of defendants is likely to be. 249 Two points should be noted here. First, concerns about “litigation burdens” are sometimes thought “quaint” in light of modern communications technology and transportation infrastructure. E.g., Klerman, supra note 229, at 250 (internal quotation marks omitted). The idea presented here, that concerns for litigation burdens are essentially a proxy for deeper worries about adjudicative accuracy, may render this concern even quainter, at least in civil actions. If litigation burdens matter in and of themselves, they are a feature of virtually every case. But if litigation burdens matter because of their impact on accuracy, they matter in only a smaller subset of cases, as when they induce a default judgment or make it cost-prohibitive for the defendant to get a key witness to trial. Of course, being forced to defend in a distant forum may have a direct impact on adjudicative accuracy, as when an important but reluctant defense witness is beyond the subpoena power of the court selected by the plaintiff. But at least in purely domestic civil actions, this issue rarely occurs. Cf. Paul R. Dubinsky, Is Transnational Litigation a Distinct Field? The Persistence of Exceptionalism in American Procedural Law, 44 Stan. J. Int’l L. 301, 335 (2008) (discussing how Uniform Foreign Depositions Act allows litigants to “count on the assistance of state courts elsewhere”). Second, due process limits on personal jurisdiction can of course be grounded on other values—and, from a normative perspective, it might well be argued that they should be. For example, trying a defendant far from home might be said to impinge on the defendant’s dignity. For present purposes, though, this Article takes as a given the major features of extant due process limits on personal jurisdiction in civil cases. Those limits, this Article argues, are predicated not just on structural concerns, see supra Part III, but also on concern for adjudicative accuracy—not litigation burdens qua litigation burdens, or, for that matter, a concern for dignitary harms.
B. From Civil to Criminal
On the argument developed above, a concern for limiting the sorts of adjudicative inaccuracies that can result when a litigant is forced to defend in a distant forum shapes due process in civil cases. Can this concern be transposed to the context of federal extraterritorial prosecutions? Not in all ways, but certainly in one very important way.
A defendant in a civil case, even with a winning defense on the merits, may opt to default rather than litigate. 250 See supra text accompanying notes 227–229 (discussing defendant’s incentives to default or defend). And criminal defendants sometimes do something analogous, pleading guilty even when they are not. 251 See Jed S. Rakoff, Why Do Innocent People Plead Guilty?, N.Y. Rev. Books (Nov. 20, 2014), http://www.nybooks.com/articles/archives/2014/nov/20/why-innocent-people-plead-guilty/ [http://perma.cc/7YYM-CG95] (discussing this phenomenon and its causes). But while a civil defendant might default to avoid going back and forth to a far-off courthouse, this is not a concern for a defendant in an extraterritorial prosecution. A criminal defendant is, of necessity, already before the forum. 252 See supra notes 17–21 and accompanying text (describing personal jurisdiction rules in federal criminal context). Moreover, a non-U.S. defendant charged for extraterritorial conduct will likely be detained in jail—forcibly required to stay in the forum until the defendant’s case is resolved. 253 See, e.g., United States v. Townsend, 897 F.2d 989, 990–91, 996 (9th Cir. 1990) (noting foreign defendant’s detention without bail). And if the defendant is not detained pending trial, the court can ensure that the defendant can, at no personal expense, travel to and from the courthouse for appearances. 254 See 18 U.S.C. § 4285 (2012) (authorizing district judges to “direct the United States marshal to arrange for that person’s means of non-custodial transportation or furnish the fare for such transportation to the place where his appearance is required”). The defendant may plead guilty. And, tragically, the defendant may plead guilty even though he or she is not. But not because of the expense of going back and forth to court.
Similarly, in civil actions, costs can potentially make it too expensive for the defendant to get a witness or other evidence to trial. 255 See supra text accompanying notes 226–230. But this is not usually a substantial concern with respect to extraterritorial prosecutions. If a criminal defendant cannot afford airfare for an overseas witness, or even for an investigator, the court has the statutory tools to ensure that the defendant’s costs are paid. 256 See § 3006A(e)(1) (authorizing payment for defense investigators, experts, and “other services”); Fed. R. Crim. P. 15(d) (authorizing payment for deposition costs); Fed. R. Crim. P. 17(b) (authorizing payment for witness-fee costs). And, indeed, the Constitution itself sometimes requires defrayal of the defendant’s litigation expenses. 257 See Ake v. Oklahoma, 470 U.S. 68, 76–77 (1985) (describing state’s duty to “take steps to assure that the [indigent] defendant has a fair opportunity to present his defense” as “grounded in significant part on the Fourteenth Amendment’s due process guarantee of fundamental fairness”). Moreover, while there is no available empirical analysis, it seems that judges have worked to ensure that defendants charged with serious extraterritorial crimes are generally given ample resources with which to build a defense. 258 See, e.g., Docket Sheet, United States v. Hage, No. 1:98-cr-01023-LAK (S.D.N.Y. Sept. 21, 1998) (reflecting appointment of six lawyers for defendant Ghailani in extraterritorial prosecution).
But there is one place where the civil–criminal analogy holds. Being required to defend a civil action in a distant forum may diminish adjudicative accuracy if the necessary evidence or witnesses are beyond the court’s subpoena power. 259 See supra note 249 (describing “direct” impact of litigating in distant forum on adjudicative accuracy). And so too in a criminal case. As Justice Story put it: “[T]rial in a distant State or Territory might subject the [criminal defendant] to . . . the inability of procuring the proper witnesses to establish his innocence.” 260 Joseph Story, A Familiar Exposition of the Constitution of the United States § 386, at 278 (Regnery Gateway, Inc. 1986) (1859).
