On February 26, 2020, Judge Janet B. Arterton of the federal district court in Connecticut granted a motion under Rule 29 of the Federal Rules of Criminal Procedure to acquit a defendant of seven counts of which he had been convicted by a jury in November 2019. A post-conviction Rule 29 acquittal is uncommon: A prosecutor armed with a jury verdict only rarely is found not to meet the quite lenient standard that the factual evidence, when viewed “in the light most favorable to the government,” provided a basis for a reasonable jury to convict. Even more unusually, the issue on which Judge Arterton found the prosecution lacking did not go to whether the defendant committed the acts of which he was charged or had the requisite state of mind, but whether the criminal statute applied to him at all—an issue that in much of the world is considered to be a question of “jurisdiction” (or, in Europe, of “competence”). The decision raises troubling questions about how threshold issues in criminal cases are resolved under American criminal procedures.
The Hoskins Case
The case is United States v. Hoskins, which has now generated a virtual library of commentary during its long existence. Hoskins is a British subject who for a few years early in the century worked for French manufacturing giant Alstom SA as an officer of its U.K. subsidiary. The gist of the indictment was that he and other Alstom officers and employees engaged in a scheme—the evidence of which all the relevant decisions (and this Comment) assume to be sufficient—to hire third parties knowing that they would pay bribes to Indonesian government officials. The government conceded that Hoskins pursued this scheme in the United Kingdom, France, and Indonesia, but never committed any relevant act in the territory of the United States.
From the outset, Hoskins argued that the U.S. Foreign Corrupt Practices Act (FCPA), on which roughly half of the charges against him were based, did not apply to him at all, and that he could not be charged under it. (He also faced other charges related to money laundering; his conviction and sentence on those charges in November 2019 still stand, subject to a possible appeal, and are not discussed here.) His argument was simple: By its terms, the FCPA restricts the categories of individuals to whom it applies, categories that include U.S. citizens and residents, anyone who commits a relevant crime “in whole or in part” in U.S. territory, and an “agent of a domestic concern.” (The FCPA defines a “domestic concern” as a business that is incorporated or has its principal place of business in the United States, or a person who is a citizen, national, or resident.) Hoskins argued that he fit in no FCPA category, which if true meant that he could not be charged under the FCPA.
The government’s principal position was that even if Hoskins himself did not have one of these statuses (and the government conceded that he met neither the “citizen/resident” nor the “territorial” hook), he helped others who did, and thus could be found vicariously liable under 18 U.S.C. § 2 (aiding and abetting) or for conspiring with them, thus subjecting himself to liability for substantive FCPA violations in which he did not participate directly under Pinkerton v. United States. The arguments on that issue were complex (and have been exhaustively explored by many, including by me), but ultimately—after years of litigation, including a request for rehearing and an unusual pre-trial interlocutory appeal—the government lost: The district court, affirmed on interlocutory appeal by the U.S. Court of Appeals for the Second Circuit, ruled that Hoskins could not be charged under the FCPA as an ancillary to a citizen/resident or someone territorially active in the United States.
This left the question whether Hoskins was an “agent of a domestic concern,” which was not the subject of his energetic pretrial motion practice, nor addressed in the Second Circuit’s opinion.
The “Domestic Concern” Argument
Viewed objectively, common sense suggested that the government’s position was dubious. The “domestic concern” in question was Alstom’s U.S. subsidiary that submitted the bid on the Indonesia project. Put simply, it makes no everyday sense to argue that an officer of the parent company, or of its U.K. subsidiary, is an “agent” of a remote subsidiary of which he was never an officer or employee, and whose offices he never visited. Ultimately, the government’s evidence did not establish that Hoskins was an “agent” of the U.S. subsidiary and in some ways established the opposite: The evidence showed that he was tasked with supervising the U.S. subsidiary’s officers from Europe, not with following them. (In the context of the FCPA, the word “agent” may have (PDF: 393 KB) a specific connotation, namely local intermediaries who at the request of the “principal” actually conduct the illicit quid pro quo discussions and effect payment to the foreign “official.” In fact, the Third Superseding Indictment (PDF: 3.5 MB) against Hoskins specifies two Indonesian individuals who performed exactly such a role (described in the Indictment as “Consultant A” and “Consultant B”).)
