Territoriality in American Criminal Law

Emma Kaufman* December, 2022

It is a bedrock principle of American criminal law that the authority to try and punish someone for a crime arises from the crime’s connection to a particular place. Thus, we assume that a person who commits a crime in some location—say, Philadelphia—can be arrested by Philadelphia police for conduct deemed criminal by the Pennsylvania legislature, prosecuted in a Philadelphia court, and punished in a Pennsylvania prison. The idea that criminal law is tied to geography in this way is called the territoriality principle. This idea is so familiar that it usually goes unstated.

This Article foregrounds and questions the territoriality principle. Drawing on a broad and eclectic set of sources, it argues that domestic criminal law is less territorial than conventional wisdom holds. Although the territoriality principle is central to criminal law ideology, territorialism is a norm in decline. In reality, over the past century, new doctrines and enforcement practices have unmoored criminal law from geographic boundaries. The result is a criminal legal system in which borders are negotiable and honored in the breach.

Scholars have largely overlooked the deterritorialization of domestic criminal law, but the decline of the territoriality principle has striking implications. It undermines constitutional doctrines and academic theories built on the classic account of criminal law. It upsets foundational conceptual distinctions that structure public law. And it raises normative questions about just how far criminal laws should reach. This Article grapples with those questions and argues that borders are an underenforced constraint on the police power.

Introduction

Twenty years ago, Dustin Higgs was sentenced to death in a federal court in Maryland.¹ Federal law required Higgs to be executed “in the manner prescribed” by the state where he was tried.² But while Higgs was on death row, Maryland abolished the death penalty. So prosecutors asked the court to designate Indiana, a state that permits capital punishment, as the site of Higgs’s execution.³ The trial court refused, but the Supreme Court intervened.⁴ The United States government put Dustin Higgs to death during an unprecedented “spree of executions,” four days before President Trump left office.⁵

Eleven months later, it became a crime to mail abortion pills to Texas.⁶ Under a new Texas law, a doctor in Vermont who prescribes two pills online faces years in a Texas jail.⁷ A Texas court can issue a warrant for that doctor’s arrest, and the governor could seek the physician’s extradition.⁸ The current governor, Greg Abbott, hailed the statute as “a celebration of Texas values.”⁹ A lobbyist behind the bill described it as an effort to ensure that Texas prosecutors can reach people “outside of the state’s strict limits.”¹⁰

These stories seem to have little in common. But both challenge the intuition that criminal law is tied to a particular place. In the first case, the government selected Indiana as the right site to impose a punishment handed down in Maryland. In the second, the Texas legislature criminalized conduct across the country. These cases feel strange—legal, perhaps, but unconventional. To those trained in Anglo-American criminal law, it is odd to think that a Texas criminal statute can reach Vermont and that a crime committed in Maryland can result in an execution in Indiana, hundreds of miles and several states away.

These cases are unsettling because they conflict with a basic assumption about how domestic criminal law works. It is a bedrock principle of American criminal law that the authority to try and punish someone for a crime arises from the crime’s connection to territory. Thus, we assume that a person who commits a crime in some place—say, Philadelphia—can be arrested by Philadelphia police for conduct deemed criminal by the Pennsylvania legislature, prosecuted in a Philadelphia court before a jury of Philadelphia residents, and punished in a Pennsylvania prison. As a corollary, we also tend to assume that Pennsylvania criminal law stops at the state’s borders and that Ohio courts cannot apply Pennsylvania criminal law.¹¹

The idea that criminal law is bound to geography in this way is called the territoriality principle.¹² This idea is deeply embedded in American legal thought. The Model Penal Code (MPC) cites the territoriality principle as a “maxim of American jurisprudence” that differentiates criminal from civil law.¹³ Philosophers identify territoriality as one of the central tenets of criminal law in common law regimes.¹⁴ Conflicts experts presume that criminal laws are territorial and thus “obviously” fall outside their field.¹⁵ Constitutional scholars describe the territorial approach to criminal law as a “very long tradition” in American jurisprudence,¹⁶ and the Supreme Court treats the territoriality of domestic criminal law as self-evident.¹⁷ These sources rarely specify why criminal law is or must be territorial,¹⁸ but they invoke territorialism as a foundational and distinctive feature of domestic criminal law. Everyone else simply assumes the territoriality principle. In practice, the idea that domestic criminal law is territorial is so familiar that it usually goes unstated.

This Article foregrounds and questions the territoriality principle. Drawing on a broad and eclectic set of sources—from penal codes and constitutional cases to private contracts, police manuals, interviews, and materials obtained through open records requests—it argues that domestic criminal law is less territorial than conventional wisdom holds. Territorialism is central to criminal law ideology and a good deal of legal doctrine. But the territoriality principle is a norm in decline.

In reality, over the past century, new laws and enforcement practices have unmoored criminal law from geographic boundaries. This phenomenon has unfolded at every phase of the criminal legal process, from criminalization to punishment. Over time, legislatures have redefined crimes to stretch across territorial borders; courts have expanded their own criminal jurisdiction and recognized states’ authority to prosecute extraterritorial acts; police have pooled their resources and formed interjurisdictional task forces; and prison superintendents have agreed to trade prisoners between states.¹⁹ The result is a legal system in which state criminal law exceeds state borders. In some cases, criminal law need not be territorial at all.

Legal scholars have largely overlooked these developments. But the decline of the territoriality principle has striking implications. As a matter of doctrine, the shift away from territorialism upsets legal rules built on the classic account of criminal law. The claim that domestic criminal law is territorial runs throughout substantive criminal law and constitutional jurisprudence. The territoriality principle shapes federalism and criminal procedure doctrine and surfaces in everything from habeas and abstention cases to Commerce Clause disputes.²⁰ The erosion of territorialism makes these lines of doctrine feel anachronistic and strained.

At a more practical level, the evolution of criminal jurisdiction creates basic good governance concerns. As this Article explains, the territoriality principle receded over the course of the twentieth century as police and prison officials began to enter private agreements—contracts, compacts, memoranda of understanding—that extend the reach of criminal laws. These agreements effectively rewrite the borders of domestic criminal law. Their proliferation raises pressing questions about transparency in criminal law enforcement.

The decline of territorialism also has broad theoretical implications. Although it is rarely discussed, the territoriality principle plays a critical role in defining and legitimating criminal law. In treatises and legal opinions, territoriality is understood as a unique feature of domestic criminal law—a background principle that distinguishes the field from both civil and international law.²¹ On the standard account, domestic criminal law takes a more territorial approach to jurisdiction than international criminal law, and criminal law’s commitment to territorialism explains why choice of law exists in only the civil domain.²² In theory, territorialism also helps to justify sanctions imposed in the name of “the people.”²³

If territorialism is receding, these claims make less sense. If, as this Article shows, domestic criminal law contains nonterritorial theories of jurisdiction and the administration of criminal law blurs territorial boundaries, then criminal law is not so different from other fields. And if criminal law is not distinctive, it is difficult to explain some of the more peculiar aspects of our criminal legal system. Without territorialism, for example, it is hard to understand the prohibition on choice of criminal law and the presumption that state criminal law stops at state lines. It also becomes trickier (even more impossible than it already is) to defend penal sanctions. To echo Professor Richard Ford, if criminal jurisdiction is not territorial, “all that is solid melts into air.”²⁴

That insight, in turn, opens up some interesting questions about whether territorialism is good for criminal law. This Article’s basic thesis is that the borders of domestic criminal law are up for grabs. Questions of criminal jurisdiction are much more contested and flexible than many scholars and critics of American criminal justice seem to have realized. Recognizing this fact reveals some provocative possibilities for criminal justice reform, such as redistricting police forces, regionalizing criminal lawmaking, and permitting criminal defendants to forum shop. The erosion of territorialism has the potential to transform the design and administration of criminal law.

Yet, at the same time, the story outlined below should give reformers pause. The decline of the territoriality principle has not been a boon for criminal defendants. On the contrary, one of this Article’s core lessons is that the erosion of territorialism has empowered police, prosecutors, and prison wardens, who have benefited from growing flexibility about the outer boundaries of their authority. The history of criminal jurisdiction demonstrates that, although they are fraught and arbitrary, borders are a crucial restraint on the police power.²⁵ Those concerned about abuse of power in the criminal justice system might then want to revive territorialism rather than embrace its decline. Animated by that concern, this Article makes a case for more territorialism in American criminal law.

This argument proceeds in three Parts. Part I defines the territoriality principle and traces its history. At first, it is unclear why so many authoritative sources seem to believe that domestic criminal law has to be territorial. Part I sets out to answer that question. It locates the origins of the territoriality principle in substantive criminal law, constitutional law, and the institutional design of the criminal justice system.

Part II documents territorialism’s decline over the course of the twentieth century. Bringing together a wide variety of sources, Part II depicts a criminal justice system that has been deterritorialized in two respects. First, courts have embraced nonterritorial conceptions of criminal jurisdiction—holding, for example, that states can predicate their criminal laws on citizenship rather than presence on state soil.²⁶ Second, the administration of criminal law has become more centralized and collaborative. Together, these trends have produced a legal regime that is less territorial than one might expect, and, where it remains tied to geography, is much less wedded to state borders than one steeped in doctrine and rhetoric about American criminal law would imagine.

Part III explores the implications of this account. It begins by explaining how the deterritorialization of criminal law undermines doctrines and conceptual distinctions that separate criminal law from other fields. Part III then turns to the normative questions implicit in the erosion of the territoriality principle. In the end, the Article offers a defense of territorial jurisdiction.

Its real aim, though, is to demonstrate that the territoriality principle is worth debating. At its heart, this is a piece about the importance of criminal jurisdiction. Scholars and students of American law tend to think that the study of jurisdiction is technical and dry, or as Professor William Prosser put it in 1953, “a dismal swamp” inhabited by “eccentric professors who theorize about mysterious matters in a strange and incomprehensible jargon.”²⁷ Perhaps it is unsurprising, then, that there has been little scholarship on territoriality in domestic criminal law.²⁸ American criminal law textbooks omit jurisdiction, and the leading article on the topic remains a piece published in 1926.²⁹ In the meantime, criminal law scholars have turned their attention toward the political incentives and institutional dynamics that make the country’s criminal justice system so harsh and indefensible.³⁰

But the territoriality principle persists in doctrine, theory, and ideology about criminal law, and it plays a pivotal role in justifying the power to punish. In other areas of law, academics are having lively debates about the merits and drawbacks of territorialism. Scholars of constitutional law, federal courts, civil procedure, immigration law, international law, and conflicts have all asked whether the law they study is (or should be) based on a territorial theory of state power.³¹ Criminal law scholars have said less about this question. Yet the shift away from territorialism is no less evident in criminal law, and it may even be more destabilizing to the field.

Moreover, this seemingly obscure topic has serious stakes. The decline of the territoriality principle is why Dustin Higgs could be executed in Indiana.³² It is why Texas can criminalize prescriptions written in Vermont.³³It is also—to use a looming example—why it is less clear than one might think whether Missouri can ban its residents from getting abortions outside the state now that the Supreme Court has overturned Roe v. Wade.³⁴ The legal system’s waning commitment to territorialism raises disquieting questions about the limits of criminal law.

I. The Origins of the Territoriality Principle

The territoriality principle is so entrenched in domestic criminal law as to feel unremarkable. It is uncontroversial to think that a crime committed in Maine will be policed, prosecuted, and punished in Maine. This Part asks why this approach to criminal law seems straightforward—how, in other words, the conventional wisdom came to be.

