Copyright © Villanova
University; Henry H. Perritt, Jr
1996
JURISDICTION IN CYBERSPACE [FNa]
41 VILL1. L. REV. 1
Henry H. Perritt, Jr. [FNaa]
.
. Introduction
THE Global Information Infrastructure, also known as the "GII" or "Information Superhighway," links suppliers and users of electronic resources around the world, making it more difficult to localize wrongdoing for purposes of criminal or civil litigation. The same lack of localization makes it difficult to enforce judgments of criminal and civil courts, and to organize civil discovery and tools of criminal investigation, such as search warrants and subpoenas. Conduct with potentially serious legal consequences is difficult for traditional sovereigns to control in the GII because it is ephemeral, invisible and crosses geographic boundaries easily. Geographically based concepts of sovereignty must be squared with the nature of open networks, which are indifferent to geographic boundaries. Conventional doctrines of jurisdiction to prescribe, to adjudicate and to enforce legal decisions [FN1] must evolve to handle new disputes *3 in cyberspace.
Whose substantive legal rules apply to a defamatory message that is written by someone in Mexico, read by someone in Israel by means of an Internet server located in the United States, injuring the reputation of a Norwegian? Whose courts have jurisdiction to adjudicate claims of injury or violation of national rules? Must a Norwegian go to Mexico or to the United States to find a legal institution with power over one of the two potential sources of compensation? If not, and if jurisdiction exists in Norway, the most convenient forum for the victim, how is a favorable decision by a Norwegian tribunal ordering that the Mexican originator or the American intermediary pay damages to be enforced outside of Norway?
Similar questions, addressed by different rules, arise in the criminal context. Suppose the message is criminal instead of defamatory, involving child pornography or indecency, or representing some sort of financial fraud, forgery or terrorism. Must the wrongdoer be tried only where he or she is physically found? If the answer is yes, how should the wrongdoer be apprehended and moved to the place of trial? Whose substantive criminal law should apply?
Arguably, in the hypothetical above, Norway has the greatest interest in punishing the conduct, while Mexico has only a weak interest, but legislatures usually focus on conduct occurring within the jurisdiction in defining crimes, and criminal courts rarely exercise power to prosecute for a crime against another jurisdiction's laws. While conflict of law jurisprudence has wrestled with problems like these for centuries, [FN2] the transnational character of cyberspace will increase the frequency with which choice of law questions arise. Also, questions may arise as to the suitability of traditional choice of law formulations and their possible extension into the criminal context.
*4 This Article explores how the legal system deals with information technology procedurally, beginning with well-recognized problems of civil and criminal jurisdiction, [FN3] and discussing closely associated problems in choice of law, [FN4] enforcement of judgments [FN5] and discovery. [FN6] Next, it explores new legal institutions designed to deal with the trans-border difficulties caused by the GII. [FN7] Of the several jurisdictional problems, the problem of criminal jurisdiction is the most serious. In the civil sphere, personal jurisdiction and choice of law pose no insuperable difficulties, because there is growing agreement on the basic principles of both, which adequately accommodate most of the special circumstances of the GII.
Enforcement of civil judgments presents somewhat greater difficulty, because the international legal framework is not sufficiently comprehensive to allow parties confidence in enforcing a civil judgment in another country where assets may be located. Nevertheless, the availability of arbitration as an alternative dispute resolution forum, the decisions of which are nearly universally enforceable, provides a readily acceptable alternative to civil judgment enforcement. [FN8]
In contrast, the power of criminal courts is rigidly localized; they lack the flexibility afforded civil courts by the doctrine of personal jurisdiction supported by minimum contacts and choice of law. Criminal jurisdiction based on extradition treaties may not include computer crimes. Moreover, the power to conduct criminal searches and seizures across national boundaries is unevenly defined through the treaty process and foreign intelligence cooperative agreements, although the availability of civil discovery on the American model is similarly circumscribed.
The underlying transactions in the GII are not entirely new. International trade, international communications and international payments have been occurring for several hundred years. If changes in the legal system are needed to permit the GII to realize *5 its potential, it must be because the new technologies for trade, communications and payments diminish the suitability of existing legal arrangements. The new technologies may make it more difficult to identify the time and place of legally significant events. Additionally, the new technologies may reduce the cost of transactions to the point where the cost of utilizing existing legal measures, including the cost of uncertainty with respect to outcomes, may become so important relative to the value of the transaction that they discourage commerce. Finally, the new technologies may open up international commerce to types of transactions that heretofore mostly occurred only within a single country. For these reasons, the GII will force the international legal community at least to modify the way in which it deals with transnational conflict.