Consider, for example, a hypothetical defendant charged in federal court with murder who wants to call in defense an eyewitness to the killing. If the murder took place in the United States, there are two basic possibilities. If the witness wants to testify, the witness will come to court; if the witness does not want to come to court, the court will make her, by enforcing a subpoena. 261 See Fed. R. Crim. P. 17(a)–(d), (g) (describing process for issuing subpoenas and criminal penalties for noncompliance). But if the murder took place outside the United States, 262 For a federal murder statute with extraterritorial reach, see, for example, 18 U.S.C. § 1116(c) (2012). eyewitnesses will likely also be outside the United States. And if those witnesses do not want to take the trip to an American court, there is no ready way to force them. U.S. courts have subpoena power over American nationals who are abroad. 263 See 28 U.S.C. § 1783(a) (2012) (permitting subpoenas of “national or resident of the United States who is in a foreign country”). But no one else. 264 See United States v. Moussaoui, 382 F.3d 453, 463–64 (4th Cir. 2004) (describing this “well established and undisputed principle”). And while the Sixth Amendment stipulates that “[i]n all criminal prosecutions, the accused shall enjoy the right . . . to have compulsory process for obtaining witnesses in his favor,” 265 U.S. Const. amend. VI. it does not require the court to try to compel the attendance of foreign witnesses. 266 See United States v. Beyle, 782 F.3d 159, 170 (4th Cir. 2015) (“[T]he right to compulsory process does not scorn practicality . . . . [F]ederal courts lack power to secure the appearance of a foreign national located outside the United States.”); United States v. Zabaneh, 837 F.2d 1249, 1259–60 (5th Cir. 1988) (holding defendant “was not denied a constitutionally protected right of compulsory process, in that such right does not ordinarily extend beyond the boundaries of the United States”); cf. United States v. Greco, 298 F.2d 247, 251 (2d Cir. 1962) (holding defendant did not have absolute right to compel attendance of foreign witness).
Thus, earlier this year, the Fourth Circuit affirmed the conviction and life sentence of a defendant who had been charged with committing various crimes abroad. 267 Beyle, 782 F.3d at 173. In the trial court, the defendant had unsuccessfully sought to “access or subpoena” several foreign witnesses whose testimony, he said, would establish his innocence. 268 Id. at 169. The Fourth Circuit held that this was of no moment. “‘[T]he Sixth Amendment can give the right to compulsory process only where it is within the power of the federal government to provide it,’” and the federal government “lack[s] power to secure the appearance of a foreign national located outside the United States.” 269 Id. at 170 (quoting Greco, 298 F.2d at 251).
This problem recurs, 270 For other examples of defendants trying, and failing, to get potentially important evidence from abroad, see United States v. Reyeros, 537 F.3d 270, 280–85 (3d Cir. 2008) (rejecting defendant’s demand that United States turn over Colombian evidence); United States v. Paternina-Vergara, 749 F.2d 993, 998 (2d Cir. 1984) (holding U.S. law enforcement only to duty of good faith to acquire materials from Canada); United States v. Friedman, 593 F.2d 109, 120 (9th Cir. 1979) (holding prosecution need not turn over Jencks Act documents held in Chile). For examples of defendants trying, and failing, to get potentially important witnesses from abroad, see United States v. Croft, 124 F.3d 1109, 1116–17 (9th Cir. 1997) (affirming conviction where trial took place before deposition of foreign defense witness could go forward); United States v. Bastanipour, 697 F.2d 170, 177–78 (7th Cir. 1982) (affirming conviction where trial took place before written testimony of foreign defense witness was obtained). and that is no surprise. In extraterritorial prosecutions, important evidence will virtually always be outside of the United States. Under current law, such evidence can be very difficult for defendants to get in front of the jury—because they cannot get their hands on it in the first place. The potential consequences for adjudicative accuracy are plain. Criminal juries, like civil ones, cannot consistently return accurate verdicts if there are large gaps in the evidence that they see.
C. Accuracy’s Added Impetus in Criminal Cases
In civil actions, due process doctrine is shaped by a concern that trying a defendant in a “distant” forum can lead to diminution in adjudicative accuracy. 271 See supra section IV.A. This concern is present in criminal cases. 272 See supra section IV.B. And, indeed, this section argues, this concern bears much more weight in criminal cases than in civil ones.
There are three reasons why. First, and most fundamentally, our law has long been—rightly—harrowed by fear of convicting the innocent. 273 See 4 Blackstone, supra note 115, at *358 (“[B]etter that ten guilty persons escape than that one innocent suffer.”). See generally Daniel Epps, The Consequences of Error in Criminal Justice, 128 Harv. L. Rev. 1065, 1081–87 (2015) (tracing history and influence of Blackstone’s principle). But there is no similarly intense concern for false positives in tort or contract suits. “We acknowledge that wrongfully depriving an innocent man of his liberty is a worse outcome than wrongfully picking his pocket with an erroneous civil judgment,” 274 J. Harvie Wilkinson III, In Defense of American Criminal Justice, 67 Vand. L. Rev. 1099, 1112 (2014). and that acknowledgement is reflected in many of the basic differences between the law in criminal and civil cases. 275 See Epps, supra note 273, at 1074–75, 1088–89 (collecting examples). There is generally no reason to think that accuracy enough for civil cases should suffice for criminal ones.