Before beginning, it is helpful to introduce my key term. As I explain below, the territoriality principle is a shorthand for two ideas: first, that a government’s power to define and punish crime arises from its authority over a bounded geographic region; and second, that criminal laws apply to any person within that region. The territoriality principle is thus a claim about both the source and the subjects of criminal law. In a territorial criminal law regime, governments with legally defined borders have authority to criminalize conduct within those borders. That authority extends to anyone who is physically present, regardless of her personal traits or place of residence.

Accordingly, it undermines the territoriality principle for state legislatures to criminalize out-of-state conduct. It would also offend the territoriality principle to apply criminal laws only to citizens or residents and not, for instance, to noncitizens and visitors. In the first example, the government extends the reach of criminal law beyond its borders. In the second, the government predicates criminal law on legal status rather than presence on state soil.³⁵

This Article is interested in both kinds of violations of the territoriality norm because both upend traditional assumptions about how domestic criminal law works. My basic claim is that criminal law has become less territorial over time—more expansive, more flexible, and more receptive to forms of jurisdiction based on identity rather than geography. To understand why this development is notable, it helps to begin by asking why territorialism has so much purchase in the classic account of criminal law.

A. Criminal Law

Territorial borders are built into the definitions of American criminal law. In the first weeks of law school, students learn that a crime is an act that happens in a given place as a result of voluntary bodily movement.³⁶ This axiom, known as the voluntary act requirement, embeds a “where” question in the foundation of criminal law. To know that a crime occurred, one must decide where the salient conduct—the actus reus—took place. This is why courts describe venue as “a necessary, if often subtle, element of every criminal statute.”³⁷ At its core, crime is an idea tied to a location.³⁸

Traditionally, the relevant location of a crime is a legally defined territory whose political body has deemed the action criminal. Stating this principle out loud makes it sound complicated, but the idea is familiar. When I commit an assault in Battery Park, what matters is that I am in New York because states define and prosecute assaults. (One could make the same point about federal crimes or city ordinances. The observation is simply that the ambit of domestic criminal law is territorial, and territories have formal legal boundaries.) In other words, when defining crimes, American criminal law generally cares about where you are rather than who you are. The relevant fact is that you are on New York land, not that you are an American citizen, a New York citizen, a senior citizen, or a person who identifies as female, Black, or Jewish. And because of the way criminal law has developed in the United States, the operative “where” is most often the state. Thus, the Model Penal Code provides that criminal law typically extends to any offense committed “within the [s]tate.”³⁹

The territorial scope of criminal law is supposed to be one of the things that makes it distinctive. In a section of the Model Penal Code called “territorial applicability,” the MPC’s reporters explain that criminal law is fundamentally different from civil law because in the criminal context, “jurisdiction and choice of law . . . are merged.”⁴⁰ Again, the principle sounds tricky but feels simple. In a New York criminal court, New York criminal law applies. Whereas in civil actions a court can apply a sister state’s rules, in criminal court the forum state’s law governs.⁴¹ A New York judge cannot choose to apply Connecticut criminal law to your case even if you live in Connecticut or planned your crime there. No, the MPC tells us, “it has long been a maxim of American jurisprudence that a state will not enforce the penal laws of another state.”⁴² So if you committed your crime in New York, you are prosecuted there. And if you committed a crime in Connecticut and drove to New York, you must go back to Connecticut to be prosecuted in the right place.

Conflicts scholars call the prohibition on interstate enforcement of criminal laws “the public law taboo.”⁴³ The intuition driving this taboo is that there is something special about public law, and “penal” law in particular, that makes it offensive and illegitimate to apply a criminal law outside the place it was enacted. It is not entirely clear why penal law is so special—why, for instance, it is more problematic for a state to impose another state’s criminal rules than its civil rules. The answer seems to lie in deeply held (if not especially concrete) beliefs about criminal law’s relationship to concepts like sovereignty and democracy. Later Sections of this Article elaborate and question those beliefs.⁴⁴ For now, the critical point is that criminal law is supposed to be territorial in ways that civil law is not.

To be more precise, domestic criminal law is supposed to be territorial. When it comes to international law, scholars and courts have long recognized several bases for criminal jurisdiction, including forms of extraterritorial jurisdiction in which nations can criminalize conduct by and against their citizens abroad.⁴⁵ For international law scholars, it is not unusual in the least to think that criminal law could be pegged to a personal trait like citizenship status rather than to the place the crime was committed. That idea is also familiar to those who prosecute federal crimes such as terrorism and human trafficking, which often involve extraterritorial conduct and out-of-country defendants.⁴⁶ Territoriality is one option among many for international law theorists and national security lawyers.

But when it comes to the criminal law that governs conduct in the United States—the criminal law under which tens of millions of state prosecutions proceed each year—territorialism is entrenched and rarely questioned. As the Introduction noted, criminal law textbooks omit jurisdiction.⁴⁷ When the subject is discussed, territoriality is understood to be the prevailing rule. In 1909, the Supreme Court announced without fanfare that an act “done within the territorial limits of [one state] . . . cannot be prosecuted and punished by [another state].”⁴⁸ A century later, criminal law philosophers describe territoriality as the principle “we currently observe,”⁴⁹ a central tenet in “classic[] accounts [of] domestic criminal law,”⁵⁰ and a norm that distinguishes criminal from international law, a field in which ideas about jurisdiction are more varied and flexible.⁵¹ A commitment to the territoriality principle is thus thought to be what makes domestic criminal law both domestic and criminal.

To be clear, the territoriality principle is neither a settled doctrine nor an absolute rule. Cases and treatises on criminal law never quite specify what sort of law territoriality is—a doctrine, a practice, or simply an assumption. (This Article is one attempt to solve that mystery.) There are also recognized exceptions to the territoriality principle in state criminal law. The Model Penal Code acknowledges that states can criminalize out-of-state conduct that affects their “legitimate interests,” and the MPC’s reporters take pains to note that the Code sets “forth a number of alternative bases for jurisdiction, thus rejecting the old common law doctrines of strict territoriality.”⁵² But most of those exceptions are territoriality adjacent; the MPC imagines, for example, criminalizing out-of-state conspiracies that involve an overt act “within the State.”⁵³ These are, moreover, exceptions to a baseline rule that domestic criminal law is territorial. That rule guides the MPC and explains the strong presumption against extraterritoriality when interpreting state criminal statutes.⁵⁴

The idea, in short, is that paradigmatic criminal law—the stuff taught in first-year criminal law classrooms—is territorial. Much of this Article is devoted to challenging that idea, to exploring when state criminal law fails to be territorial and how the deterritorialization of domestic law undermines claims about the distinctiveness of criminal law and its special relationship to democracy. These critiques make more sense and land harder if one pauses at the outset to observe just how ingrained territorialism is in Anglo-American criminal law, and state criminal law in particular. On the conventional account, the relationship between criminal law and territory is uncomplicated. Extraterritorial criminal law is exceptional and technical, the sort of material one learns in upper-level courses and specialist subjects like international or national security law. And territorialism is the characteristic that makes criminal law identifiable as a form of domestic (rather than international), criminal (rather than civil), and public (rather than private) law.

B. Constitutional Law

One finds a similar commitment to the territoriality principle in American constitutional law. The federal Constitution contains several provisions that appear to require criminal law to be territorial. The two most obvious are the Venue Clause in Article III, which fixes the place of criminal trials “in the State where the said Crimes shall have been committed,”⁵⁵ and the Vicinage Clause of the Sixth Amendment, which requires juries to be drawn from “the State and district wherein the crime shall have been committed.”⁵⁶ There are distinctions between these two provisions: venue is the location of a criminal trial; vicinage is the catchment area for the jury.⁵⁷ But the clauses are related conceptually and historically, and both are responsible for the strong strain of territorialism that runs through American criminal law.

The Venue Clause was a response to British laws that permitted colonists to be transported to England—“3,000 miles to an alien environment”—for criminal trials.⁵⁸ In a magisterial article published in 1976, Professor Drew Kershen explained that the Constitution’s authors adopted the Venue Clause with two ideas in mind. First, the Founders believed that limiting venue would protect defendants because the “legal and moral support” necessary to mount a defense was most readily available close to home.⁵⁹ Second, and more fundamentally, the drafters espoused a common law theory of criminal jurisdiction in which courts may “enforce criminal laws only with respect to crimes committed within the territory assigned to the court.”⁶⁰

Under this common law theory, venue (the place of a criminal trial) and jurisdiction (the power to hear a criminal case) were one and the same. Courts had no authority to enforce the criminal laws of another place, so of course criminal trials would be held within a court’s territory. This understanding of criminal law—in which venue and jurisdiction are merged—is what early Americans meant when they described criminal law as “local.”⁶¹ To say that criminal law was local was not merely to recount a fact about criminal adjudication. Calling criminal law local was a claim about the source and limits of judicial power.

The Founders expected that the twin goals of the Venue Clause—protecting defendants and constitutionalizing territorial jurisdiction—would align. Because people usually committed crimes near their homes,⁶² the Constitution’s authors assumed that tying venue to the site of a crime would have the incidental benefit of protecting criminal defendants. But when that assumption faltered, the Founders chose territoriality as the touchstone of the Venue Clause.⁶³ Rather than linking venue to a defendant’s residence, the drafters tied venue to the crime’s location. The theory behind this choice was that the government’s authority to enforce criminal laws arises from a power of self-determination that starts and stops at territorial borders—that is, from sovereignty.⁶⁴ From this perspective, the choice between tying venue to a defendant’s residence and predicating venue on a crime’s situs was no choice at all. Venue had to track the crime because the government’s power to enforce criminal law flowed from territory, not from facts about the defendant.⁶⁵

The Venue Clause thus constitutionalized territorialism. The Vicinage Clause did too, though in slightly different ways. The Sixth Amendment’s vicinage requirement began as a proposed amendment to the text of the Constitution, which James Madison introduced to the House in 1789.⁶⁶ Madison’s proposal would have replaced Article III’s Venue Clause with language requiring trials before a jury “of the vicinage” in the county where the crime occurred, except when insurrectionists controlled that county (in which case the trial could be moved nearby).⁶⁷ This proposal morphed several times during the First Congress and ultimately landed in the Bill of Rights rather than Article III.⁶⁸ At a pivotal moment in the process, the Senate rejected the term “vicinage” on the ground that it was “either too vague or too strict”⁶⁹—too vague because it did not refer to a “particular geographical territory recognized as a political or governmental unit,”⁷⁰ and too strict because it might require trials to be held in rebellious counties where juries would not convict.⁷¹

The dispute over the word “vicinage” is illuminating. Supporters of the vicinage provision liked the idea that vicinage could mean “neighborhood” or “community,” loose concepts that captured a preference for localism and a belief that juries were “the conscience of the community” and had a substantive role to play in interpreting criminal laws.⁷² Opponents of the word “vicinage”—the Federalists—were worried about national uniformity and local rebellions, so they wanted a clear term that would not require “neighborhood” juries in every case.⁷³ The final language of the Sixth Amendment was a negotiated settlement between these camps. The Amendment requires juries to be drawn from the “State and district wherein the crime shall have been committed, which district shall have been previously ascertained by law.”⁷⁴

This rule retains and entrenches a basic form of territorialism: it makes the site of a crime the threshold issue when determining the proper criminal procedure. But the Sixth Amendment calls for juries from “states” and “districts” rather than the “vicinage.”⁷⁵ The final vicinage provision is thus not really a vicinage provision. Instead, it summons juries from formal political units with legally defined boundaries, and it refers to districts “ascertained by law,” a phrase that empowers the legislature to determine the jury’s geographic origins within a state. In this respect, the Sixth Amendment was a significant grant of discretion to Congress.⁷⁶ The Amendment also represented the triumph of legal notions of jurisdiction—and Anglo-American legal definitions of territory⁷⁷—over looser and more longstanding conceptions of political community.