II. Hypotheticals
A hypothetical illuminates the relationship between the various aspects of civil and criminal jurisdiction. Suppose someone in country X operates a World Wide Web server on the Internet and intentionally makes false statements in Web pages accessible through this server for the purpose of inducing customers in Virginia to buy inferior merchandise through the server. The purposeful direction of the misleading statements to Virginia would, under World-Wide Volkswagen [FN9] and Asahi, [FN10] support personal jurisdiction in a Virginia state court. [FN11] Of course personal jurisdiction would not exist unless the actor were served with appropriate process, which probably would not include electronic notice alone under current rules. The doctrine of forum non conveniens, however, might lead to the dismissal of a Virginia action because witnesses and other evidence necessary to prove the mental state elements of the fraudulent misrepresentation are more likely to be found in country X. [FN12] Choice of law arguments would focus on the competing interests of country X and Virginia. Virginia has legitimate interests in protecting its citizens from fraudulent misrepresentation, but country X has an interest in discouraging fraudulent conduct by persons operating from within its territory while also allowing a sufficient ambit of legitimate or mistaken commercial *6 activity. Virginia substantive law probably would be applied, although there is room for argument.
As to criminal prosecution, no court in Virginia would have personal jurisdiction over the actor unless the actor could be made physically available before the Virginia court. This could be accomplished through arrest while the actor is in Virginia, through extradition from country X, or from extra-legal physical seizure and transportation to Virginia followed by arrest there. Appropriateness of criminal venue in Virginia would depend on the commission of a crime in Virginia; the commission of a crime in country X would be insufficient, unless the fraudulent misrepresentation through the Internet were viewed by Virginia law as an international "universal" crime like piracy. Most likely, a crime has been committed in Virginia only if the Virginia or federal criminal statutes against fraudulent misrepresentation of this sort (perhaps wire fraud or consumer fraud) have extraterritorial effect, in which case the Virginia targeting of the conduct in country X probably would satisfy the effects test under international law and justify extraterritorial application of Virginia or federal criminal law to the conduct. Any civil discovery or search and seizure in country X would depend on the cooperation of country X officials in responding to U.S. requests.
Further, if there were less evidence that the purpose of the statement was to induce persons in Virginia, as opposed to persons in general, then there would be greater difficulty with personal jurisdiction in the civil case, and there also would be greater difficulty with substantive application of Virginia or federal law in the criminal case--the criminal equivalent of choice of law. Of course, the courts of country X would have personal jurisdiction over the actor with respect to both civil and criminal litigation, but could entertain a prosecution only for violation of criminal laws of country X (unless an international crime is involved as noted earlier in this section), or unless the criminal jurisprudence of country X allows prosecution of a citizen of country X for violating foreign law (a relatively unlikely prospect, though one that enjoys some case law support in France and Germany). In the civil case, the courts of country X would apply country X choice of law rules, which probably do not differ materially from those that would be applied by a court sitting in Virginia. If the actor were an American citizen, that would give the courts in Virginia a stronger basis to exercise personal jurisdiction in the criminal prosecution, but that citizen or she still could not be tried in absentia.
*7 Finally, the analysis would be essentially the same, if, for instance, the conduct involved intentional public distribution of pirated material, infringing the copyright of a citizen of Virginia, except that the civil action would be for copyright infringement instead of fraudulent misrepresentation, and the criminal prosecution would be for criminal copyright infringement. There is no reason that the outcome on any of the dimensions of civil or criminal jurisdiction would be different, although the availability of the many reported cases could make resolution of the civil personal jurisdiction and choice of law questions more certain. Also, congressional intent with respect to extraterritorial jurisdiction might be different as between the copyright crime and the wire fraud or consumer fraud crimes. These hypotheticals illustrate the problems created by the GII. The contours of these problems, as well as some possible solutions, are explored below.
III. Relationship Among Personal Jurisdiction, Service of Process, Venue,
Choice of Law, Enforcement of Judgments and Discovery
Criminal and civil jurisdiction proceed from common concepts, although the detailed requirements are significantly different in modern law. The two sections in this part explain the conceptual relationship among different aspects of the power of courts over controversies, comparing and contrasting civil and criminal doctrines, historically and in modern law.
A. Introduction to Determinants of Judicial Power
Four procedural doctrines interact in circumscribing the power of any court to decide a controversy according to particular substantive rules: personal jurisdiction; notice by arrest, service of process or otherwise; choice of law; and venue. A fifth doctrine, enforcement of judgments, encompasses all of the other four. The res judicata effect of a judgment, and therefore its enforceability, [FN13] depends on the court rendering the judgment having personal jurisdiction, being the appropriate venue and giving appropriate notice to those bound by the judgment. Inappropriate choice of law *8 also is a basis for challenging the efficacy of a judgment, albeit a much weaker one.