Once the government’s aiding and abetting/conspiracy theories had been rejected, the factual issue whether Hoskins was an “agent” became completely determinative of whether the FCPA applied to Hoskins at all. It was, however, never the subject of pretrial fact-finding—because, as this Comment will summarize, U.S criminal procedures provide little or no basis for a pretrial evaluation of contested facts sufficiently alleged in an indictment. Rather, the facts were explored by the jury and the court following this sequence:
- Before trial, the parties submitted lengthy arguments to Judge Arterton as to how to “instruct” the jury to guide them in determining whether Hoskins was an “agent of a domestic concern,” which it would need to find beyond a reasonable doubt in order to convict. Judge Arterton issued an opinion (PDF: 449 KB) setting out her approach, which notably was not to limit the definition to local accomplices, but rather to use general—and quite vague—descriptors of what constitutes an “agent”—generally, anyone who in any way “took direction” from, or was “controlled by,” another (the principal) was its “agent.”
- In their summations, the two parties argued to the jury, inter alia, whether Hoskins was an “agent of a domestic concern.”
- The jury, after deliberation, returned a verdict of guilty on seven FCPA charges. Because it was a “general” verdict, we do not know the jurors’ reasoning or factual basis on the question of Hoskins’s status as an “agent of a domestic concern.”
- Hoskins then filed a motion under Rule 29 arguing, on the FCPA counts, that the evidence that Hoskins had been “an agent of a domestic concern” was “insufficient to sustain [the] conviction.” The government’s response brief (PDF: 207 KB) did not point to evidence that showed any direct basis to assume such a status; rather, it referred to a number of email and other communications involving Hoskins, and argued that the jury could have “inferred” from them that he acted as an agent of the U.S. subsidiary under the general terms pursuant to which the jury had been instructed.
- In her opinion (PDF: 213 KB) granting Hoskins’s Rule 29 motion on the FCPA counts, Judge Arterton reviewed in detail the evidence upon which the government relied, and ultimately concluded that the Government “has identified no evidence introduced at trial which, even when drawing inferences favorable to the Government, could entitle a rational finder of fact to conclude beyond a reasonable doubt that there was an understanding between Mr. Hoskins and API [the U.S. subsidiary] that API would be in control of Mr. Hoskins’s actions on the Tarahan Project or that API did control Mr. Hoskins’s actions in a manner consistent with agency relationships.”
The government has filed a notice of appeal from this decision, which if successful could reinstate the guilty verdicts (or under certain circumstances lead to a new trial). Hoskins has also filed a notice of appeal.
A Uniquely American Process of Addressing an Issue at the Core of the Case
The ultimate result is thus this: A factual issue that has nothing to do with “guilt” in the normal sense of defining impermissible conduct, but to many is “jurisdictional” because it governs whether the case can proceed at all (and whether the court was “competent” to hear it), was not resolved until eight years after indictment, after horrendously expensive proceedings including a lengthy appeal and a complete jury trial (still subject to yet another appeal), and ultimately without any “fact finding” in the normal sense.
From many perspectives, a process that does not allow “threshold” issues to be decided at the “threshold”—that is, before a full jury trial—appears irrational: It wastes resources; by hiding the factual determinants of territorial power behind a general jury verdict, it inhibits allocation of responsibilities among sovereign nations (Hoskins could easily have been prosecuted in France or the United Kingdom); and as noted below it can be unfair to a small but growing class of defendants, namely, non-citizens with little or no connection with the United States who wish to contest its power to enforce its laws against them. In fact, the series of procedures that played out in Hoskins and in other international cases occurs because of an interplay of distinctively American procedures.