One could go on at some length here. The subsequent history of the venue and vicinage provisions is interesting, not least because it involves protracted disputes about what happens to criminal jurisdiction⁷⁸ and federal juries⁷⁹ when Congress tinkers with the boundaries of judicial districts. Part II explores some of that history. Most notably, it examines the emergence in the twentieth century of the idea that jurisdiction and venue are distinct such that the location of a criminal trial is a waivable right possessed by the defendant rather than a structural requirement for the legitimacy of criminal adjudication.⁸⁰ As Part II explains, the fracturing of jurisdiction and venue marks an important moment in the deterritorialization of American criminal law.⁸¹

But at this stage, that history is a detour from the core point: the commitment to territorialism in American criminal law can be traced to the U.S. Constitution. The creators of the federal Constitution treated criminal law as territorial when they debated and drafted the rules on criminal trials. As a result, courts have understood it to be a problem of constitutional significance when a crime is tried outside the place it was committed.⁸² The claim here is not that it is unconstitutional to try crimes extraterritorially. The whole purpose of this Article is to show that the legality of extraterritorial criminal law has shifted over time. Rather, the observation is that we tend to think domestic criminal law is territorial because the Constitution’s drafters assumed it had to be.

State constitutions contain the same background assumption. The Sixth Amendment vicinage right is one of the few that is not incorporated,⁸³ but most state constitutions have Sixth Amendment analogues that build territoriality into the criminal legal process.⁸⁴ Montana, for example, requires criminal trials to be held in “the county or district” where the crime occurred, before a jury from the same area.⁸⁵ Massachusetts limits criminal prosecutions to “the vicinity where [the facts] happen.”⁸⁶ In Louisiana, criminal trials must be in the “parish where the offense or an element of the offense occurred.”⁸⁷ There are distinctions between these provisions, but they all reflect and reinforce a territorial understanding of criminal law. When searching for the origins of the territoriality principle, constitutional rules on criminal trials prove to be a critical source.

Territorialism appears in other parts of the federal Constitution too. The Extradition Clause obliges states to “deliver[] up” alleged criminals who “flee from Justice.”⁸⁸ This Clause, which immediately prompted disputes about the rendition of fugitive slaves,⁸⁹ requires interstate cooperation in criminal law enforcement. More to the point, it presumes that cooperation means sending fugitives back rather than trying them in the state where they are found. One could imagine a legal regime in which states authorized each other to enforce their criminal codes—so, for example, Georgia would say, “Massachusetts, try that defendant for me,” rather than, “Send him back to my criminal courts.” But that is not how the U.S. Constitution works. Because early Americans had sharp disagreements over the criminal laws used to regulate slavery,⁹⁰ and because the Founders assumed the common law conception of criminal jurisdiction,⁹¹ the Constitution provides for extradition. As a result, when people “flee from Justice,”⁹² we ship them back to the place where justice is supposed to happen.

American courts have also read territorialism into the Due Process Clause.⁹³ In cases involving crimes that cross state lines—for example, a cross-border shooting or a theft in one state where stolen goods wind up in another—courts have suggested that the Due Process Clause imposes an outer limit on extraterritorial prosecutions.⁹⁴ As one New York court described the doctrine, the Fourteenth Amendment prohibits states from “overreaching” when they “exercise [criminal] jurisdiction over out-of-state events.”⁹⁵ Part II explains why this doctrine is not especially restrictive in practice,⁹⁶ but here the invocation of the Constitution is what matters. In due process cases, as in venue and vicinage cases, courts believe it to be a constitutional problem when state criminal law exceeds state borders.

Together, these examples portray a constitutional order wedded to territorialism. Indeed, they suggest that territorial criminal law was a critical part of the project of constituting the United States. The rules on extradition and criminal procedure in American constitutions are not just rules of the road for efficient trials. These rules are how states announced their borders and enacted their power to govern within them. (You can really tell you have a state once its police arrest and imprison you.) Early Americans used criminal law to instantiate state power and to navigate interstate relations, including disputes over slavery. From this perspective, the territoriality principle is bound up with the history of American federalism and reflects foundational ideas about what it means to be a union of sovereign states.

On a more prosaic level, the territoriality principle clearly has roots in constitutional law. The previous Section explored how territoriality is ingrained in substantive criminal law, in concepts like the actus reus requirement and the prohibition on interstate enforcement of “penal” laws. The observations in that Section apply broadly to criminal law in common law regimes.⁹⁷ Here, we might add that territorialism is also distinctively American. The country’s federal and state constitutions contain clauses mandating territorial criminal procedures, limiting extraterritorial criminal jurisdiction, and requiring states to facilitate territorial criminal law enforcement. These provisions are one key reason we tend to think criminal law must be tied to a particular place.

C. Institutional Design

Territorialism is also built into the institutional design of the criminal legal system. The Federal Rules of Criminal Procedure require crimes to be prosecuted “in a district where the offense was committed” unless narrow exceptions apply.⁹⁸ State criminal procedure rules mirror this requirement, usually mandating prosecution in the county or district where a crime occurs.⁹⁹ This summary glosses over some important variation—for instance, states use different words to limit venue, which, as the “vicinage” debate suggests,¹⁰⁰ reflects competing views about the values protected by criminal procedure. Technically, state venue rules fall along a spectrum from more to less strictly territorial. But they all embrace a baseline territorialism in which criminal procedure hinges on the site of a crime.

Criminal law enforcement is organized around territory as well. Typically, police are clumped into territorial units, such as the New York or Los Angeles Police Departments. (The Bergen County Sheriff’s Office, the New Jersey State Police, and the Federal Bureau of Investigation also work as examples here. All of these law enforcement agencies are defined by geography and empowered to act within some legally bounded area.) The same is true of state prisons and county jails, which have names like Ohio State Penitentiary and Cook County Jail and receive funding through county and state corrections budgets.¹⁰¹ Each of these law enforcement units purports to act on behalf of a certain place.

The United States also has a complex interstate extradition apparatus, which developed from the Extradition Clause.¹⁰² Before a person can be sent to face criminal charges in another state, he must go through an elaborate legal process that involves written notices, communication between state governors, approval from secretaries of state, and hearings before an Article III court.¹⁰³ In Pennsylvania, for example, interstate extradition requires two warrants, three hearings, and twenty-two distinct steps.¹⁰⁴ Defendants have a federal constitutional right to this process, which they can sue to enforce.¹⁰⁵

All of which is to say there is a considerable edifice built up around the idea that criminal law is territorial. American governments have organized their law enforcement bureaucracies, allocated resources, and recognized enforceable rights to effectuate the presumption that criminal law has territorial boundaries. Put simply, territorialism is shot through the entire system. In addition to being a foundational premise of substantive and constitutional law, the territoriality principle is a basic feature of how criminal law is operationalized in the United States.

And then there is the practice of punishing noncitizens, which is perhaps the clearest example of territorialism in American criminal law. The United States has always subjected noncitizens within its borders to its criminal laws. Foreign nationals have been prosecuted in state and federal criminal courts—and imprisoned in state and federal prisons—for as long as those institutions have existed.¹⁰⁶ The assumption behind this practice is that criminal law attaches as soon as a person enters the country whether or not she is a legal member of the polity. This assumption is usually implicit, though when courts need a citation they turn to the discussion of states’ police power in Mayor of New York v. Miln, in which the Supreme Court observed that “[t]he right to punish, or to prevent crime, does in no degree depend upon the citizenship of the party who is obnoxious to the law.”¹⁰⁷ Rather, the Miln Court explained, a state has criminal jurisdiction “over all persons . . . within its territorial limits.”¹⁰⁸ Thus, the “alien who shall just have set his foot upon the soil of the state, is just as subject to the operation of the law, as one who is a native citizen.”¹⁰⁹

The idea that a person who is not a citizen can nonetheless be subjected to criminal law because the right to punish arises from “soil” is the essence of the territoriality principle. In a regime based on this principle, criminal law is egalitarian (because it is blind to personal traits like citizenship status) and state-building (because it depends on and reaffirms state borders). Later Parts of this Article discuss these features of territorial criminal law and consider whether they are desirable. The goal here is simply to observe that territorialism is embedded in the institutional design of American criminal justice. Though it often goes unremarked, the territoriality principle shapes everything from the organization and funding of police departments to the criminal prosecution of hundreds of thousands of noncitizens each year.¹¹⁰

D. Rhetoric

Territoriality is also a prominent theme in constitutional case law. Constitutional opinions are full of rhetoric about the inherently local nature of criminal law. In the Rehnquist era, the Supreme Court invoked the localness of criminal law to restrict congressional power under the Commerce Clause.¹¹¹ In abstention and habeas cases, the Court has cited the local nature of criminal law to explain why federal courts ought not police state criminal proceedings.¹¹² Claims about criminal law’s localness also surface in Eighth¹¹³ and Fourteenth¹¹⁴ Amendment cases, in which the Supreme Court has upheld harsh sentences and prosecution-friendly procedural rules on the ground that criminal law is local and “belongs exclusively”¹¹⁵ to states.

There is enormous nuance in these lines of doctrine. Each forms its own field of study. But grouping them together demonstrates just how much work rhetoric about the localness of criminal law is doing in constitutional jurisprudence. The idea that criminal law is local—naturally, fundamentally, traditionally, essentially—runs throughout constitutional thought. This idea has been cited to support a state-centric conception of American federalism, a restrictive view of criminal defendants’ rights, and limits on the Article III docket (which in turn maintain the prestige of federal courts).¹¹⁶ The localness of criminal law is, in other words, part of the foundation on which American constitutionalism is built.

Courts are not especially clear about what it means for criminal law to be inherently local. Usually, the claim simply seems to mean that the federal government should let states manage an issue. But as the previous Section explained, the assertion that criminal law is local has a rich history: it refers to an eighteenth-century conception of criminal jurisdiction in which the power to define crimes and enforce criminal laws was limited to a particular territory.¹¹⁷ At the founding, “crimes [were] considered ‘local’ in nature, i.e., local to the territory of the enacting sovereign and local to the territory of the enforcing court.”¹¹⁸ Modern references to the “essential” and “traditional” localness of criminal law call forth this early American understanding of criminal jurisdiction, which was premised on the territoriality principle. Thus, when courts say criminal law is local, they are in an important sense saying that criminal law is territorial.

This is yet another way in which territorialism lurks beneath American law. The constitutional cases described above rarely include discussions of the proper scope of criminal jurisdiction. These are cases about the Commerce Clause, due process, and debt collection.¹¹⁹ But they rely on rhetoric about the territoriality of criminal law and indeed gain some of their force from the looseness of that rhetoric. As we will see in Part II, the claim that criminal law is quintessentially local becomes less persuasive the more one examines it. But as a truism, this claim functions powerfully to support a slew of different constitutional doctrines. The territoriality principle is prevalent and potent in the background of constitutional law as an unquestioned “maxim of American jurisprudence.”¹²⁰ Its presence in so many corners of constitutional doctrine helps to explain why it seems so strange to think that New York could criminalize gambling in Nevada or authorize Missouri to prosecute violations of the New York criminal code.