The four doctrines were not particularly distinct in early English procedure, [FN14] but later diverged in nineteenth century and early twentieth century American jurisprudence. Now, they are beginning to converge again in civil procedure (though not in criminal procedure). [FN15] The fair play and substantial justice test for personal jurisdiction [FN16] requires interest analysis similar to that used to resolve choice of law disputes. It also overlaps considerably with forum non conveniens analysis used to determine venue. More *9 flexible rules for personal jurisdiction necessitate an expansion of the types of notice that are acceptable. [FN17] Venue rules also interact with choice of law rules. Choice of law analysis usually begins with the forum's choice of law rule; thus, venue may determine the outcome of choice of law. [FN18]
As commerce, and therefore litigation, becomes more international in character, American personal jurisdiction, venue, choice of law and notice rules must be reassessed and perhaps harmonized with corresponding rules in other countries. [FN19] Analysis and harmonization can be simplified if the convergence eventually results in the atrophy of independent personal jurisdiction inquiry, replaced by stronger forum non conveniens rules and clearer limitations on choice of law. [FN20] International reassessment and harmonization of criminal jurisdiction must take place in the context of ongoing efforts to define codes of international crimes and to develop concepts for international criminal courts.
B. Historical Relationship Among Jurisdiction, Venue and Choice of Law
The prospects for unifying personal jurisdiction, venue and choice of law rules are influenced by the historical separation of the rules, and their historical links to private international law concepts. Historically, venue rules not only determined where a suit *10 could be brought, subsuming personal jurisdiction inquiries, [FN21] but also determined what law would be applied. Until Pennoyer v. Neff, [FN22] territorial limitations were imposed mostly by state (or national) venue statutes. Venue, in turn, depended on whether the cause of action was "local" or "transitory." Local causes of action had only one acceptable venue, while transitory causes of action might have several. All crimes except piracy were local. Choice of law meanwhile was hardly an issue; it was assumed that any court would apply its own substantive law. Choice of venue thus determined choice of law. [FN23] There was, under these early formulations, little difference between criminal and civil jurisdiction and venue. For both civil and criminal cases, courts exercised jurisdiction only over those physically before the court. [FN24] Once a court had jurisdiction over the defendant, it applied its own substantive law.
One of the best explanations of the original relationship among venue, personal jurisdiction and enforcement is found in Livingston v. Jefferson, [FN25] a trespass action against Thomas Jefferson. The alleged trespass occurred in New Orleans, and Jefferson was found in Virginia. The Virginia circuit court dismissed the action, finding that it was local and therefore could be brought only where the trespass occurred. [FN26] Jefferson was not found in that district and *11 therefore could not be sued there. Under the prevailing analytical framework, he apparently could not be sued anywhere. Nevertheless, both John Marshall, sitting as Circuit Justice, and Judge Tyler agreed that the rule was long established that trespass actions were local rather than transitory, thereby limiting venue to Louisiana. [FN27]
As important as the holding is, the explanations offered by Chief Justice Marshall and Judge Tyler are even more illuminating. In trespass actions, Marshall explained, the title and bounds of land might come into question, and only a jury from the vicinage of the land could appropriately determine such facts. [FN28] Judge Tyler reasoned that a judgment for the plaintiff would potentially necessitate execution by the sheriff and his posse to remove the trespasser, if necessary: "And suppose the sheriff and jury should deny the power of the court, could they be coerced?" [FN29] In other words, the power of the court should be circumscribed by the court's power over the officer that must execute the court's judgment. [FN30] If execution could be had only in one place, then only the court of that place had the power to try the case. [FN31] Judicial power was limited both by convenience analysis (proximity to the subject matter of suit), the focus of modern venue analysis, and also by considerations of judgment enforceability.