Second, while in civil cases due process narrows the scope of personal jurisdiction to preserve adjudicative accuracy, it need not be too focused on that goal. This is because accuracy concerns in civil cases are also accommodated by the forum non conveniens doctrine. That doctrine permits the trial court to decline to hear a case over which it has jurisdiction, in favor of the litigation being resolved in an alternative court, outside of the United States. 276 See generally Piper Aircraft Co. v. Reyno, 454 U.S. 235, 251 (1981) (discussing forum non conveniens doctrine). And forum non conveniens is closely focused on accuracy. Thus, for example, a forum non conveniens dismissal may be appropriate when “crucial witnesses are located beyond the reach of [the] compulsory process” of one court. 277 Id. at 258; see also Travelers Health Ass’n v. Virginia ex rel. State Corp. Comm’n, 339 U.S. 643, 649 (1950) (“[S]uits on alleged losses can be more conveniently tried in Virginia where witnesses would most likely live and where claims for losses would presumably be investigated. Such factors have been given great weight in applying the doctrine of forum non coveniens.”); Gulf Oil Corp. v. Gilbert, 330 U.S. 501, 508 (1947) (“Important considerations [include the] availability of compulsory process for attendance of unwilling, and the cost of obtaining attendance of willing, witnesses . . . .”). Or such a dismissal may be warranted when “access to sources of proof” 278 Gulf Oil Corp., 330 U.S. at 508. such as documents 279 See Piper Aircraft Co., 454 U.S. at 257–58 (affirming lower court conclusion “fewer evidentiary problems would be posed if the trial were held in Scotland” as “large proportion of the relevant evidence is located in Great Britain”). or “premises” evidence 280 Gulf Oil Corp., 330 U.S. at 508. is more straightforward in one court than another. Where forum non conveniens directly addresses accuracy concerns, there is no need for due process doctrine to be robust enough, itself, to carry the whole load. 281 See, e.g., Burger King Corp. v. Rudzewicz, 471 U.S. 462, 477 (1985) (“Most . . . considerations [that “would render jurisdiction unreasonable”] usually may be accommodated through means short of finding jurisdiction unconstitutional. For example . . . a defendant claiming substantial inconvenience may seek a change of venue.”); cf., e.g., Daimler AG v. Bauman, 134 S. Ct. 746, 767–69 (2014) (Sotomayor, J., concurring in the judgment) (alluding to trade-off between personal jurisdiction and forum non conveniens); Stafford v. Briggs, 444 U.S. 527, 553–54 (1980) (Stewart, J., dissenting) (noting civil litigation in an inconvenient forum permitted by due process might nonetheless be transferred under venue transfer statute). And all the more so given that a court may dismiss a case on forum non conveniens grounds even before determining that it has personal jurisdiction over it. 282 Sinochem Int’l Co. v. Malay. Int’l Shipping Corp., 549 U.S. 422, 430–35 (2007).
But forum non conveniens has no bearing on extraterritorial prosecutions. Federal Rule of Criminal Procedure 21 permits transfers of a federal prosecution within the United States from one judicial district to another. 283 See Fed. R. Crim. P. 21(b) (describing transfer “to another district”). But Rule 21 does not purport to permit the transfer of a criminal case from New York to London or to Nairobi. And no common law doctrine does either. 284 This is not surprising. It is controversial enough for federal courts to decline to exercise jurisdiction that Congress has vested in them. See, e.g., Hyde v. Stone, 61 U.S. (20 How.) 170, 175 (1857) (“[T]he courts of the United States are bound to proceed to judgment . . . in every case to which their jurisdiction extends. They cannot abdicate their authority or duty in any case in favor of another jurisdiction.”); Cohens v. Virginia, 19 U.S. (6 Wheat.) 264, 404 (1821) (“It is most true that this Court will not take jurisdiction if it should not, but it is equally true, that it must take jurisdiction if it should.”). And it would be more controversial yet to do so directly against the wishes of the Executive, which has, after all, initiated the prosecution. Moreover, a country that has extradited a defendant to the United States would be surprised (and maybe disturbed) to learn that the defendant is being sent back, unprosecuted. And finally, a criminal transfer of jurisdiction to another country could not pass muster under current forum non conveniens doctrine. Forum non conveniens presupposes the existence of an available alternative forum, and a forum “generally is deemed available if . . . all of the parties [can] come within that alternative court’s jurisdiction.” 14D Charles Alan Wright et al., Federal Practice and Procedure § 3828.3, at 639 (4th ed. 2013). But the United States, a party in any federal criminal case, cannot simply “come within” another country’s courts and prosecute a transferred case. This is because countries do not allow other countries to mount prosecutions in their courts. See supra text accompanying note 75 (noting general understanding that governments may only prosecute violations of their criminal law in their own courts). Without forum non conveniens to pick up the slack, due process must do more work in extraterritorial prosecutions than it needs to do in civil ones. 285 Note that Mathews v. Eldridge, 424 U.S. 319 (1976), like forum non conveniens, helps to ensure that there is a solid backstop of adjudicative accuracy in civil litigation. But Mathews v. Eldridge does not generally apply in criminal cases. See Medina v. California, 505 U.S. 437, 443–44 (1992) (concluding Mathews balancing test is not “appropriate framework” in criminal procedure context).