Judge-Made Extraterritoriality Principles
First, unlike criminal justice systems in “code-based” regimes in Europe and the many legal systems inspired by them, there is no codified, clear, or consistent set of principles in the United States to inform us (and to direct prosecutors and courts) on the limits of the territorial reach of U.S. criminal statutes. Taking France as an illustrative European example, in an early chapter of the French Criminal Code (PDF: 332 KB) called “on the territorial applicability of a criminal law,” one finds a series of specific provisions, generally articulating well-known principles of international prescriptive law relating to territoriality, detailing with a fair degree of precision when French criminal law does, and does not, apply to allegedly criminal conduct. In the United States, there is no set of generally applicable legislative principles. Rather, as Professor Julie O’Sullivan has explored in her excellent article on The Extraterritorial Application of Federal Criminal Statutes, principles restricting U.S. prosecutors’ reach on territorial grounds in the United States are “judge made,” not terribly coherent, vague, frequently evolving, and of uncertain application.
The Extent of Jury Decision-Making
Second, the United States appears unique in the degree to which a wide array of factual determinations in a criminal case are made by a jury. The Sixth Amendment provides that in a criminal case “the accused shall enjoy the right to a speedy and public trial, by an impartial jury . . . .” While some other countries provide for jury trials on the issue of guilt (that is, whether the defendant committed an illegal act), none allocates to jurors the responsibility to be the ultimate deciders on so many questions not central to guilt as such. In a series of decisions including Apprendi v. New Jersey, the Supreme Court has ruled that a wide range of issues must be submitted to a jury if they dictate a conviction or contribute in certain meaningful ways to a sentence. This includes some important sentencing parameters, but also questions whether “jurisdictional hooks” such as the use of interstate commerce or the mails have been proven beyond a reasonable doubt. Professor O’Sullivan suggests that in cases where the territorial reach of a U.S. criminal statute is contested, the factual basis for a finding of permissible extraterritorial application must also be submitted to a jury and included in its general verdict. Wrapping factual resolution of such “threshold” issues into a general jury verdict is not only unknown in the rest of the world, but may be antithetical to European (and many other) principles of justice: In a 2010 decision of the European Court of Human Rights called Taxquet v. Belgium, the court essentially held that a pure “lay jury” verdict similar to the U.S. model did not respect principles of justice under the European Convention on Human Rights because a general jury verdict does not provide a coherent “statement of reasons” sustaining guilt, capable of being reviewed. If, as Professor O’Sullivan suggests, “territorial reach” issues are included in general jury verdicts without specific and explained fact-finding, from a European perspective the confusion about the extent of U.S. prosecutorial power will simply be magnified.
What Constitutes a “Jurisdictional” Issue Under U.S. Law
Third (and linked to the foregoing), this issue is exacerbated in the United States because the very categorization of certain issues as “jurisdictional” is different from the analysis in other countries. In Morrison v. National Australia Bank Ltd. (PDF: 298 KB), a 2010 civil litigation arising under federal securities law and an important milestone in the current development of “extraterritoriality” analyses, Justice Scalia noted that the “extraterritorial reach” of the operative statute was not a “jurisdictional question” but was rather “a merits question.” In his analysis, a “jurisdictional” inquiry is limited to whether the court has the “power to hear the case” at all; in a federal criminal case, this inquiry amounts only to whether an indictment has been filed charging an “offense[] against the laws of the United States,” which alone is sufficient to trigger a federal district court’s jurisdiction under 18 U.S.C. § 3231. Categorizing essentially all other issues as “merits” questions, as the Morrison Court did, complicates the development of appropriate procedures for addressing “threshold” issues, among other reasons by implying that such issues may only be decided by a jury.