Transitory actions were different. For them, venue existed more or less wherever the defendant could be found. Contract was the archetypal transitory action, based on a legal and moral obligation in the person to perform his promises, and such actions thus were not tied to any particular place. [FN32] There was a loose correspondence between real actions and local actions, and between personal actions and transitory actions. When enforcement required *12 doing something with respect to a res, only the local sheriff could do it. Conversely, when enforcement involved a person who might move around, enforcement could occur wherever the person was found, through imprisonment for debt, or simply through a "capias ad satisfaciendum" [FN33] or "capias ad respondendum." [FN34]
These limitations, however, were purely common-law limitations. [FN35] The distinction between real and personal actions achieved constitutional status in Pennoyer [FN36] some sixty-six years later, in which an action proceeded as though it were personal when the remedy sought was in rem. [FN37] In Pennoyer, the United States Supreme Court reasoned that the territorial limits on the power of state courts are like the territorial limits on the power of courts in nation-states. If a sovereign purported to decide a personal action when it lacked the power to serve process on the person of the defendant, its judgments were not entitled to recognition or enforcement in other states. [FN38] Conversely, the same court could exercise power over a real action because it had de facto power over the res located within its territorial limits. [FN39] Of course, an important feature of this distinction was that an in rem judgment could be enforced only to the limits of the value of the thing. [FN40] The Pennoyer Court held that purporting to decide cases against persons not within the personal jurisdiction of the court violated due process under the Fourteenth Amendment. [FN41]
*13 The distinctions between personal and real actions and between local and transitory actions have largely disappeared from personal jurisdiction analysis for civil actions. Further, the inherently non-local nature of information technology might seem to make the distinction entirely irrelevant in analyzing civil procedure issues for the GII, but that is not so. The traditional distinction between local and transitory actions appropriately sought to conform doctrine to practical limits on judicial power. The rationale used by Judge Tyler in Livingston is a good example. Even though interests analysis is the centerpiece of the law of personal jurisdiction [FN42] and choice of law analysis, [FN43] the pragmatic power dimension, along with an assessment of efficiency and convenience, becomes more important in deciding, as a practical matter, where a lawsuit over electronic conduct should be litigated and where judgments resulting from such a lawsuit can be enforced practicably. The simple answer, as in Livingston and Pennoyer, is that judicial power may be exercised over the GII wherever a human defendant may be found and wherever some assets are located. The GII may span geographic boundaries, but its human actors are present in some traditional jurisdiction, and the hardware, software and financial assets used to operate each part of the GII are located in some traditional jurisdiction. Lawsuits will be efficacious in those jurisdictions. The practical interaction between judicial power and electronic actors and assets appropriately guides application of modern personal jurisdiction, choice of law, venue and enforcement doctrines considered in the next several sections.
IV. Personal Jurisdiction
Personal jurisdiction in both the criminal and civil context is the basic precondition of jurisdiction to adjudicate. Personal jurisdiction refers to the power of a court over the defendant. As the historical sections of this Article explain, personal jurisdiction originally derived from physical control or custody over the defendant. It has been relaxed in the civil context to allow power to be asserted over persons with certain connections with the forum state who are given formal notice through service of process. In the criminal context, however, personal jurisdiction still signifies physical custody. *14 Exploring application of personal jurisdiction concepts in cyberspace requires review of the "minimum contacts" and "fair play and substantial justice" concepts of civil personal jurisdiction, initially articulated in International Shoe; [FN44] an understanding of the forum non conveniens doctrine that sometimes pretermits exercise of personal jurisdiction when it exists; and the various techniques, including extradition and extralegal obtaining of custody that extend into the international sphere, involved in criminal personal jurisdiction.
A. Personal Jurisdiction in Civil Cases
Personal jurisdiction of an American court, state or federal, depends on the interaction of an affirmative statutory or common law source of jurisdiction and limitations imposed by constitutional due process. Traditionally, as the preceding section explains, state courts asserted jurisdiction over persons physically present within their territorial boundaries and served with process while they were there, and asserted jurisdiction over things found within their territorial boundaries and attached while they were there. [FN45]
These affirmative bases of jurisdiction are constitutionally permissible merely because they are traditional under the Supreme Court's decision in Burnham v. Superior Court. [FN46] But states, in addition *15 to asserting jurisdiction over persons or things found within the forum state, also frequently assert jurisdiction in other cases under long arm statutes. [FN47] Typical of long arm statutes are the provisions of the Uniform Interstate and International Procedure Act, [FN48] which authorizes jurisdiction over persons served (usually by mail [FN49]) out-of-state when they have certain connections with the forum state. Long arm statutes typically authorize the exercise of jurisdiction to the extent permitted constitutionally [FN50] under a line of Supreme Court cases beginning with International Shoe. [FN51]