Finally, in civil cases, due process serves two competing masters. To reduce the danger of adjudicative inaccuracy, tight jurisdictional rules would often be optimal—allowing jurisdiction, perhaps, only where the relevant events took place or where the defendant lives. But if jurisdictional rules are too restrictive, some plaintiffs may not realistically be able to sue, because the only permissible place for filing an action may be far from the plaintiff’s home. 286 See McGee v. Int’l Life Ins. Co., 355 U.S. 220, 223–24 (1957); Travelers Health Ass’n v. Virginia ex rel. State Corp. Comm’n, 339 U.S. 643, 648–49 (1950). Too many fora, and the defendant may not get adjudicative accuracy in court. Too few fora, and the plaintiff may never get into court in the first place. These are powerful weights and counterweights, a centripetal force (adjudicative accuracy) that localizes jurisdiction and a centrifugal force (court access) that tends to broaden its scope—and in the years since International Shoe, due process doctrine in civil cases has come to rest somewhere between them.
But with respect to extraterritorial prosecutions, there is no need to seek out a fine balance between adjudicative accuracy (for the defendant) and court access for the moving party (the plaintiff in civil cases, the government in criminal ones). This Article has argued that due process checks on personal jurisdiction in extraterritorial prosecutions should be understood as shaped not by structural concerns, but rather by concerns for adjudicative accuracy. 287 See supra Part III. But if that is the case, due process doctrine in criminal cases does not need to be built the way it is in civil cases. In civil cases, past a certain mark—“minimum contacts” in the modern formulation—due process requires that a court divest itself of jurisdiction. But if adjudicative accuracy is the sole touchstone, past a certain mark, whatever it might be, due process would simply require that the court take appropriate steps (again, whatever those might be) to protect adjudicative accuracy. If accuracy is the sole problem due process need set itself against, once that problem is solved due process’s work is done. When structural concerns drop out of the equation—as, this Article has argued, they do in the context of extraterritorial prosecutions 288 See supra Part III. —“carefully crafted judicial procedures [can] . . . protect the defendant’s interests,” without any need to dismiss the case so long as those procedures are complied with. 289 J. McIntyre Mach., Ltd. v. Nicastro, 131 S. Ct. 2780, 2789 (2011) (plurality opinion); cf. Burger King Corp. v. Rudzewicz, 471 U.S. 462, 483 n.27 (1985) (denying due process challenge to personal jurisdiction when “[t]he only arguable instance of trial inconvenience occurred when [the defendant] had difficulty in authenticating some corporate records” and “the court offered him as much time as would be necessary to secure the requisite authentication”).
This means something important. In civil cases, due process limits on personal jurisdiction are somewhat relaxed, even at the potential cost of some adjudicative accuracy for defendants, to allow plaintiffs to bring cases in the first place. But in extraterritorial prosecutions, there is no similar need. In the absence of any focus on structural concerns, due process limits are not a matter of whether the United States will be able to bring cases—but how it must proceed when it does so. And if protections for adjudicative accuracy do not potentially come at the expense of bringing large numbers of cases, they can be more robust. 290 Some might argue that to safeguard accuracy, extraterritorial prosecutions must always be dismissed. Adjudicative accuracy presumably requires an “impartial jury,” U.S. Const. amend. VI, and, on this argument, an American jury can only be impermissibly “[]partial” when it sits in judgment over a foreign defendant for alleged crimes committed abroad. But this may be backwards. It is distance from a serious crime that is usually thought of as contributing to accurate decisionmaking, not proximity to it. See, e.g., United States v. McVeigh, 918 F. Supp. 1467, 1471–74 (W.D. Okla. 1996) (transferring Oklahoma City bombing trial to Colorado, citing, among other things, “impairment of the deliberative process of deductive reasoning from evidentiary facts” likely felt by potential Oklahoma jurors). And, in any event, this argument proves too much. It suggests that extraterritorial prosecutions are categorically impermissible because of the gulf they necessarily involve between where a crime happened and where it is tried. But this argument is not viable. See supra note 31 (noting constitutional provisions contemplating extraterritorial prosecutions).
V. What Due Process Requires
The prior Part argued that the contours of due process limits on extraterritorial prosecutions should be shaped by a concern for the decline in adjudicative accuracy that can result from the defendant being required to litigate in a “distant” forum. This Part aims to operationalize that idea, proposing a concrete due process standard.