Prosecutorial Discretion in the United States
Fourth and finally, the United States is in a class by itself in giving unreviewable discretion to prosecutors and limiting judicial review of prosecutorial discretion during pretrial investigations and charging decisions. The “judicial hands-off” approach to criminal investigations applies not only to decisions whether to prosecute or not (which in many countries must be approved or reviewed by a judge), but also to such important issues as to whether to dismiss an indictment voted by a grand jury and whether many forms of a negotiated outcome are “in the public interest.” The federal courts in particular have ruled that with the notable (but limited) exception of accepting guilty pleas, judicial oversight of a criminal investigation and of negotiated outcomes such as increasingly prevalent deferred prosecution agreements is limited under separation of powers principles essentially to procedural questions.
Taken together, these U.S. principles give an extraordinary carte blanche to prosecutors interested in pursuing crimes that have little or even no real connection to the United States at all: If prosecutors draft indictments that track the language of the relevant statute sufficiently to survive a motion under Rule 12(b)(2) or (3) of the Federal Rules of Criminal Procedure (which is “rarely granted” and accepts as true all factual allegations), U.S. federal criminal justice provides virtually no basis upon which a judge may explore prior to trial whether the territorial basis for the prosecution is sufficient under current (and still evolving) rules and make a fact-finding record.
International Implications of These U.S. Issues
In addition to the case of Hoskins, a number of recent, unrelated cases (some currently pending) suggest that this is neither a rare nor an inconsequential occurrence.
- In January 2019, Lebanese citizen Jean Boustani was arrested in the United States under a federal indictment charging him with money laundering violations for activities in Mozambique. As in Hoskins, his lawyers early and energetically argued that U.S. laws did not apply to his conduct since he was not a U.S. citizen and never came to the United States. A pretrial motion on this point was rejected (PDF: 1.1 MB) by the district judge on the basis that an indictment “tracking the language of the statute is sufficient.” In December 2019, Boustani was acquitted after a six-week jury trial. A review of the transcript suggests that the jurors were motivated in significant part by concerns about whether the “territorial hooks” had been demonstrated, a conclusion bolstered by a published news report of interviews with three jurors. While one could say that justice was ultimately done, it came at a significant price to Boustani, who remained in jail for almost a year pending his trial.[1]
- In 2017, two French women were indicted in federal court in Brooklyn for alleged violations of the Commodity Exchange Act relating to the London Interbank Offered Rate (LIBOR). One of the defendants hired an attorney to argue that the indictment against her should be dismissed because the Act did not apply to her extraterritorial conduct, since she never committed a relevant act in the United States. The trial court rejected the motion, in significant part because the defendant had not heeded a summons to come from France to the United States, be arrested, and appear in court, and thus was barred by the “fugitive disentitlement principle,” ruling that the concept of a “fugitive” applied to people who did not “flee” the jurisdiction at all but simply refused to come to it. The French defendant has appealed the decision to the Second Circuit.
- In 2019, Turkish bank Halkbank was indicted in federal court for alleged money laundering. It attempted to argue that the relevant laws did not apply to its conduct, but was concerned that if it “appeared” in New York to so argue, it would be considered to be “present” and thus, in essence, to have submitted to the jurisdiction that it wished to contest. It thus asked for a “special appearance,” a concept sometimes used in civil litigation to describe a situation where a civil litigator contests jurisdiction but does not waive its rights by doing so. The trial court ruled that “if [the defendant] wishes the district court to decide its jurisdictional motion, [it] holds the key to unlock its dilemma: travel to New York and answer the charges or have its legal counsel do so.” On the bank’s request for mandamus, the Second Circuit declined to interfere, and the case is proceeding in the district court.
Cases like these seem irrational and unfair to many observers outside the United States, and the virtual non-existence of a straightforward way to contest threshold issues (viewed by many as “jurisdictional”) prior to an expensive jury trial risks creating enormous bad will with important trading countries—whose cooperation on cross-border criminal matters is essential—if the Department of Justice continues to take an aggressive view of its territorial powers. It also appears to some to be inconsistent with the spirit of Article 4 of the OECD Convention on Combating Bribery of Foreign Officials in International Business Transactions (PDF: 813 KB), entitled in its English version “Jurisdiction,” Article 4.3 of which provides that countries investigating overseas bribery should “consult with a view to determining the most appropriate jurisdiction for prosecution.”