Under this line of cases, jurisdiction over a person or a corporation not served with process within the jurisdiction is permissible, but only when the defendant has minimum contacts with the forum state [FN52] and only when the assertion of jurisdiction satisfies considerations of fair play and substantial justice. [FN53] Within this analytical *16 framework, contacts such as sending goods into the forum state or permitting instrumentalities that cause injury to enter the forum state count as minimum contacts only if they are purposeful; mere foreseeability of contact with the forum state is not enough. [FN54] Also, entering into a contract with someone in the forum state may satisfy minimum contacts requirements when the contract provides for a repetitive series of transactions with the forum state. [FN55]
Personal jurisdiction analysis in Europe places less emphasis on service of process while a defendant is physically present and greater emphasis on factors that fit comfortably within the minimum contacts concept in American jurisprudence. [FN56] Thus, a defendant always may be sued where he or she lives, and nonresident defendants also may be sued when they have intentionally engaged in transactions with someone in the forum state. [FN57] Beyond that, some European jurisdictions, such as France and Scotland, assert what is known as "exorbitant bases of jurisdiction." [FN58] One exorbitant basis permits courts in the plaintiff's place of residence to hear suit against nonresident defendants, at least when the claim has some relationship to the jurisdiction. Another allows what Americans would call quasi in rem jurisdiction. A third allows jurisdiction over anyone served with process while temporarily present within the forum--so-called "tag" jurisdiction. [FN59]
The Constitution permits in rem jurisdiction when a civil defendant has property located in the forum state and the dispute relates to the property. Quasi in rem jurisdiction--when a defendant has property in the forum state but the dispute is unrelated to the property--probably is permissible only when the plaintiff can show exigent circumstances, such as a practical inability to effect *17 personal service on the defendant. [FN60] While it may be tempting to suppose that quasi in rem or in rem jurisdiction could be used expansively to obtain meaningful relief by a wrongdoer by computer, Internet technologies decrease, rather than increase, the likelihood of valuable assets being located in the jurisdiction of the plaintiff. Even if an argument could be sustained that an Internet server used to disseminate the harmful information is vicariously "present" in the forum state, turning the property into money requires the practical exercise of dominion and control over something that can be sold, and vicarious presence is not sufficient. If the server is a SPARC station and is located in Luxembourg, the plaintiff and judgment creditor will get money only if the SPARC station can be sold and that requires obtaining control over it in Luxembourg. Vicarious presence of that computer in Virginia because it is easily accessible from Virginia is not enough to permit it to be sold in Virginia.
Longarm jurisdiction in civil cases knows no real counterpart in criminal personal jurisdiction because of the rule that persons may not be tried in absentia. Thus, while it is tempting to compare the service of process in another jurisdiction with extralegal acquisition of custody, the resources required to do the latter are substantially greater than those required for the former, and the affront to the sovereignty of the place of service or acquisition of custody is much greater for the latter. On the other hand, the effect of the apparent difference is narrowed by similarities in the problems of judgment enforcement. [FN61]
B. Obtaining Personal Jurisdiction over Publications [FN62]
A number of cases apply the minimum contacts and fair play and substantial justice factors to print publications. These cases are helpful in applying personal jurisdiction doctrine to the GII because print publications, like electronic ones, spread out and come in contact with a multiplicity of jurisdictions, based on conduct that is concentrated at the place of the author or publisher. A publication usually satisfies the minimum contacts analysis if it has a substantial *18 circulation in the jurisdiction [FN63] or if the defendant publisher intended to cause injury in the jurisdiction. [FN64]
Similar minimum contacts analysis applies when the defendant makes an allegedly tortious statement which is later published by another, although the maker of the statement may have less responsibility for where circulation occurs. In Madara v. Hall, [FN65] the plaintiff, a California resident, brought suit in a Florida district court against the defendant for an allegedly libelous statement made by the defendant in California during a telephone interview with a magazine reporter. [FN66] The statements were later published in a magazine which was circulated in Florida. [FN67] The United States Court of Appeals for the Eleventh Circuit applied the traditional minimum contacts analysis in considering whether the defendant had sufficient contacts with Florida to warrant personal jurisdiction over him. [FN68]
The court determined that the defendant was "not subject to personaljurisdiction in Florida." [FN69] First, the forum was not convenient to either of the parties. [FN70] Second, Florida had little interest in adjudicating the dispute because neither party resided in the state. [FN71] Finally, "neither the interstate judicial system's interest in obtaining the most efficient resolution of controversies nor the interest of the states in furthering fundamental social policies would be served by subjecting Hall to the jurisdiction of Florida courts." [FN72]