The proposed standard: In a federal extraterritorial prosecution, the defendant should be given roughly the same access to evidence and witnesses as the defendant would have had if, instead of committing the criminal acts abroad, the defendant had acted inside the United States and sought judicial assistance with respect to the evidence and witnesses. (The basic kind of domestic judicial assistance available to defendants is Federal Rule of Criminal Procedure 17, which allows the defendant to obtain a court subpoena for evidence, to secure a deposition, or to compel trial testimony. 291 Fed. R. Crim. P. 17(a), (c), (f). In deciding whether to enforce a Rule 17 subpoena, courts generally consider, among other things, the specificity of the defendant’s request, the import of the information sought, and the burden on third parties. 2 Wright & Henning, supra note 33, at § 275 n.29, at 269–71 (collecting cases regarding quashed subpoenas). Moreover, even before a postconviction appeal is taken and the harmless error doctrine kicks in, the willingness of a district judge to issue a requested Rule 17 subpoena may turn, as a practical matter, on whether the government’s case is so overwhelming that the subpoena will make no difference. Cf. United States v. Al Fawwaz, No. S7 98-cr-1023 (LAK), 2015 WL 711737, at *13 (S.D.N.Y. Feb. 19, 2015) (denying adjournment in part because “absence of any showing that purported evidence would be helpful to the defendant at trial”). )
If a local phone record would have been successfully subpoenaed by the defendant under Rule 17 if the charged federal crime had taken place inside the United States, then under the proposed due process standard, the same sort of local phone record (if it exists) would need to be provided to the defendant if the charged crime took place outside of the United States. 292 This is not currently the law. See supra text accompanying notes 259–269. Similarly, if an eyewitness would have been successfully subpoenaed under Rule 17 for trial testimony if the crime had taken place in the United States, then under the proposed due process standard, the same sort of eyewitness testimony would need to be preserved (presumably by video deposition) and made available for the defendant’s use at trial. 293 This, too, is not current law. See supra text accompanying notes 259–269. Finally, under the proposed standard, if law-enforcement reports would have been provided to the defendant if the charged crime had taken place in the United States (and, therefore, was initially investigated by the FBI), due process would require those same sorts of reports be provided to the defendant if the crime took place outside of the United States (and, therefore, was initially investigated by foreign police). 294 This is also not current law. See, e.g., United States v. Reyeros, 537 F.3d 270, 280–85 (3d Cir. 2008) (concluding prosecution had no obligation to disclose documents gathered by Colombian police force); United States v. Paternina-Vergara, 749 F.2d 993, 998 (2d Cir. 1984) (requiring only “good faith effort” by prosecution to obtain evidence held by foreign government). Law-enforcement reports must generally be provided to the defendant if they contain (1) a government trial witness’s prior statements, Jencks v. United States, 353 U.S. 657, 668 (1957), or (2) exculpatory information. Brady v. Maryland, 373 U.S. 83, 87 (1963). Adjudicative accuracy is a core purpose of these rules. For more on this point as to prior statements, see United States v. Nixon, 418 U.S. 683, 710–13 (1974); Jencks, 353 U.S. at 666–69; United States v. Burr, 25 F. Cas. 187, 190–92 (C.C.D. Va. 1807) (Marshall, C.J.). As to exculpatory materials, see Brady, 373 U.S. at 86–88; Corinna Barrett Lain, Accuracy Where It Matters: Brady v. Maryland in the Plea Bargaining Context, 80 Wash. U. L.Q. 1, 22 (2002) (“Brady disclosure is perfectly aligned with accuracy interests . . . .”).
The logic of the proposed due process standard is straightforward. For defendants being prosecuted for extraterritorial acts, potential declines in adjudicative accuracy are principally a matter of what Samuel Issacharoff has called a “mismatch.” 295 Samuel Issacharoff, Settled Expectations in a World of Unsettled Law: Choice of Law After the Class Action Fairness Act, 106 Colum. L. Rev. 1839, 1844 (2006). U.S. criminal law reaches the defendant, and so the defendant is prosecuted in a U.S. court for, say, a murder committed in Thailand. But the subpoena power of the U.S. court pulls up short; it does not reach into Thailand along with federal criminal law. 296 See supra section IV.C (describing limited reach of subpoena power). Because of this structural gap, the U.S. court cannot do what it normally would—assist the defendant in putting hands on potentially important evidence that is located where the crime was committed. 297 See generally Issacharoff, supra note 295, at 1844 (describing “mismatch between the scale of the regulated conduct (across a national market) and the jurisdictional reach of the regulator (within a single state)”). The proposed due process standard undoes the mismatch, by requiring that the defendant be treated, for purposes of information gathering, as if the defendant had acted within the subpoena power of the court where he is being tried.
How would this work in practice? In a domestic criminal case, the court does two distinct things with respect to Rule 17 subpoenas. First, the judge decides whether the defendant is entitled to certain evidence or testimony under the standards of Rule 17; and second, the judge enforces the decision, requiring compliance with the subpoena on pain of contempt. 298 Fed. R. Crim. P. 17(a), (c), (g). In the first step, the court decides that the defendant is entitled to, say, certain local bank records; and in the second step, the court ensures that the bank actually provides them.
Under the proposed due process standard, the first step would remain the same. The judge would decide whether the defendant has satisfied the standards of Rule 17 with respect to a request for a certain piece of evidence or a certain witness. If the judge decides the defendant has satisfied the Rule 17 standard, but the evidence and witnesses are abroad, the court would have no way to make its decision count, as it would domestically; the court cannot enforce its writ outside of the United States. 299 See supra notes 259–270 (discussing limited reach of subpoena power).
This would put things in an odd spot. It would be clear from the court’s decision what evidence and witnesses the defendant would have had if he or she had acted in the United States. But, at the same time, the court would not be able to give the defendant actual access to such evidence. And so, under the proposed standard, a due process violation would clearly loom, with whatever remedy it might carry for the prosecution (potentially including dismissal of the case itself).
But that would hardly be the last word. The prospect of a due process violation would strongly incentivize prosecutors to try to make good on the court’s decision, to work to obtain the foreign evidence and witnesses on behalf of the defendant. In the domestic Rule 17 context, the court makes a decision and then enforces it for the defendant. Under the proposed due process standard, the usual process would become bifurcated: The court would make a decision, and the prosecutor would, as a practical matter, enforce it by obtaining evidence on behalf of the defendant.