Possible Solutions
There is no easy solution. Rule 12 would at first blush seem to invite a pretrial inquiry of the sort described here: It provides that “a motion that the court lacks jurisdiction may be made at any time while the case is pending,” and that a number of other issues (such as venue) “must be raised by pretrial motion,” adding the qualifier “if . . . the motion can be determined without a trial on the merits.” The Rule also provides that a court “must decide every pretrial motion before trial unless it finds a good cause to defer a ruling,” and that “[w]hen factual issues are involved in deciding a motion, the court must state its essential findings on the record.” The cases are clear, however, that the courts lack authority to hold a pretrial evidentiary hearing on any fact that is or constitutes part of an “element of the offense;” they regularly hold (PDF: 137 KB), for example, that courts cannot engage in pretrial fact-finding on issues such as venue since the defendant has a right to a jury trial and a jury is tasked with finding whether venue exists.
I see no impediment—in principle, in practice, or under the Constitution—to a judge faced with a plausible threshold or other “territorial” claim inviting the government to make a “proffer” of the proof supporting territorial reach. The Second Circuit would seem to support this approach. In United States v. Alfonso, 143 F.3d 772 (2d Cir. 1998), the court held that, “[u]nless the government has made what can fairly be described as a full proffer of the evidence it intends to present at trial to satisfy the jurisdictional elements of the offense, the sufficiency of the evidence is not appropriately addressed on a motion to dismiss an indictment.” It then went on to note:
While the formal consent of both parties would not be required for the district court to undertake a pretrial determination of the sufficiency of the jurisdictional evidence where the government has made a sufficient proffer to permit such a ruling, it bears noting that in some cases the government may actually favor such a pretrial ruling. If the district court questions the sufficiency of the evidence satisfying the jurisdictional element of an offense, the government may prefer a pretrial ruling on the issue since that would permit an immediate appeal from a ruling adverse to the government. In the absence of a pretrial ruling, the government faces the risk that the court may ultimately grant a motion to acquit, pursuant to Federal Rule of Criminal Procedure 29, for failure to satisfy an element of the offense, in which case the government would have no opportunity to appeal. [Citation omitted.] It may therefore be to the government’s advantage to make a full proffer and to obtain an appealable pretrial determination of the sufficiency of the evidence on the jurisdictional element.
As stated, such an approach requires the consent of both parties. While such consent might not be provided in all cases, “jurisdictional” facts by their nature tend to be somewhat less vulnerable to partisan disagreement. I can also envision circumstances where a trial judge might make it sufficiently clear to a prosecutor that the judge will be looking carefully at this issue when procedurally appropriate, such as on a Rule 29 motion during or after trial, causing a prosecutor to think twice before refusing consent.
Footnotes
[1] Notably the Second Circuit upheld the trial court’s refusal of bail on the ground that, as a wealthy non-American, Boustani posed a risk of flight. Although Boustani offered to demonstrate his commitment to return to court by employing a private company to assure his appearance through a combination of surveillance and technology such as electronic monitoring, the court rejected that proposal, ruling that allowing bail under those circumstances would not be fair to defendants lacking such resources. While admirable in its concern for fairness, the bail refusal would seem to apply particularly to non-American defendants accused of overseas economic crimes, who can expect to have to wait in jail for months or years while essentially “jurisdictional” issues are decided by a jury.
Frederick T. Davis is a member of the Bars of Paris and New York, and is a Lecturer in Law at Columbia Law School. In July 2019, Cambridge University Press published his book American Criminal Justice: An Introduction.
Disclaimer
The views, opinions and positions expressed within all posts are those of the author alone and do not represent those of the Program on Corporate Compliance and Enforcement (PCCE) or of New York University School of Law. PCCE makes no representations as to the accuracy, completeness and validity of any statements made on this site and will not be liable for any errors, omissions or representations. The copyright of this content belongs to the author and any liability with regards to infringement of intellectual property rights remains with the author.