In Ticketmaster-New York, Inc. v. Alioto, [FN73] the plaintiff brought a defamation action in the United States District Court for the District*19 of Massachusetts against the defendant, a California resident, for statements he made during a telephone interview which were later published in a Massachusetts newspaper. [FN74] The court held that two factors determine purposeful availment: foreseeability and voluntariness. [FN75] It stated that if "the source of an allegedly defamatory remark did not initiate the pivotal contact, and the in-forum injury is not reasonably foreseeable, jurisdiction may not be asserted over the source based on the comment." [FN76] The defendant did not make the initial contact, but the in-forum injury was foreseeable. [FN77] But then the Court of Appeals "evaluate[d] the fairness of asserting jurisdiction in the totality of the circumstances" [FN78] and found that it would be fundamentally unfair to assert personal jurisdiction over the defendant, [FN79] in part because of the burden on the defendant of defending in Massachusetts. Requiring "a California resident to appear in a Massachusetts court is onerous in terms of distance, and there are no mitigating factors to cushion that burdensomeness here." [FN80] It would not be fair "on the strength of a single remark uttered in the course of a single unsolicited telephone call from a Massachusetts-based journalist, to compel a California resident to defend a tort suit in a court 3000 miles away." [FN81]
The touchstones of personal jurisdiction analysis drawn from print publication cases are appropriate for GII cases. An electronic publisher should be subject to personal jurisdiction in any place to which the electronic publisher intentionally sends its publication. Thus, subscription-based commercial systems like CompuServe or America Online should be subject to personal jurisdiction in places *20 where significant numbers of their subscribers reside. The residence of subscribers is known to these services; they derive revenue from those subscriptions, and there is little reason to distinguish between the electronic subscriber and the print subscriber. If an electronic publisher (including an individual poster) publishes a statement intended to injure someone, the publisher should be subject to personal jurisdiction in the place where the injured party is located, under the precedent represented by Calder v. Jones, [FN82] although the Gestalt factors might be brought into play and lead to a different result as in Madara.
There are many conceivable GII cases in which publication does not support the exercise of personal jurisdiction so strongly. For example, one might post a message to a list. The list causes dissemination to all those subscribed to the list. But the poster usually has no knowledge of the extent of the list and thus the dissemination of this posting to a particular person is usually neither purposeful nor foreseeable, unless other facts indicate specific knowledge of a particular recipient of messages posted to that list. Absent such special facts, the exercise of personal jurisdiction is not appropriate merely based on the dissemination of messages through the list. An even weaker case for the assertion of personal jurisdiction arises from placement of material on servers connected to the Internet or Internet-like open architectures. Then, the act resulting in the receipt of the message in a particular place is the act, not of the publisher, but of the retriever. Publication in these circumstances should not subject the publisher to personal jurisdiction in places where the information is retrieved, under the rationale of Hansen v. Denckla. [FN83]
The Internet, and especially its World Wide Web application, creates a publishing infrastructure in which responsibility for "sending" information depends on the particular technological configuration. Many of them, unlike print publishing, rely on the consumer to come to the publisher and to trigger the transmission to that particular consumer. An exchange the author had online through one of the Counsel Connect (CC) electronic seminars with another participant fairly frames the characterization issue. The other CC participant said that the system operator in cases like the one shown in Figure 1 below is like someone who leaves a number of objects *21 on the floor. The requester is like someone who uses a vacuum cleaner to pick the objects up. The vacuum cleaner bag is like the jurisdiction in which the effect of the information is felt. No sensible person, he said, would assert that the person leaving the objects on the floor had placed them in the bag of the vacuum cleaner.
Figure 1
TABULAR OR GRAPHIC MATERIAL SET FORTH AT THIS POINT IS NOT DISPLAYABLE
TABULAR OR GRAPHIC MATERIAL SET FORTH AT THIS POINT IS NOT DISPLAYABLE
The author preferred a different metaphor, comparing the electronic service to a vending machine in which the operator had placed different types of items in particular compartments for customer selection. The electronic customer is then like the vending machine customer, who selects a particular type of item, inserts money if necessary, and pulls the handle or pushes a button causing that type of item to be delivered. The vending machine comparison lends support to the fairness of treating the operator as a seller of that particular item, subject to regulation by whomever has jurisdiction over the vending machine compartments. That is so even though the vending machine operator was in some sense passive, and the actual delivery of the item occurred only when the customer acted to cause it to be delivered. Of course both metaphors involve conscious placement of items in a place from which they will be retrieved. Both metaphors recognize that determining who causes the content in question to enter a particular jurisdiction seeking to regulate the content is important in determining *22 whether that jurisdiction may regulate it. [FN84]
The typical bulletin board/electronic publisher concept is illustrated in Figure 1. The server not only provides pointers and other finding and retrieval value, represented by the rectangles below the circle; it also publishes content, represented by the circle. The client has dealings only with the server, which looks like a conventional publisher, except that the publisher does not send anything (or even make a copy) until the client requests it.