Could prosecutors successfully accomplish this? 300 As noted above, see supra note 49, this author served for over a decade as a federal national security prosecutor. The factual and predictive assertions in this paragraph and the next rest in part on that experience. Most of the time, yes. Prosecuting an extraterritorial case means getting ducks in a row, and early. If convicting a major drug lord requires the admission at trial of drug ledgers and cocaine seized from the defendant’s compound in Colombia, U.S. prosecutors will not go forward with the indictment and extradition without a firm enough sense that they can “prove up” the search at any federal trial that may follow. And that requires a simple, usually informal set of precommitments—that, when the time for trial comes, Colombia will make available the original evidence from the search, the agents who conducted the search, and the people who logged it and created a chain of custody. This system of informal precommitting works because of personal relationships between repeat players. 301 Cf. Charles Doyle, Cong. Research Serv., No. 94-166, Extraterritorial Application of American Criminal Law 24 (2012), https://www.fas.org/sgp/crs/misc/94-166.pdf [http://perma.cc/B8MZ-J9YL] (noting Drug Enforcement Administration has offices in eighty-five overseas cities and FBI has offices in seventy-five). And it works because convicting the defendant is in the shared interest of Colombia (which has extradited him) and the United States (which will prosecute him). There is little reason to think that American prosecutors could not generally get additional precommitments from their foreign counterparts. If U.S. prosecutors can obtain evidence from country X for use against defendant Y, there is a strong likelihood that, properly incentivized, prosecutors will be able to obtain evidence from country X for use by defendant Y. And it will be that much easier for U.S. prosecutors to make informal agreements to obtain evidence if foreign officials understand that a conviction may not “stick” in the United States if certain materials are not provided. If due process law changes, the implicit bargains struck in its shadow will change, too. 302 See generally Robert H. Mnookin & Lewis Kornhauser, Bargaining in the Shadow of the Law: The Case of Divorce, 88 Yale L.J. 950, 968–70 (1979).
Also important here is the dense web of Mutual Legal Assistance Treaties (MLATs), of which there are over sixty. 303 Doyle, supra note 301, at 22. These MLATs function as a kind of transnational Rule 17 regime, generally empowering the United States to ask other countries to gather documents within their own borders, or to compel a witness to attend a deposition. 304 See id. at 22–23 (providing background on MLATs). In the absence of an MLAT, a litigant may call upon a judge in the United States to issue a letters rogatory, which seeks assistance with respect to evidence-gathering from a foreign judge. Letters rogatory are often thought to work at a much slower pace than MLATs. Id. at 24. MLATs are widely thought of as tools that only prosecutors can use. 305 See, e.g., Daniel Huff, Witness for the Defense: The Compulsory Process Clause as a Limit on Extraterritorial Criminal Jurisdiction, 15 Tex. Rev. L. & Pol. 129, 161 (2010) (“Most MLATs explicitly provide that they do not give rise to a right on the part of any private person to obtain . . . any evidence.” (citation omitted) (internal quotation marks omitted) (quoting Robert Neale Lyman, Compulsory Process in a Globalized Era: Defendant Access to Mutual Legal Assistance Treaties, 47 Va. J. Int’l L. 261, 288 (2006) (quoting Treaty with Ukraine on Mutual Legal Assistance in Criminal Matters, U.S.-Ukr., art. 1, July 22, 1998)). This is usually the position of the Department of Justice, 306 See Linda Friedman Ramirez, Federal Law Issues in Obtaining Evidence Abroad—Part One, Champion, June 2007, at 28, 31 (discussing Department of Justice rationale). and it is based on the fact that virtually all MLATs require the United States, not the defendant or another private citizen, to request assistance from the foreign country whose help is being sought. But especially if due process were thought to require obtaining a foreign document for the defense, it is hard to see why the United States could not, on behalf of the defendant, make an MLAT request for that document. MLATs do not generally limit why the United States might make a given request. 307 The U.S.–France MLAT is typical. Treaty on Mutual Legal Assistance in Criminal Matters, Fr.-U.S., Dec. 10, 1998, T.I.A.S. No. 13,010. It indicates which U.S. authorities may make MLAT requests of France, see id. art. 3 (“[T]he competent authorities are prosecutors and authorities with statutory or regulatory responsibility for investigations of criminal offenses . . . .”), but also makes it clear that France is not to peek behind U.S. MLAT requests, to inquire into their motives or otherwise. See id. (“The presentation . . . of a request coming from such authorities establishes the competence of those authorities.”). And, indeed, there are ample indications that prosecutors sometimes make MLAT requests for defendants, especially when pressed. 308 See, e.g., United States v. Marteau, 162 F.R.D. 364, 372 & n.5 (M.D. Fla. 1995) (noting U.S. agreed to use MLAT to facilitate depositions requested by defendant); Michael Abbell, DOJ Renews Assault on Defendants’ Right to Use Treaties to Obtain Evidence From Abroad, Champion, Aug. 1997, at 20, 22 (“[I]n every known instance in which a trial court indicated it intended to issue a Rule 15 order directing the government to make an MLAT request on behalf of a defendant in a criminal case, the Department of Justice has ‘volunteered’ to make the request.”).
In short, the due process standard proposed here would do two things. First, it would incentivize prosecutors to get precommitments from local officials to assist defendants with respect to foreign evidence-gathering. And second, it would encourage prosecutors to press MLATs into service for defendants. 309 Note that the proposed standard is, as it should be, see supra section IV.C, more demanding than anything that applies in civil cases. There is no suggestion that if a Connecticut car accident is the basis of a Pennsylvania tort suit, the Pennsylvania suit must somehow be the equal of a hypothetical Connecticut lawsuit from the perspective of adjudicative accuracy (or, for that matter, from the perspective of litigation burdens or anything else). It also bears noting that there may be purely domestic criminal cases—cases in which the defendant acted wholly within the United States—where exculpatory evidence is also beyond the subpoena power of the court because it is located abroad. The due process issues potentially implicated in such cases are beyond the scope of this Article.