II. The Erosion of the Territoriality Principle

After reading Part I, the territoriality principle should seem central to domestic criminal law, perhaps even required by the Constitution. Many aspects of the American legal system, from the basic definition of crime to the organization of criminal justice institutions, work together to make criminal law feel naturally and necessarily territorial. Yet, over the past century, criminal law’s commitment to the territoriality principle has faltered.

This Part traces the trend away from territoriality in domestic criminal law. It depicts a very different criminal justice system than we saw in Part I. The legal regime outlined above was defined by geographic borders and was particularly committed to state lines. The one below is flexible, negotiable, and remarkably unconcerned about territorial limits on the police power.

To trace this system’s emergence, this Part proceeds chronologically through the criminal legal process, from criminalization to policing, prosecution, and then punishment. In some ways, this approach is atypical. To the extent that legal academics discuss criminal jurisdiction, they tend to focus exclusively on criminalization and specifically on whether legislatures can criminalize extraterritorial acts.¹²¹ This Part departs from that norm by examining both criminalization and criminal law enforcement. It thus explores jurisdiction not just in the narrow terms of legislative intent but—to borrow the Supreme Court’s phrase—in “its popular sense of authority to apply the law to the acts of men.”¹²²

I take this approach because “the criminal law” is more than the set of liability rules created by legislatures. One of the core insights of criminal law scholarship of the past thirty years is that, given the extraordinary scope of the criminal code, the real substance of criminal law is made by law enforcers—by the police and prosecutors who decide which offenses are actually criminal.¹²³ From this perspective, it misses the point to study only criminalization. To know if criminal law is territorial, one has to ask not only what conduct legislatures decide to criminalize but also which crimes courts claim the power to adjudicate, when police can cross borders, where criminal prosecutions happen, and when punishment is tied to the place a crime occurred. The answer, it turns out, is that territorialism has eroded at every turn.

A. Criminalization

Criminal law begins when an authoritative legal body, typically a legislature,¹²⁴ decides that some conduct is criminal. At this stage of the criminal legal process, the key question is how far a criminal law reaches. Theorists call this the question of criminal law’s “geographical ambit.”¹²⁵ (So to review: venue is where a criminal trial happens, vicinage is where the jury comes from, and ambit is where the substantive criminal law applies.) As Part I noted, the traditional view is that criminal statutes reach offenses within some defined territory, like New Jersey or Pennsylvania.¹²⁶ But legislatures and courts have expanded criminal law beyond these boundaries in several ways.

First, they have redefined crimes to stretch across multiple places. Take the continuing offense doctrine, under which crimes such as fraud and possession of contraband continue to occur over time “and, possibly, in a number of different places”¹²⁷ as a person moves around. Or consider inchoate offenses like attempt and conspiracy, whose definitions and relationship to the actus reus requirement are notoriously fluid.¹²⁸ Because continuing offenses travel with a person and the precise location of an inchoate offense is difficult to pin down, these crimes have an attenuated relationship to geography. (This problem is exacerbated when coconspirators can be charged for each other’s actions and tried wherever an accomplice acted.)¹²⁹ Ongoing and inchoate crimes make up a significant portion of the modern criminal docket.¹³⁰ Often, they can be prosecuted in several different places.

Continuing and inchoate crimes are territorial insofar as a court must determine where they occurred before a criminal prosecution can proceed. Asking where a conspiracy or an ongoing theft took place is a territorial question. But in modern criminal law, the answer to that question is often a legal fiction based on artificial extensions of time and concepts like constructive presence.¹³¹ The approach to territoriality when conceptualizing these crimes is a far cry from the eighteenth-century understanding of criminal jurisdiction described in Part I.¹³² Ongoing and inchoate offenses are at best quasi-territorial—which is why, when legislatures invented continuing offenses in the early 1900s, defense attorneys argued that they violated the Venue Clause and the Sixth Amendment.¹³³ That argument resurfaced in the 1970s, when critics protested that the very “idea of a continuing offense must be declared unconstitutional.”¹³⁴ This objection has never been successful, but its persistence illustrates just how much ongoing crimes undermine the territoriality principle.

Courts have also relaxed territorial rules by expanding their own jurisdiction over out-of-state conduct. Think here about a Texas criminal case in which prosecutors charged someone who forged a deed for Texas property while in Louisiana¹³⁵ or a California court adjudicating the criminal prosecution of a Colorado-based doctor who prescribed pills that someone later ingested in California.¹³⁶ In these sorts of cases, criminal prosecutions are based on the in-state effects of out-of-state activity. American courts first began to recognize this species of criminal jurisdiction in the 1860s,¹³⁷ and the Supreme Court upheld effects-based jurisdiction in 1911.¹³⁸ In embracing the effects doctrine, Justice Holmes explained that a “civilized world” required a flexible approach to the territoriality principle, so that even a criminal who had “never had set foot in the State” could be punished if his conduct caused in-state harm.¹³⁹ To put this precedent into some historical context: at the turn of the twentieth century, as the federal government began to take shape and new forms of transportation enabled mass mobility,¹⁴⁰ courts grew bolder about their power over offenses committed outside the state.

The Constitution was no barrier to this development. As Part I explained, the Due Process Clause is supposed to prevent states from “overreaching” when they exercise criminal jurisdiction over extraterritorial activity.¹⁴¹ But one can search in vain for a meaningful due process limit on criminal prosecution of out-of-state conduct. Even the New York court that boldly denounced jurisdictional “overreaching” was eventually overturned.¹⁴² In practice, any effect inside a state—harm to a state resident, a piece of state property, and so on—can ground a state criminal prosecution.¹⁴³ Again, this approach to criminal law is basically territorial in that courts must conjure up some connection to state soil to proceed. But the invention and expansion of effects-based jurisdiction loosened nineteenth-century assumptions about the limits of judicial power over criminal cases.

Courts have also developed nonterritorial theories of criminal jurisdiction. The most significant case in this category is Skiriotes v. Florida, a 1941 Supreme Court precedent.¹⁴⁴ Skiriotes arose when a Florida resident named Lambiris Skiriotes went sponge fishing in the Gulf of Mexico.¹⁴⁵ The Sheriff of Pinellas County arrested Skiriotes under a state law that banned sponge fishing in Florida; he was then prosecuted and convicted under state law.¹⁴⁶ The Supreme Court upheld the conviction on the ground that Florida could “govern the conduct of its citizens upon the high seas with respect to matters in which the State has a legitimate interest”—specifically, Florida’s “interest in the proper maintenance of [a] sponge fishery.”¹⁴⁷ The Court thus concluded that Florida could predicate its criminal laws on state citizenship rather than state territory.

Skiriotes represents a remarkable shift away from the territoriality principle. The Supreme Court has never clarified the case’s outer boundaries, and because the defendant was in international waters, it is not clear what would happen if Florida criminalized its residents’ behavior in Georgia or Alabama.¹⁴⁸ While scholars have opined that such a statute could implicate “multiple complicated doctrines”—including due process, the Sixth Amendment, the Dormant Commerce Clause, and the right to travel¹⁴⁹—Skiriotes established that criminal law can attach by virtue of citizenship rather than soil. If there is a limit on the ambit of criminal law, it will have to come from as-yet-unannounced constitutional rules, not a requirement that criminal law must be territorial.
Skiriotes is the most prominent example of deterritorialized domestic criminal law. But the basic proposition in the case is that state criminal law can proceed without a clear territorial hook. This precept guides more run-of-the-mill cases too. Sometimes, for example, courts ground criminal jurisdiction on a state’s general interests in some industry or value rather than on the effects of a crime. In 2017, Florida invoked its interest in the cruise ship industry to try an attempted sexual assault between two noncitizens at sea.¹⁵⁰ Alaska has succeeded with similar prosecutions undertaken in the name of the state’s “tourism industry.”¹⁵¹ These sorts of cases extend the theory of effects-based criminal jurisdiction to include abstract harms to a state’s economic and political order. At that point, territoriality has more or less disappeared.

Courts have also extended the scope of state criminal law by relaxing the rules around extradition. Part I outlined the development of an elaborate legal apparatus to facilitate interstate extradition.¹⁵² Initially, state governments used that apparatus to regulate and denounce each other’s criminal legal systems. In 1860, Ohio’s governor refused to extradite a man who helped to free a slave in Kentucky.¹⁵³ In the 1930s, New Jersey’s governor declined to extradite Robert Elliott Burns after reading his memoir, I Am a Fugitive from a Georgia Chain Gang!.¹⁵⁴ In 1975, the governor of Tennessee ignored Oklahoma’s request to extradite country singer Faron Young, who was “wanted on a morals charge.”¹⁵⁵ Until the 1970s, state courts policed extradition too, denying requests when they felt a sister state had failed to show probable cause.¹⁵⁶ The Supreme Court put a stop to that practice in 1978 when it held that state courts may not opine on the sufficiency of extradition requests.¹⁵⁷ Instead, the Court emphasized, judicial approval of extradition is “summary and mandatory.”¹⁵⁸

This development in extradition doctrine undermined the territoriality principle. As Part I pointed out, extradition is quintessentially territorial; it is the practice of sending fugitives back to the “right” state to be tried. Extradition is criminal law’s substitute for choice of law and in that respect is the lynchpin of territorialism. But making extradition mandatory weakens the territoriality principle, because it means that the borders of state criminal law matter less. Imagine two countries: one that refuses to extradite a criminal and one where extradition is routine. Territorial borders are more meaningful in the first country—the one where you are safe once you touch its soil. The scope of criminal law feels more territorial there too, since the offended country’s laws cannot reach you.

In other words, denying extradition is one way to make clear that criminal law has geographic limits. By requiring “summary and mandatory” extradition between states, the Supreme Court has ensured that Arizona’s criminal laws reach New Hampshire, and vice versa. The Court made this choice explicitly. In Michigan v. Doran, a leading extradition case, Chief Justice Burger explained that mandatory extradition “foster[s] national unity” by preventing any state from becoming “a sanctuary for fugitives” and “‘balkaniz[ing]’ the administration of criminal justice.”¹⁵⁹ Writing for the Court, Chief Justice Burger articulated a national vision of criminal law: “In the administration of justice, no less than in trade and commerce,” unity is “served by de-emphasizing state lines.”¹⁶⁰ Modern extradition doctrine thus extends the reach of state criminal laws to the edges of the nation. As a practical matter, summary extradition means that it is not particularly difficult to prosecute someone who flees across a state border.¹⁶¹

There are important technical distinctions between the examples in this Section. Conceptually, there is a difference between a legislature’s power to deem conduct criminal (legislative jurisdiction) and a court’s power to adjudicate criminal cases (judicial jurisdiction). The former power is what the term “criminalization” calls to mind, while the exercise of judicial jurisdiction is rarely described as an instance of criminalization.¹⁶² Both types of jurisdiction are different from questions about the functional reach of Maine or Texas criminal law given modern extradition rules.