Figure 2
TABULAR OR GRAPHIC MATERIAL SET FORTH AT THIS POINT IS NOT DISPLAYABLE
TABULAR OR GRAPHIC MATERIAL SET FORTH AT THIS POINT IS NOT DISPLAYABLE
In Figure 2, the operator simply sets aside an electronic space within which others can exchange material. The content originator posts a file to the server, from whence it is retrieved by the client. In that situation, the operator may have no knowledge of what items are being offered and retrieved, and thus it is more difficult than in Figure 1 to conclude that the operator is responsible for the contact between the item and Jurisdiction B. Not only was the operator not the initiator of the transmission to Jurisdiction B, but it may have lacked knowledge that the particular item was available for transmission to Jurisdiction B. Still, the server has control over its facilities and could have blocked the content originator from posting material there.
Figure 3
TABULAR OR GRAPHIC MATERIAL SET FORTH AT THIS POINT IS NOT DISPLAYABLE
TABULAR OR GRAPHIC MATERIAL SET FORTH AT THIS POINT IS NOT DISPLAYABLE
*23 The typical Web server is in yet a different situation, as shown in Figure 3. Web server X never has possession of the content. It points to server Y, which in turn points to the content on the server of the content originator. Transactions 1-4 with servers X and Y are requests for and transmissions of pointers only. The eventual request for and retrieval of the content transpires entirely between the client and the content originator in transactions 5 and 6. The Web server that simply points to other (content) servers is rather like one who gives directions to the dirty movie theater or to the drug dealer. While the supplier of the pointers can be said to be involved in the distribution chain, he or she never comes in contact with the accused material.
In the Web often the information is transmitted indirectly as a result of interactions between user/requesters and intermediaries who point to particular information objects--sometimes without the knowledge of the person making the information object available. Such an infrastructure is not only two-layered; in many cases a pointer points, not directly to the full information resource, but to another collection of pointers, which may point to still other collections of pointers and so on, as in Figure 3, collectively marking a trail to the complete resource. The computer programs involved assemble a trail from the three pointers and then retrieve the desired *24 content from wherever it resides, directly into the computer of the requester, without the content having to traverse all the intermediary computers. Whether an intermediary points directly or indirectly to the desired resource is inherently an engineering decision driven by performance considerations. In many cases, the decision to maintain copies of a particular information resource is made entirely automatically without any human intervention. A clear example of this is in the caching of recently-retrieved resources within a World Wide Web browser such as Netscape.
Unless one is to accept a broad stream of commerce theory of personal jurisdiction, which was rejected in World-Wide Volkswagen [FN85] and Asahi, [FN86] none of the publishers, save the one in Figure 1 above, are within the personal jurisdiction of courts in Jurisdiction B. Moreover, it is dubious whether the content originator in Figure 2 is within the jurisdiction of the courts of Jurisdiction B. While the originator--or his or her computerized agent--purposefully sent the information item into Jurisdiction B, the sending occurred instantaneously in response to the address supplied by the client, without any advance arrangement by the content originator. The only place that clearly has jurisdiction over the content originator is Jurisdiction D. Jurisdictions A and C have jurisdiction over the intermediaries X and Y, but the victim may not have a cause of action against them. [FN87]
In Pres-Kap, Inc. v. System One, Direct Access, Inc., [FN88] a split panel of the intermediate Florida court held that Florida courts lacked personal jurisdiction over a New York travel agent who contracted to use an airline reservations system with its main database located in Miami, Florida. [FN89] The majority opinion expresses rhetorical concern over the implications of holding that the location of computer databases could determine personal jurisdiction over suits against *25 those who use them, but offers little analysis as to why personal jurisdiction should not have existed under Burger King. [FN90] The dissent is more persuasive. [FN91]
Plus System, Inc. v. New England Network, Inc. [FN92] provides more guidance than Pres-Kap. In Plus System, a Colorado-based ATM network sued its New England affiliate for refusing to implement a new royalty charge. [FN93] In rejecting the challenge to personal jurisdiction, the district court found, among other things, that the defendant's regular use of the plaintiff's computer system located in Colorado was an availment of Colorado and its law. [FN94] The defendant benefited from the services provided by plaintiff's computer system, and it made no difference that the defendant's connection occurred indirectly through a service organization in Wisconsin. [FN95] There were, to be sure, other contacts, including a visit and tour of the Colorado computer facility by defendant's personnel, a contractual choice of law clause that pointed to Colorado and signature by the plaintiff, at least, of the contract in Colorado. [FN96]
C. Venue
Even when personal jurisdiction exists, venue may not be proper. [FN97] Civil actions in federal court based on diversity jurisdiction may be brought in a judicial district where any defendant resides, where a substantial part of the events or omissions giving rise to the claim occurred, where a substantial part of the property that is the subject of the claim is situated or in "a judicial district in which the defendants are subject to personal jurisdiction at the time the action is commenced, if there is no district in which the action otherwise may be brought." [FN98] The rules for actions based on *26 federal question jurisdiction are the same, except that the catch-all allows litigation in "a judicial district in which any defendant may be found," as opposed to one in which the defendant is "subject to personal jurisdiction." [FN99] These venue rules are typical; most state rules are similar. These basic venue rules would permit litigation in any forum in which personal jurisdiction exists, because the transactional link necessary to support personal jurisdiction, a fortiorari, would support venue.