This would greatly expand the range of evidence that defendants in extraterritorial prosecutions are able to put in front of the jury—putting such defendants in roughly the same place as domestic defendants. That would be a boon to the cause of adjudicative accuracy. Between them, informal precommitting and MLAT use could well carry the whole load of ensuring that defendants who acted abroad get their hands on roughly the same sort of information they would have obtained had they acted inside the United States. 310 The federal prosecutors entrusted with extraterritorial prosecutions often already bend over backward, as they emphatically should, to ensure defense access to foreign witnesses and evidence. See, e.g., United States v. Beyle, 782 F.3d 159, 172 (4th Cir. 2015) (describing government’s efforts to make witnesses who had already pleaded guilty available to defendant). There are of course unfortunate counter-indications, e.g., United States v. Theresius Filippi, 918 F.2d 244, 246–48 (1st Cir. 1990) (acknowledging prosecutor “did not act in accordance with the obligations imposed on her as an agent of justice” when prosecutor failed to request Special Interest Parole from INS to allow witness to testify), and these tend to underscore the need for defendants’ rights to be protected by law, and not just by a healthy prosecutorial culture. And if that were to be the case, adjudicative accuracy would have been improved without even having to consider potential due process remedies, of which dismissing an important extraterritorial prosecution is one.
To be sure, there are added costs for the government if its prosecutors are required to work to secure evidence for the defendant. But these costs will almost always be quite low. In the scheme of things, asking prosecutors to prepare additional MLAT requests, for example, is no serious burden. And all the more so when measured against the high stakes for the defendant (for whom the evidence may be the difference between conviction and acquittal) and the low stakes for the government (which can have no legitimate interest in securing a conviction of a defendant who, on a reasonably fuller record, would have been acquitted). 311 Cf. Mathews v. Eldridge, 424 U.S. 319, 335 (1976) (determining whether due process requires certain procedure to be provided in given case by considering, among other things, respective interests of litigants in dispute).
It is true that the added burdens and costs of having to obtain evidence for defendants might lead federal prosecutors to be more selective, at least at the margins, about the sorts of extraterritorial cases they opt to bring. This should happen infrequently, for the reasons set out above. But when it does happen, that would be a cost for public safety. Fewer prosecutions, though, are almost always the price of fairer prosecutions. 312 See generally William J. Stuntz, The Distribution of Fourth Amendment Privacy, 67 Geo. Wash. L. Rev. 1265, 1274–77 (1999) (providing examples in criminal justice context of instances where legal regulation raises “quality of the regulated activity” while lowering “quantity of the regulated activity”). And, on balance, that is generally a worthwhile tradeoff. 313 Id. (deeming such practices “reasonable”). If a foreign country does not permit, say, exculpatory evidence to be supplied to the defendant, better to deal with that in advance, as the due process standard proposed in this Article would require—rather than dealing with it, if at all, after the fact, after a potentially innocent person has been convicted.
In addition, some of the practical issues spun off by the proposed approach would seem to be ones that can be readily solved. For example, it might be thought that defendants will opportunistically manufacture due process issues by requesting information from abroad just before trial. But the incentives for an already-incarcerated defendant 314 See supra text accompanying note 253 (explaining extraterritorial criminal defendants are often detained pending trial). to needlessly delay trial are low. And, in any event, this can already be an issue under existing law. A defendant might seek to enforce a Rule 17 subpoena on the eve of trial. But courts have had no serious trouble solving these problems by setting real deadlines and by differentiating between genuinely needed information and everything else. 315 See, e.g., United States v. Al Fawwaz, No. S7 98-cr-1023 (LAK), 2015 WL 711737, at *13–16 (S.D.N.Y. Feb. 19, 2015) (highlighting this distinction in review of case law).
Finally, it should be noted that the due process standard proposed here represents the minimum that the Constitution requires—but perhaps not the maximum. The proposed due process standard sets itself against declines in adjudicative accuracy, and concerns for such declines are, this Article has argued, at the core of what animates worries about the “burden” of litigating in a distant forum. But jurisdictional due process, some might argue, should also be shaped by an additional set of concerns—for the dignity of the defendant, for example, or for the defendant’s ability to foresee the possibility of facing trial in a certain location. On those arguments, the due process standard set out in this Part would function as a floor, indicating what due process minimally requires to protect against adjudicative accuracy, but not what due process might also require to vindicate other, additional interests (like the defendant’s dignity, or foreseeability).
Conclusion: Jurisdiction to Adjudicate—And to Legislate
Fifty years ago, Arthur T. von Mehren and Donald T. Trautman published Jurisdiction to Adjudicate: A Suggested Analysis. 316 Von Mehren & Trautman, supra note 243. It is easily the most influential article ever written on personal jurisdiction, and it is encyclopedic. But the article says nothing about criminal jurisdiction. This has long been typical, 317 See John Bernard Corr, Criminal Procedure and the Conflict of Laws, 73 Geo. L.J. 1217, 1217 n.1 (1985). but it is no longer tenable. A surge of important extraterritorial prosecutions has forced us to ask foundational questions about criminal personal jurisdiction, just as cross-border lawsuits during the first decades of the twentieth century required a rethinking of basic questions about civil personal jurisdiction. 318 See supra text accompanying notes 9–12 (describing shift in civil personal jurisdiction framework from Pennoyer v. Neff to International Shoe). This Article, part of a larger project to sketch out the constitutional limits on criminal jurisdiction, 319 See Farbiarz, supra note 15. has developed the following argument.