In a course on criminal jurisdiction, one would parse these distinctions. But here the salient observation is that a range of doctrinal shifts have come together to extend the reach of domestic criminal law. When legislatures define crimes to cover extraterritorial conduct and courts exert jurisdiction over out-of-state activities, both acts expand state criminal law beyond its traditional boundaries. The result is a corpus of criminal law that is detached from territorial borders—not wholly, but more than one might expect.

B. Policing

The deterritorialization of American criminal law is even more pronounced in law enforcement. As Part I noted, police are typically organized into territorial units. Police officers wear badges engraved with city insignia¹⁶³ and swear oaths to state constitutions.¹⁶⁴ Yet, in practice, cooperative agreements and permissive doctrines mean policing is only loosely tied to boundary lines.

Sometimes the deterritorialization of policing is formal, as when police departments sign agreements to provide “mutual aid.”¹⁶⁵ Many states permit local law enforcement agencies to enter partnerships under which they can police each other’s territory.¹⁶⁶ These mutual aid agreements, which are essentially private contracts between police departments, create standing commitments to share resources and jointly enforce criminal laws.¹⁶⁷ Police departments also enter into memoranda of understanding (MOUs) and more informal agreements to combine their powers.¹⁶⁸ These contracts expand the footprint of local police departments—or, as one 1988 white paper put it, mutual aid agreements “knit jurisdictions together.”¹⁶⁹

The standard mutual aid agreement merges the staff and resources of two small, neighboring police departments within a single state.¹⁷⁰ But mutual aid compacts can extend “local” policing much further. In some states, municipal police can contract to provide services outside the state. In Nebraska, for example, “[a]ny city or village” can authorize its police to assist another state’s police force.¹⁷¹ Local police departments can also join multijurisdictional task forces that span state boundaries.¹⁷² These task forces are designed to overcome territorial jurisdictional limits. They are a mechanism to blur legal boundaries so that “city and suburban cops in Illinois and Indiana” can “work[] the case together.”¹⁷³

At a high level of generality, cooperative policing is nothing new. American police departments have always shared equipment and “backup forces,” particularly during emergencies.¹⁷⁴ But police sharing practices were interlocal and “strictly informal” until the 1930s, when increased mobility prompted a movement to expand and institutionalize interjurisdictional policing.¹⁷⁵ In 1934, citing concerns about “the interstate nature of crime and the growing complexity of law enforcement,” Congress passed the Crime Prevention Compact Act (CPCA).¹⁷⁶ That statute authorized states to enter compacts “for cooperative effort and mutual assistance in the prevention of crime and in the enforcement of their respective criminal laws and policies.”¹⁷⁷ The CPCA encouraged states to establish “such agencies, joint or otherwise, as they may deem desirable for making [cooperative policing] effective.”¹⁷⁸

A flurry of state laws followed. In the wake of the CPCA, states passed statutes enabling cooperative policing and establishing interjurisdictional criminal task forces.¹⁷⁹ States also began to enter regional compacts that provided for joint policing and created centralized “criminal intelligence bureau[s]” to gather and disseminate information about crime.¹⁸⁰ This cooperative model flourished during World War II as local law enforcement agencies coordinated their training programs and emergency plans for “civil defense.”¹⁸¹

Cooperative policing then expanded in the 1960s, when police turned to mutual aid agreements to manage civil unrest, Vietnam protests, and “rock festivals.”¹⁸² Between 1960 and 1972, twenty-five states and the District of Columbia passed laws enabling interjurisdictional policing.¹⁸³ By the 1980s, federal commissions were studying how best to coordinate local police forces,¹⁸⁴ and the federal government was awarding grants to encourage multijurisdictional task forces.¹⁸⁵ Federal funds for these local partnerships ballooned after September 11, 2001, as police associations argued that only an “interjurisdictional enforcement approach” could “address the threats of international and domestic terrorism.”¹⁸⁶

The rise of cooperative policing deserves its own separate study. The birth and expansion of mutual aid agreements is a fascinating case of subfederal coordination and bureaucratic state-building. This model of policing provides a counternarrative to the dominant account of twentieth-century criminal law, which emphasizes the proliferation of federal criminal laws and the emergence of a federal enforcement bureaucracy that (the story goes) displaced state and local police.¹⁸⁷ Mutual aid agreements complicate this federalization story. They show that criminal law expanded during the twentieth century not just because the federal government grew but also because—in the shadow of a growing federal government, and perhaps in resistance to it—states and localities formed regional unions that detached criminal law from territorial borders and upended nineteenth-century ideas about criminal jurisdiction.

This history is nuanced, but even the quick version illustrates that modern policing is significantly less tied to jurisdictional boundaries than the phrase “local police” suggests. Over the course of the twentieth century, contracts to provide emergency aid became broader agreements to combine both the resources and the legal authority of local police forces. Today, mutual aid agreements are routine.

Sometimes, moreover, agreements are not even necessary. Although compacts and task forces are particularly clear examples of interjurisdictional policing, local law enforcement is deterritorialized in other ways as well. Some jurisdictions, for example, empower nominally local police to act anywhere in the state.¹⁸⁸ In California, a peace officer’s authority extends to any offense he witnesses;¹⁸⁹ in Connecticut, “active members of any . . . police force in a town, city, or borough” can execute warrants statewide.¹⁹⁰ Texas permits out-of-state police officers to enter its territory to make arrests for felonies committed elsewhere.¹⁹¹ Tennessee has a “special enforcement unit” focused on Medicaid fraud, whose police routinely arrest people “in Alabama, Mississippi, and Georgia.”¹⁹² These practices abrade territorial jurisdictional boundaries.

Most states also have “citizen’s arrest” statutes that allow private citizens to apprehend alleged criminals.¹⁹³ Courts have relied on these statutes, which are controversial relics of the Reconstruction era,¹⁹⁴ to uphold arrests made by police officers outside of their territorial jurisdiction.¹⁹⁵ Some states even relax “the requirements for a citizen’s arrest” when the citizen in question is a police officer—a doctrine that turns police into sort of roving super-citizens.¹⁹⁶ Courts use “hot pursuit” doctrines to a similar effect.¹⁹⁷ These doctrines create exceptions to “the territorial limits of . . . [police] jurisdiction” and expand the scope of criminal law enforcement.¹⁹⁸ They do so deliberately. In adopting a wide interpretation of the state’s hot pursuit law in 1991, the Pennsylvania Supreme Court explained that “constructing impenetrable jurisdictional walls benefit[s] only the criminals hidden in their shadows.”¹⁹⁹

One could go on and on with gripping stories of extraterritorial policing—of late-night arrests outside city limits²⁰⁰ and NYPD raids in Newark.²⁰¹ Together, these examples demonstrate that the borders of police jurisdiction are negotiable. Territorial jurisdiction is not meaningless; the technical limits of police authority still matter to police, to defendants, and to some courts. But mutual aid agreements, task forces, compacts, citizen’s arrest statutes, and hot pursuit doctrines are all legal devices that detach policing from geography. In practice, the insignia on police uniforms belie a nuanced, interjurisdictional reality in which “local” and “state” police have more than local and state power.

C. Prosecution

It is no surprise that police cross borders. Policing is, after all, an infamously discretionary activity.²⁰² But one would expect prosecution to be different. Given the Constitution’s focus on criminal trials—the Venue Clause, the Vicinage Clause, all the debate over criminal trials at the founding²⁰³—it would seem as if prosecution must have strict geographic boundaries.²⁰⁴ Even at this phase of the criminal legal process, though, territorialism has receded.

Several historical developments drove the deterritorialization of prosecution. First, in the late nineteenth century, courts began to distinguish jurisdiction from venue.²⁰⁵ As Part I explained, early American legislators and courts adopted a common law theory of criminal jurisdiction in which the power to hear a criminal case and the location of a criminal trial were coextensive because crime was “inherently local.”²⁰⁶ That conception of criminal jurisdiction started to splinter in the 1870s as Congress began to separate federal judicial districts into smaller divisions.²⁰⁷ The subdivision of judicial districts raised a new question: Could federal trial courts adjudicate criminal cases involving crimes in another division within the same district? The Supreme Court answered that question in the affirmative in a trio of cases between 1892 and 1897.²⁰⁸ Those cases began to carve out a distinction between the power to conduct criminal proceedings and the proper place for a criminal trial.

That distinction deepened between 1925 and 1931, when appellate courts—first the Fifth Circuit,²⁰⁹ then the D.C. Circuit²¹⁰—held that venue was a criminal procedure right that could be waived. Once venue was a right that belonged to criminal defendants rather than a limit on judicial power or a requirement for the legitimacy of criminal trials, the separation of jurisdiction and venue was “complete.”²¹¹ Defendants could agree to move trials (or simply fail to make a timely venue objection),²¹² and so long as the court could find a hook for its jurisdiction, the criminal prosecution could proceed.

Recall, moreover, that the Supreme Court expanded the bases for criminal jurisdiction to include the effects of a crime in 1911.²¹³ As the previous Section explained, courts began to adopt a bolder and more flexible approach to judicial power over out-of-state crime during the progressive era. Alongside that doctrinal shift, venue’s transformation into a criminal procedure right meant that criminal trials could be prosecuted in a wider range of places. By the middle of the twentieth century, venue was a right that defendants could forfeit or waive. Jurisdiction was a separate matter, satisfied so long as a crime caused harm or implicated an interest in the place where the court sat.²¹⁴

This new approach to criminal jurisdiction enabled a second development: the spread of special venue laws. Once courts had abandoned strict territorial ideas about their power in criminal cases, legislatures began to pass more laws dislocating criminal prosecution.²¹⁵ These laws take several forms. The most common is the “specific” venue statute, which designates the place of trial for a certain type of offense. Specific venue statutes can provide for venue in a particular court or in multiple places—for example, anywhere a stolen item is taken, carried, or found.²¹⁶ Their close cousin, the “buffer” statute, allows prosecutors to choose between neighboring counties when crimes occur in multiple jurisdictions or near a county boundary line.²¹⁷ State laws and procedural rules can also allow prosecutions to proceed where a person is arrested or resides.²¹⁸ These laws are exceptional; as Part I explained, venue is usually proper in the county where the crime occurs.²¹⁹ But special venue statutes demonstrate that, in modern criminal law, legislatures can move trials where they wish. (Courts participate here too, by treating venue defects as harmless.)²²⁰

Prosecution also grew more coordinated and centralized during the twentieth century. Like police, prosecutors can join task forces to investigate and try offenses that span formal boundary lines. In New York City, for example, a “Special Narcotics Prosecutor” appointed by the district attorneys from the city’s five boroughs can prosecute drug and related crimes across the city.²²¹ The state created this interjurisdictional office in 1971 “to address [the] free flow of narcotics across county lines.”²²² Illinois began to establish similar “multijurisdictional drug prosecution units” in the early 1980s.²²³ California uses an analogous model for identity theft crimes, which can be combined with “all associated offenses,” consolidated, and tried in the county of the prosecutor’s choice.²²⁴

It is important not to overstate the trend toward centralized prosecution. Criminal prosecution of state law remains one of the most localized and varied parts of the American justice system.²²⁵ States routinely delegate prosecutorial power to local actors, who often act without centralized oversight.²²⁶ The impression here should not be one of a highly consolidated legal regime. But there is no question that technological developments and relaxed conceptions of criminal jurisdiction encouraged cooperation between prosecutors and spurred the creation of interjurisdictional prosecution offices.²²⁷ As with special venue statutes, these innovations depart from the territoriality norm and permit a measure of forum shopping in criminal prosecution.²²⁸

Finally, and perhaps most important, the twentieth century witnessed the rise of the plea bargain. The exponential growth of negotiated pleas is one of the most documented and lamented features of the American criminal justice system.²²⁹ Today, plea bargains account for almost all criminal convictions in the United States.²³⁰ Professor John Langbein famously observed that this system “recapitulate[s] much of the doctrinal folly of the law of torture.”²³¹ Plea bargaining also displaces the jury, which transforms vicinage into a rather weak criminal procedure right. Ironically, given the Founders’ fixation on juries, the Constitution requires territoriality at the least important moment of the modern criminal process.