It has been thought that due process need not limit personal jurisdiction in federal criminal cases because of Article III’s venue requirements, and that due process cannot limit personal jurisdiction in any criminal cases because of the Ker–Frisbie doctrine. But these arguments are not persuasive, 320 See supra sections I.B–C. and, operating from a clean slate, it is clear that just as due process limits civil personal jurisdiction, it should be understood to limit criminal personal jurisdiction. If the lesser coercive power associated with ordinary civil cases requires the imposition of due process limits on jurisdiction, then the more forceful coercive power brought to bear in criminal cases should also require due process limits. 321 See supra Part II.
How might a due process doctrine be built to limit personal jurisdiction in the specific context of extraterritorial prosecutions? Two arguments have been developed.
First, structural concerns should have no role to play in shaping such a due process doctrine. Indeed, from the perspective of structural concerns, such a doctrine is largely superfluous. Article I already supplies the requisite justification for a criminal prosecution, without the need to layer on due process protections. 322 See supra section III.A. And the major consequences for other sovereigns of the United States taking jurisdiction over a case are either immaterial in the criminal context (in the case of ouster and commandeering) or can be avoided simply by not extraditing the defendant (in the case of cross-border control). 323 See supra section III.B.
Second, while a focus on “litigation burdens” has been part of due process law for generations, that is best understood as a proxy for a deeper concern with the decline in adjudicative accuracy that can occur when a defendant is forced to litigate in a “distant” forum. 324 See supra section IV.A. That concern for adjudicative accuracy should shape criminal due process doctrine. 325 See supra section IV.B. Indeed, concern for adjudicative accuracy looms especially large in the context of federal extraterritorial prosecutions, where there is a structural gap between the reach of federal criminal law and the reach of federal courts’ subpoena power. 326 See supra section IV.C. To close that gap, this Article has proposed a due process standard that, by incentivizing prosecutors to gather evidence abroad on behalf of defendants, would improve adjudicative accuracy—and likely without the need to forego large numbers of extraterritorial prosecutions. 327 See supra Part V.
But there is more to the proposed due process standard than a hoped-for win-win situation. One of the deepest anxieties about democracy is that majorities will use their power to mistreat minorities. 328 See, e.g., The Federalist No. 10, at 77 (James Madison) (Clinton Rossiter ed., 1961) (describing concern “measures are too often decided, not according to the rules of justice and the rights of the minor party, but by the superior force of an interested and overbearing majority”). The volumes of logic that have said so have only been buttressed by some of the most terrible pages of our history. But there is, or should be, a similar anxiety in the context of extraterritorial prosecutions. And that is that majorities will mistreat foreigners—strangers to our political system, but subject to our law when they are brought into the United States to be prosecuted for violations of it. We are concerned about minorities because they can have too little say. Constitutional law, though imperfectly, has been on its guard against this problem. Substantive laws that propose to treat outsiders differently than insiders have been struck down in a variety of contexts. 329 See, e.g., Supreme Court of N.H. v. Piper, 470 U.S. 274, 278–82 (1985) (applying Privileges and Immunities Clause to strike down statute restricting bar admission to state residents); Hughes v. Fetter, 341 U.S. 609, 612 (1951) (holding, under Full Faith and Credit Clause, Wisconsin courts required to hear cases arising under wrongful death statutes of other states). This makes sense. “[T]here is no more effective practical guaranty against arbitrary and unreasonable government than to require that the principles of law which officials would impose upon a minority must be imposed generally.” 330 Ry. Express Agency v. New York, 336 U.S. 106, 112–13 (1949) (Jackson, J., concurring). In this respect, federal criminal law has been as it should be. Material support is material support and narco-terrorism is narco-terrorism. Standards are not somehow set lower for Americans than for Spaniards or Egyptians, for domestic conduct than for international. 331 See 18 U.S.C. § 2339B (2012) (defining material support); 21 U.S.C. § 960a (2012) (defining narco-terrorism). But what substance gives, procedure can take. Formal equality in the substantive law is not as meaningful as it needs to be if, as a practical matter, defendants who acted abroad can be convicted of violating a criminal law under a standard that is lower than the standard that applies to crimes committed in America. That is a deep danger, and it is the one that the proposed due process standard protects against. If it is unthinkable that a defendant who acted in Massachusetts might be convicted of murder without somehow being able to bring a bona fide alibi witness into court, then it should be similarly unthinkable if the defendant acted in Morocco. Subjecting extraterritorial defendants to domestic standards of adjudicative accuracy helps ensure that they are treated fairly—and that the law we apply to others is not, in practice, quietly more stringent than the law we apply to ourselves.
I have argued elsewhere that, with respect to legislative jurisdiction, due process imposes only very light and very loose limits on what Congress can criminalize extraterritorially. 332 See Farbiarz, supra note 15, at 545, 552–57 (suggesting limits would only come into play in actual-conflict cases). Under that argument, wisdom and humility must constrain policymakers, but not due process checks on legislative jurisdiction. Federal criminal law can constitutionally reach into distant corners—battlefields in Syria, where chemical weapons have been used on civilians; 333 Id. at 556–57. or into West Africa, where local terrorists have murdered and raped and kidnapped at an awful scale. 334 Id. at 552–53.
Properly understood, due process limits on legislative jurisdiction are loose enough to allow the United States to pursue serious wrongdoing around the world. 335 Id. at 552–57. But, this Article has argued, due process checks on personal jurisdiction must be tight enough to require the United States to pursue global crime with scrupulous concern for adjudicative accuracy. 336 See supra Parts IV–V. “[G]uilt shall not escape,” and “innocence [will not] suffer.” 337 Berger v. United States, 295 U.S. 78, 88 (1935) (describing these as “the twofold aim” of criminal law). This is no contradiction. It is, rather, our criminal law’s most basic creed.