These four developments—the splintering of jurisdiction and venue, the proliferation of special venue rules, the creation of multijurisdictional prosecution units, and the plea-bargaining revolution—mean that even prosecution is not as territorial as one would think. In the modern criminal justice system, a court’s power to hear a case is different than where it may be heard. Legislatures can designate special places for criminal trials. Prosecutors can forum shop. Venue defects are harmless errors. Plea bargains have largely undermined the relevance of the territorial limits on trials and juries. The cumulative result is a criminal justice system in which even prosecution, the lodestar of constitutional criminal procedure, is less territorial than it used to be.

D. Punishment

Finally, there is punishment. The last phase of the criminal legal process is the least tied to territorial boundaries. Take imprisonment, the most well-known criminal sanction. American law does not require people to be imprisoned in the jurisdiction whose laws they violated, nor do prisoners need to be held near the site of their crime.²³² Prison officials may send federal prisoners across the country, no matter where they were convicted. State prisoners can also be transferred, sometimes thousands of miles, to serve their sentence in another state’s prisons.²³³ And noncitizens convicted in American courts can be repatriated to their country of origin—so a Massachusetts or Texas prison sentence can lead to prison time in Australia or Germany.²³⁴

My previous research has explored the history of these prisoner transfer systems. As that work explains, the prison was originally conceived in the late eighteenth century as a territorial alternative to transportation punishment, the reigning noncapital sanction at the time.²³⁵ Early Americans believed imprisonment was a local and therefore more democratic alternative to transportation, a “monarchical” sanction used by “kings and despots.”²³⁶ (One can hear echoes of how the term “local” was used in early debates about criminal jurisdiction here.)²³⁷ The first American prison systems were built on this “positive republican theory of crime,” in which territorial punishment connected prisons to the polity that enacted criminal laws and promised to reintegrate prisoners into that polity upon release.²³⁸

This philosophy of punishment ebbed with the rise of the administrative state and the professionalization of a prison bureaucracy.²³⁹ As prison systems expanded in the 1940s and 1950s, state prison officials began to sign contracts to share prisoners and prison bed space.²⁴⁰ Those contracts turned into regional compacts and eventually a national prison network in which states can send their prisoners across the country.²⁴¹ The Supreme Court upheld interstate punishment in 1983 when it ruled that prisoners have no due process right to in-state confinement, even if prison transfers “involve[] long distances and an ocean crossing.”²⁴² Thus, between 1930 and 1990, it became both normal and legal to punish prisoners far from the site of their crime.

International prisoner repatriation proceeded along a similar timeline. In the mid-1970s, after high-profile news reports and congressional hearings on the treatment of Americans imprisoned abroad, the United States entered its first treaty for international prisoner transfers, a bilateral agreement with Mexico.²⁴³ A series of other treaties followed, and by 1985 the United States was a party to agreements providing for repatriation to and from sixty-seven countries.²⁴⁴ That number has since grown to more than one hundred.²⁴⁵ Under the terms of these treaties, the United States can repatriate both federal and state prisoners to serve prison sentences in their countries of origin. The United States can also receive and “resentence” Americans convicted abroad,²⁴⁶ though in practice a treaty regime born for that purpose is used mainly to export noncitizens out of the country.²⁴⁷

Interestingly, the systems for both interstate and international prison transfers are underused.²⁴⁸ Prisoners are outsourced less than the law would permit—a testament to the durability of the territoriality norm.²⁴⁹ But the emergence of prisoner transfer systems represents a notable shift away from the belief that punishment has to be local to be lawful or democratic. Over the course of the twentieth century, courts and prison bureaucrats concluded that punishment need not be connected to the place where criminal law is made.

By now, the arc of this narrative should feel predictable. The history of American imprisonment parallels the deterritorialization of policing and prosecution, with glimmers of interjurisdictional cooperation in the early twentieth century giving way to more systematic coordination by the 1970s. In each of these contexts, increased mobility and state capacity put pressure on the territoriality principle and encouraged more creative—which is to say extraterritorial—forms of criminal law enforcement.

Once again, moreover, the Due Process Clause was no real impediment to the expansion of criminal law.²⁵⁰ The Supreme Court has never policed the location of prisons, and the Court explicitly declined to constitutionalize prison placement in the early 1980s.²⁵¹ As a result, constitutional law imposes very few limits on where prisoners may be held. American prison officials can move prisoners all over the country (in some cases all over the world), and the site of a crime is barely related to where punishment occurs. Instead, personal facts about a prisoner—his sentence length, disciplinary record, family’s location, citizenship status—determine where he will be confined.²⁵² This is a legal regime that has long since given up on the idea that punishment must be local to be legitimate.

E. Post-Territorial Criminal Law

In the end, then, every stage of criminal law has been deterritorialized. This transformation has been uneven. Some parts of criminal procedure, such as jury trials, remain more territorial than others. If asked to rank the system by its commitment to borders, the right response is probably that criminalization is the most territorial phase of American criminal law, followed by prosecution, policing, and then punishment. To be even more granular: legislative jurisdiction is slightly more territorial than judicial jurisdiction; both are a bit more territorial than prosecution; and prosecution has clearer geographic limits than either policing or punishment, which are deeply discretionary. But these are differences in degree rather than kind. At every point in the process of making and applying domestic criminal law, the territoriality principle has declined.

This observation should be surprising. In thinking about when American criminal law is territorial, one might expect to discover a disconnect between criminalization and criminal law enforcement—that is, a system in which legislatures define crimes with territorial borders in mind and courts understand their jurisdiction in territorial terms, but police and prison officials go rogue when implementing legal rules. On this account, the deterritorialization of criminal law would be an enforcement pathology. This story would be consistent with the view that criminal law “on the books” always looks different than criminal law “in action,” particularly in a country where courts afford law enforcers so much latitude.²⁵³

Alternatively, one could imagine a criminal legal system that was territorial before conviction but unconcerned about the location of punishment. In such a system, policing and prosecution would be territorial, but imprisonment would not be. This system would line up with the prominent (though not uncontested) theory that criminal procedure rights exist to protect the innocent.²⁵⁴ On this account, territorial policing and prosecution would be a mechanism for providing fair and accurate criminal adjudication, and deterritorialized punishment would simply reflect the Constitution’s fixation on innocence. This version of criminal law would also accord with the basic thesis of the sociology of punishment, which is that “criminal law and punishment are distinct social and cultural practices.”²⁵⁵ For both the constitutional theorist and the sociologist, it would not be especially remarkable to discover that punishment looks different from pre-conviction criminal justice.

Given the stories we tell about criminal law, it would make sense if deterritorialization were the product of unruly officers or a general disregard for the convicted. If these molds fit, the conclusion would be that criminal law is territorial, notwithstanding some predictable nuance. Yet neither of these two models captures the current legal system. In reality, criminal law is more disjointed: it is sometimes territorial, sometimes predicated on alternative ideas like citizenship or abstract state interests, and most territorial at the parts of the criminal process that matter the least. It strains credulity to call this an essentially territorial legal regime. In fact, American criminal law is split between territorialism and a different theory of state power in which criminal law applies to a state’s members and interests wherever they go.

Recall, for instance, that after Skiriotes state criminal laws can follow citizens outside state territory; that the separation of venue and jurisdiction enables both prosecutors and defendants to move criminal trials; that local police forces can pool their powers across state lines; and that state prison officials can trade prisoners and rent prison beds in another state.²⁵⁶ These are formal, legal instances of deterritorialized criminal law, not aberrations or pathologies so much as evidence of a post-territorial legal regime that has been warped over time by new rules and enforcement practices. Over the course of the twentieth century, tectonic social shifts—increased mobility, the birth of a criminal enforcement bureaucracy, the rise of interstate compacts, a turn to private contracting—changed the character of criminal justice and made territorialism seem cumbersome and antiquated. These social developments have produced a legal system that is home to several competing ideas about the source and outer limits of the power to enact criminal law.

III. The Stakes of Territorialism

We are left with a conflicted criminal justice system. On one hand, the territoriality principle is entrenched in constitutional law and institutional design. On the other hand, the borders of criminal law have faded in clarity and import. Over time, both the substance and the enforcement of domestic criminal law have become less territorial: more regional, more discretionary, and more flexible about permissible forms of jurisdiction. Our state-based criminal justice system is decreasingly tethered to state soil. And criminal courts have long since abandoned nineteenth-century ideas about the limits of their authority.

These developments lead to some concrete conclusions. As the Introduction noted, the emergence of the modern criminal legal system—with its contractual, overlapping boundaries—embarrasses constitutional doctrines premised on the inherent localness of criminal law. The doctrines discussed in Part I look outdated and contrived in light of Part II. Meanwhile, the rise of task forces and ad hoc enforcement agreements demands more transparency about the reach of criminal laws. People subject to law enforcement ought to know which laws govern them, which police can detain them, what practices their taxes are funding, and which criminal prosecutions can be brought in their name.

This Part focuses, though, on the deeper implications of territoriality’s decline. It explores two observations. First, the evolution of criminal jurisdiction makes it more difficult to explain why criminal law is distinctive and thus why we have special rules for criminal law administration. The more one examines the territoriality principle, the more one wonders why Massachusetts can’t enforce Nevada’s criminal code. Second, the erosion of territorialism raises normative questions about whether borders—and, in particular, state borders—are good for criminal law. This Part considers that question and concludes with a cautious call to revive the territoriality principle as a strategy for criminal justice reform.

A. Distinguishing Criminal Law

Part II described a legal regime shaped by private agreements—contracts, compacts, MOUs—that alter the boundaries of criminal law. This is not the criminal legal system of ideal theory, in which the power to enact criminal law depends on the consent of a polity with clearly defined, stable borders.²⁵⁷ Instead, we live in a system of private ordering in which government officials negotiate the scope of criminal laws, often behind closed doors.²⁵⁸ This system developed over the course of the twentieth century as mobility and technology undermined territorial rules, encouraged more relaxed conceptions of criminal jurisdiction, and demanded flexible institutional arrangements.

In many respects, this system’s birth is a familiar story. The account of criminal law outlined in Part II lines up with the standard narrative of legal development in other disciplines, such as civil procedure and conflict of laws. In those fields, it is not at all shocking to say: “This body of law changed around the middle of the twentieth century, as social pressure to modernize and privatize made territorial boundaries seem antiquated and generated new, less formalist theories of law.” This statement broadly describes the trajectory of personal jurisdiction jurisprudence²⁵⁹ and the rejection of territorial approaches to choice of law in the 1950s.²⁶⁰

Indeed, the transformations of civil and criminal law occurred along strikingly similar timelines. The Supreme Court decided International Shoe Co. v. Washington²⁶¹—the case credited with unmooring civil jurisdiction from territoriality²⁶²—just four years after Skiriotes,²⁶³ the sponge-fishing case that recast criminal law. This timeline suggests that the Court was engineering dramatic shifts in both civil and criminal jurisdiction in the middle of the twentieth century. Yet only the civil story has received extended treatment in legal scholarship. Scholars who study procedure and conflicts—that is, academics who study the civil side of American law—widely recognize and routinely debate the decline of territorialism.²⁶⁴ Criminal law scholars, by contrast, write very little about territoriality.

But as Part I made clear, territorialism is meant to be one of the field’s core attributes. Everyone from the drafters of the Model Penal Code to constitutional theorists identifies the territoriality principle as a “maxim of . . . jurisprudence” that distinguishes domestic criminal law.²⁶⁵ And the Supreme Court treats the territoriality of criminal law as an implicit and longstanding rule.²⁶⁶ In Bigelow v. Virginia, a First Amendment case that invalidated the conviction of a Virginia newspaper editor who printed an advertisement for a New York abortion clinic, the Court deemed it “obvious” that Virginia could not proscribe abortions in New York or prosecute its residents for going there.²⁶⁷ For this obvious proposition, the Court cited Huntington v. Attrill, an 1892 case in which it had observed that penal laws are territorial because crimes offend “the public justice of the state.”²⁶⁸

In all of these accounts, territorialism is what makes criminal law identifiable as a distinct genre of public law. Territorialism is also what makes domestic criminal law real law. As Professors Jack Goldsmith and Daryl Levinson have observed, American legal scholars tend to treat domestic law as real and concrete, by contrast to the purportedly unstable and unenforceable world of international law, where legal rules inevitably devolve into power relations.²⁶⁹ In this framework, domestic law is taken to be the “paradigm” version of law, while international law is a “lesser species of law—if it qualifies as law at all.”²⁷⁰ To extend Goldsmith and Levinson’s argument, domestic criminal law is supposed to be the paradigm sort of law, with clear rules, boundaries, and sources of authority. It belongs in the category of “real” law because of its dedication to the territoriality principle, which, as I noted at the outset, is a shorthand for the idea that a government’s power to enact criminal law stems from and stops at its borders.

The territoriality principle is thus central to how scholars conceptualize criminal law. It is what divides domestic from international criminal law. It is what distinguishes criminal from civil law. It is what makes domestic criminal law into the sort of content that belongs in a first-year legal curriculum rather than some upper-level seminar on the limits of sovereignty. But as Part II demonstrated, domestic criminal law is much less territorial than it seems at first pass. Modern criminal law is riddled with exceptions to the territoriality principle: criminal laws that follow Florida citizens into the Gulf of Mexico; prosecutorial forum shopping; regional police task forces; a prison system that pays no attention to the place a person is convicted.²⁷¹ The list goes on, and as these exceptions stack up, domestic criminal law starts to seem much less committed to the principle that allegedly defines it.

My claim is not that American criminal law is completely or even mostly deterritorialized. Criminal law is often territorial. Rather, the point is that even “normal” criminal law is less tied to territory than one would expect given that territoriality is supposed to be what makes this body of law normal. In reality, states and localities cooperate with each other, domestic criminal law is split between competing theories of jurisdiction, and the criminal justice system is so thoroughly saturated with discretion that territorial borders can be ignored by both law enforcers and courts. As a result, state criminal law is just as unsettled and prone to power dynamics as “lesser” bodies of law,²⁷² and its justification and scope are just as contested. When all is said and done, domestic criminal law is not all that territorial—which means it is not all that distinctive, either.

It would take a separate article to explore the full implications of the observation that criminal law is not distinctive. The big, underlying thesis here is that the principle that divides criminal law from civil law, and domestic law from international law, has eroded over the last century such that these conceptual divisions no longer withstand scrutiny. This thesis is unnerving given how much rides on the distinction between criminal and civil law. Our basic legal infrastructure (not to mention the entire constitutional rights regime) is predicated on the idea that criminal and civil law are meaningfully different. The harshness of criminal sanctions is also supposed to be justified by the distinctiveness of criminal law. The decline of territorialism puts real pressure on these claims.

Waning territorialism also prompts questions about why criminal law operates differently than other fields. For instance, if domestic and international criminal law are not especially distinct, why can’t states (like nations) criminalize their residents’ out-of-state behavior? What is wrong with a Texas statute that prevents a Texas resident from gambling in Las Vegas? What stops Alabama from criminalizing its citizens’ conduct in California? To those immersed in Anglo-American criminal law and American constitutionalism, these sorts of statutes feel intuitively problematic—imperial, dismissive of tradition, and threatening to national unity. That intuition may be right and worth solidifying in legal doctrine. But as the law stands, it is not clear whether or why these statutes would be unlawful, aside from the “long-held maxim” that criminal law doesn’t work that way. That maxim looks precarious after Part II. It is hard to argue that state criminal law must be territorial when state legislatures, courts, police, prosecutors, and prison officials all routinely act beyond state boundary lines.

In a similar vein, one might ask why courts continue to cite and endorse the prohibition on interstate enforcement of penal laws. If criminal and civil law are not particularly distinct, why is there no choice of criminal law? Why can’t California authorize Texas to apply its criminal code? What would be wrong with a legal system where Ohio courts imposed Pennsylvania’s criminal law? And what, exactly, is the difference between that system and one where Ohio can police and imprison Pennsylvania’s criminals?

The goal is not to answer these questions so much as to bring foundational puzzles about criminal law into view by pointing out that the territoriality principle, which is supposed to be defining, has receded. The territoriality principle undergirds the intellectual framework of public law. It separates disciplines like international and domestic criminal law, and it drives unusual features of the criminal legal system, such as the practice of extradition, the presumption against extraterritorial state criminal laws, and the prohibition on shared enforcement of criminal codes. All of these practices look stranger and less defensible once one recognizes that borders are no longer sacrosanct, not even in state criminal law.

B. Taming Criminal Law

The question that follows is whether borders are desirable. The decline of the territoriality principle raises the possibility of a more flexible, less territorial criminal justice system. Part II demonstrated that government officials have long been expanding and redrawing the boundaries of criminal jurisdiction. That history suggests that one could get even more creative about the scope of criminal laws. Or perhaps the deterritorialization of criminal law is an unwelcome development. To know how to think about this issue, one has to start by asking what purpose borders serve in American criminal law.

1. The Failure of Borders

There were three justifications for territorialism lurking in the doctrines surveyed in Part I. First, the territoriality principle was supposed to protect criminal defendants. This theory appeared in the Model Penal Code, which explains that geographic limits on criminal prosecution “ensure . . . []fairness to the defendant.”²⁷³ The defendant-protective understanding of territorialism also featured in early constitutional debates over venue and vicinage.²⁷⁴ There, the claim was that territorialism would benefit defendants by keeping trials close to home, which would guarantee access to counsel and sympathetic juries.²⁷⁵ The basic belief animating this theory of the territoriality principle is that borders temper the harshness of criminal law. Territorialism limits state power, which protects those subject to coercive state force.

Second, the territoriality principle was supposed to instantiate sovereignty. By marking out the place where governments can enact and enforce their criminal laws, borders are meant to preserve a right to self-determination and nonencroachment, which in turn ensures comity between (allegedly distinct and independent) states. We saw this version of territorialism at work in the interstate extradition process²⁷⁶ and in the constitutional cases that relied on claims about criminal law’s localness to limit federal power over states.²⁷⁷ In those examples, keeping criminal law within territorial boundaries was a way to show that state governments are real and deserve respect. The basic belief animating this conception of territorialism is that the power to enforce criminal law is what makes a state a state.

Third, the territoriality principle was supposed to legitimate criminal law. In theory, the reason it is problematic for Texas courts to apply Florida criminal law, or for Massachusetts police to arrest a Vermonter in Vermont, is that criminal law is authorized by a democratic process in a certain state. Accordingly, the theory goes, it is illegitimate to apply the criminal laws of that state to people who had no say in their creation. This is the understanding of territoriality that explains the penal law taboo²⁷⁸ and the rule that in criminal law, “jurisdiction and choice of law . . . are merged.”²⁷⁹ It is also the theory behind the claim that crime is an offense against “the public justice of the [s]tate”²⁸⁰ rather than any private citizen—which is why criminal prosecutors act on behalf of “the people” and criminal cases have captions like State v. Guthrie.²⁸¹ The basic belief animating this conception of territorialism is that criminal law is an especially public sort of law, which requires an especially tight connection between the law and the people. Borders provide that link.

To summarize, the territoriality principle is supposed to protect defendants, safeguard sovereignty, and sustain a connection between criminal law and democracy. The problem is that territorialism is not doing particularly well on any of these fronts.

As Part II showed, borders rarely protect criminal defendants. In modern criminal law, police and prison officials pool their powers across boundary lines; courts decline to enforce territorial jurisdictional limits because they “benefit[] only the criminals hidden in their shadows”;²⁸² and plea bargaining has diminished the relevance of geographic restrictions on criminal adjudication. Moreover, as Part I noted, territorialism was never meant to protect defendants, at least not primarily. Despite the Model Penal Code’s insistence that the territoriality principle exists to provide fairness to defendants,²⁸³ it was actually a particular, now-outdated theory of criminal jurisdiction that motivated the Venue Clause.²⁸⁴ The Founders could have required trials to be held close to a defendant’s home; instead, they tied venue to the site of the crime. This history reveals a simple but critical lesson: territorialism might help defendants, but it does so only incidentally and only when courts are willing to enforce borders against the government. In a system where officials can contract around territorial rules and courts treat them as needless formalisms, the borders of criminal law fail to protect those accused of crimes.

Borders seem to bear a closer relationship to sovereignty. In the current criminal justice system, states get to make their own criminal laws; state criminal courts refuse to apply sister states’ “foreign” criminal codes; and the Supreme Court cites variation among state criminal laws and procedures as evidence that states enjoy self-determination. These examples suggest that borders are fairly effective surrogates for the concept of sovereignty. But the idea that states are distinct and autonomous would not disappear if criminal law were less territorial. Arguably, sovereignty is enhanced when a state can criminalize out-of-state acts and contract with other jurisdictions to enforce its criminal code. (No one thinks Germany is not a country because its criminal code travels with its citizens, and its police support INTERPOL.)²⁸⁵ Surely one of the main lessons of Part II is that states have expanded their power by ignoring and relaxing territorial rules. From this perspective, territorialism inhibits sovereignty.

The issue here is that “sovereignty” is a slippery term. State sovereignty is not a thing that exists before police arrest people in the government’s name. It is an idea that is constituted by those arrests. Jurisdiction flows from the exercise of power, not the other way around.²⁸⁶ So it is slightly strange to say that sovereignty demands strict territorial borders. Moreover, it is never quite clear what sovereignty means. Professor Don Herzog recently denounced the concept as “obsolete, confused, and pernicious,” a word that either means “actor with jurisdiction” or “state,” or functions as a “vacuous adjective suitable for trotting out on formal occasions.”²⁸⁷ Others have tried to redefine sovereignty to keep up with “modern day reality” in which governments cooperate and bargain with each other.²⁸⁸ Some academics think the idea of state sovereignty is outmoded; others believe the concept still has value.²⁸⁹ One need not resolve this ongoing debate to appreciate that sovereignty and soil are not the same thing. Borders often stand in for the idea that a government has a coherent identity and legal authority. But the mere existence of extraterritorial statutes and choice of law rules does not decimate the state.

Finally, with respect to democracy, is it not at all clear why criminal law has to be territorial to reflect the will of the people. The best version of the democracy-based argument for territorialism is the public law theory outlined above: the criminal code is the pronouncement of a polity, which is defined by legal borders. Territorialism keeps laws contained within those borders and thus forges a link between criminal law and “the people.” There is something attractive about this theory. The idea that crime is a public harm runs deep in American legal culture, so it makes sense to try to line criminal law up with the geographic unit that represents the public.

But it takes an idealized vision of the polity to get this theory off the ground. As recent debates over community policing and “democratizing” criminal law have shown,²⁹⁰ legally defined government units may not be the best representation of collective will about how criminal law should work. Certainly, state lines do not always capture the relevant boundaries of a political community. In any event, even if state borders do reflect “the public,” it remains unclear why it would be undemocratic for democratically elected state legislatures to allow their neighbors to apply their criminal laws. Nor is it apparent why a citizenship-based system of criminal law would be undemocratic—after all, one could argue that citizenship is a better marker of membership in a polity than physical presence on a piece of land.²⁹¹

Ultimately, then, the territoriality principle is not a particularly good proxy for the values it is meant to serve. Territorialism is supposed to protect defendants, sovereignty, and democracy. But borders fail to shield criminal defendants in practice. States don’t lose their autonomy when they cooperate. And it takes a series of artificial assumptions to conclude that democracy requires criminal law to be territorial.

These observations reveal some new possibilities for criminal justice reform. If territorialism is declining and the theoretical defenses of the territoriality principle are weak, one might wonder whether we should develop an even less territorial criminal legal regime. Imagine, for a moment, a version of criminal law built around a capacious conception of legislative jurisdiction and a robust conflicts jurisprudence in which courts evaluated a variety of factors—facts about the crime, the defendant, and the policy interests at stake—to determine which criminal law to apply in a given case. Consider a criminal justice system in which Rhode Island judges interpreted Indiana’s criminal laws and applied Indiana’s sentencing rules. Or a system with even looser venue requirements, in which criminal defendants could move to combine multiple cases in a single proceeding or transfer their cases to a “foreign” court in the name of convenience. These reforms, which amount to permitting a degree of forum shopping by criminal defendants, might provide a check on excessive prosecutorial discretion, which scholars widely agree is one of the American criminal justice system’s main defects.²⁹²

To take the idea even further, imagine regionalizing criminal lawmaking to better reflect the groups most affected by criminal law enforcement. If the borders of criminal law are constructed and negotiable, why not have “the criminal law of Chicagoland”? Or the criminal law of Chicago’s South Side? Lest one protest that these proposals are outlandish, recall from Part II that Indiana and Illinois have created police task forces for the express purpose of overcoming territorial jurisdictional limits, and states already join compacts to harmonize their criminal laws.²⁹³ (One might also note that, given uneven policing, we effectively do have different criminal law in different neighborhoods.)²⁹⁴ It is worth asking why redistricting criminal law is unthinkable when legislatures use legal fictions to criminalize extraterritorial acts, courts can adjudicate cases involving out-of-state conduct, police can ignore territorial boundaries, and state prison officials can outsource their prisoners.

These are different sorts of proposals. Some reconsider which borders to use in criminal law, while others reject territorialism altogether. The preceding pages have shown both that domestic criminal law contains nonterritorial theories of jurisdiction and that the American criminal justice system is less state-based than doctrine and dogma would suggest. In other words, criminal law need not be territorial, and, where it is, it need not rely on state lines. One could build from either insight to advance criminal justice reform.

2. The Promise of Borders

But perhaps it is unwise to hasten territoriality’s decline. There are some real virtues to a criminal justice system with clear geographic boundaries. Think, for instance, about the upsides of a legal regime built around state borders. Allowing state governments to make their own criminal laws—and then requiring those laws to stay within state lines—constrains criminal law imperialism. In a territorial legal system, Texas cannot prevent its residents from getting abortions in New York.²⁹⁵ Given a right to travel between states and the means to do so,²⁹⁶ this approach to criminal law protects individual liberty. It also protects New Yorkers insofar as they have different policy views on abortion than Texans and have chosen to live in a state whose laws enshrine those views. To use the language of public choice, the territoriality principle minimizes the negative externalities of a government’s decisions about how to define and police crime.

For the same reasons, territorialism promotes policy variation, particularly around whether certain conduct should be criminal at all. Although there is widespread agreement that some acts are crimes—murder is the classic example here—that consensus disappears when it comes to contested topics like low-level drug crimes and immigration offenses. There is even wider debate about the appropriate punishments for these crimes.²⁹⁷ The territoriality principle means that Kansas and Colorado can disagree about marijuana, and a Kansan can go to Colorado to try out a different criminal code.

On some level, these are just the standard reasons to like federalism.²⁹⁸ These arguments are not unique to criminal law, but they are especially compelling in a field where legal rules have moral overtones, policy disagreements are sharp, and sanctions are harsh. We should be most concerned about promoting policy variation and the ability to vote with one’s feet when the stakes of law are at their apex.

There are also more fundamental reasons to like territorialism in criminal law, whether or not it tracks state lines. As Part I noted, predicating criminal law on territory is a relatively egalitarian way to decide who is subject to criminal liability.²⁹⁹ This Article began by observing that the key jurisdictional question in a territorial legal regime is where something happened rather than who committed or suffered a harm. As a result, territorial criminal law is agnostic about identity and status. Questions like whether an alleged criminal is a U.S. citizen or a legal resident are irrelevant in a territorial criminal law regime. And potential victims are protected by virtue of their presence, no matter when they arrived or how long they plan to stay.

Of course, territorialism is only relatively egalitarian. Territory is not apolitical; borders are laced with the history of colonialism and oppression.³⁰⁰ To the extent that a criminal legal regime relies on borders, it reifies that history. But there is no uncontested way to define membership for the purpose of criminal law, and we need some way to decide when criminal law applies and whom it protects. The most common alternative to physical presence is citizenship status.³⁰¹ That alternative—a regime in which the criminal code applies to citizens and travels with them—is far more status conscious and far less egalitarian than a system built around the territoriality principle, and it raises thorny questions about the government’s authority to subject noncitizens to criminal laws.³⁰² In short, in a society where citizenship status is racialized and everyone deserves the law’s protection, geography may be the least problematic metric for criminal law.

Finally, and most hopefully, territorialism could be a tool to curb abuses in the criminal justice system. Part II did not offer much cause for optimism about borders; it portrayed a legal system in which the territoriality principle rarely constrains the police power. But one has to wonder if territorialism could be more effective. In tracing the decline of the territoriality principle, Part II offered a roadmap for using borders to reform criminal law. Consider all the ways territoriality has been undermined and all the ways it might be revived. Courts could reinvigorate the Due Process Clause to prohibit extraterritorial arrests and restrict extraterritorial punishment.³⁰³ They could read the Extradition Clause to permit states to police each other’s criminal justice systems. They could use any number of different constitutional provisions to narrow the conditions under which states can criminalize out-of-state conduct. At the very least, government officials could interpret the due process guarantee to require transparency and public deliberation about the permissible reach of domestic criminal laws.

These reforms would restore some balance to the territorialism in American criminal law. At present, the territoriality principle is asymmetric: state governments invoke territorialism to justify their power to make and enforce criminal laws, but the principle does little to temper that power. Criminal law would be more fair—more coherent and restrained—if the territoriality principle worked both ways. Imagine if the territoriality principle not only authorized Tennessee to make criminal laws but also prohibited Tennessee police from arresting people in Georgia. Consider a version of territorialism that permitted Hawaii to punish people but then prevented Hawaii from outsourcing its prisoners to Kentucky. In this account of criminal law, territorialism licenses and limits state power. Borders are both the fount of government authority and a restraint on officials’ discretion. Academics frequently argue that criminal law enforcers have too much discretion.³⁰⁴ Perhaps one problem is that the territoriality principle is underenforced.

The intuition driving these suggestions is that formalism is beneficial in criminal law. These days, it can be strange to call for more formalism in any area of law. The realist turn in the American legal academy over the past half-century has made formalist arguments seem old-fashioned and slightly tone-deaf.³⁰⁵ In other branches of public law, including constitutional and administrative law, formalism is often critiqued as insincere and undue.³⁰⁶ But informality is uniquely destructive in the criminal legal system.³⁰⁷ American criminal justice is dominated by ad hoc arrangements, loose administration, and unfettered discretion. In this context, formalism—even stilted and “hopelessly arbitrary”³⁰⁸ constraints like state lines—is ameliorative.

Indeed, territorial boundaries might be just the sort of structural limits that could discipline American criminal justice. In other fields, scholars and reformers have turned to structural concepts to advance egalitarian aims. In immigration law, for instance, advocates have made preemption and commandeering claims to protect immigrants’ rights when equal protection law has proven inadequate to the task.³⁰⁹ Critics of the American criminal justice system might make a similar move, looking to the territoriality principle to constrain criminal law imperialism and law enforcement excess where arguments based on equality, liberty, and criminal procedure rights have failed.

My own inclination is toward this latter path—toward embracing territorialism as a means to regulate domestic criminal law. But I may be on the wrong side of history. The modern criminal justice system may simply be too functional, too contractual, and too interdependent to revive old ideas about criminal jurisdiction. That is certainly one conclusion to draw from Part II. For those who seek to tame the harshness of American criminal justice, there may be more mileage in abandoning or redrawing the boundaries of criminal law than in enforcing them. The real point is that all of this is possible. The territoriality principle is a norm, not a rule. And the uncritical assumption that criminal law has to be territorial has obscured some interesting and open questions about the future of the field.

Conclusion

Territorialism is an organizing principle in domestic criminal law. The presumption that a government’s power to criminalize and punish conduct stems from its authority over some piece of land dates to the founding. But over time, that presumption has weakened and warped in ways that raise foundational questions about the scope and legitimacy of American criminal law. The basic terrain of criminal law has shifted. In response, we have to ask why borders matter and whether we can sustain a criminal legal system without a clear sense of the source and limits of state power.

* Associate Professor, New York University School of Law. For helpful comments and suggestions, thanks to Greg Ablavsky, Rachel Barkow, Will Baude, Maggie Blackhawk, Mary Bosworth, Rick Brooks, Adam Cox, Kevin Davis, Justin Driver, Antony Duff, Barry Friedman, David Garland, Andrew Hammond, Don Herzog, Carolyn Hoyle, Marcel Kahan, Lucy Kaufman, Daryl Levinson, Ian Loader, Matt Lockwood, Troy McKenzie, Liam Murphy, John Rappaport, Julian Roberts, Sam Scheffler, Steve Schulhofer, Tom Schmidt, Linda Silberman, Lucia Zedner, and participants in the All Souls Seminar Series at the University of Oxford, the N.Y.U. Colloquium in Legal, Political, and Social Philosophy, the N.Y.U. Furman Scholars Seminar, and law faculty workshops at the University of Chicago, the University of Connecticut, Seton Hall, and Wayne State. For excellent research assistance, thanks to Gaelin Bernstein, Jessica Coffrin-St. Julien, Rachel Ellman, Rebecca Hutchinson, Elaina Marx, Evan Ringel, Matt Rosenthal, and Eli Watkins. For incisive and generative edits, thanks to the editors of the Michigan Law Review.