Reuber v. United States, [FN100] a case involving electronic conduct, provides some interesting guidance on venue under a specialized statute. Reuber filed a claim under the Federal Tort Claims Act (FTCA), which authorizes venue only in the judicial district where the plaintiff resided or "wherein the act or omission complained of occurred." [FN101] The language limiting venue is a more restrictive version of the language in the general federal venue statute. [FN102] The plaintiff filed one suit in the United States District Court for the District of Columbia, and another, againsta co- defendant, in Maryland state court, which was removed to the United States District Court for the District of Maryland. The District of Columbia action was transferred to Maryland, [FN103] and the United States claimed improper venue in the action against it. The court of appeals distinguished a case involving an FTCA suit over an air traffic controller transmission from Utah to a pilot flying over Montana causing that pilot to crash. [FN104] The court found that the radio transmission was directed specifically to the pilot in Montana and thus the situs of the act in that case could reasonably be perceived as including the place where the tortious radio transmission was targeted and where the foreseeable harm would occur. [FN105] But in Reuber's case, the allegedly tortious conduct involved disclosure and dissemination of a letter critical of him, and the court of appeals found that even though some communication occurred in Maryland, the tortious *27 act was complete upon the release, which occurred in the District of Columbia.
Forum non conveniens is a more significant practical limitation on venue than express limitations in statutes authorizing venue. [FN106] Forum non conveniens is a doctrine that permits the dismissal of cases on the grounds that the venue is inconvenient. It depends on a showing that venue lies in another, more convenient place. [FN107] Forum non conveniens dismissals are less common in federal *28 than in state courts because transfer to a more convenient forum is straightforward. [FN108] According to the United States Code, "[f]or the convenience of parties and witnesses, in the interest of justice, a district court may transfer any civil action to any other district or division where it might have been brought." [FN109] In addition, the multidistrict litigation procedure allows cases involving one or more common questions of fact pending in different districts to be consolidated for pretrial proceedings based on an application to the judicial panel on multidistrict litigation. [FN110] Forum non conveniens may exert considerable influence on the choice of forum for litigating GII disputes. In Figure 3 above, for example, Jurisdiction B would rarely qualify as a convenient forum, compared with Jurisdictions A, C or D.
In Creative Technology, Ltd. v. Aztech System Pte. Ltd., [FN111] the Ninth Circuit applied an analytical framework that requires the person moving for dismissal to demonstrate the existence of an adequate alternative forum and a balance of private and public factors militating in favor of dismissal. [FN112] The private factors include: "(1) relative ease of access to sources of proof; (2) the availability of compulsory process for attendance of unwilling witnesses, and cost of obtaining attendance of willing witnesses; (3) possibility of viewing subject premises; [and] (4) all other factors that render trial of case expeditious and inexpensive." [FN113] In the case before it, the court thought the first factor favored dismissal because all of the records and most of the witnesses were located in Singapore. The fourth factor favored dismissal because a parallel lawsuit in the Singapore courts was more advanced than the U.S. action. The third factor was neutral, while the second factor militated against dismissal because most of the expert witnesses resided in California. [FN114]
*29 The public interest factors identified in the court's opinion include: "(1) administrative difficulties flowing from court congestion; (2) imposition of jury duty on the people of a community that has no relation to the litigation; (3) local interest in having localized controversies decided at home; (4) the interest in having a diversity case tried in a forum familiar with the law that governs the action; [and] (5) the avoidance of unnecessary problems in conflicts of law." [FN115] The court of appeals agreed with the district court that the first and fourth factors were neutral in effect, while the second factor favored dismissal because both plaintiff and defendant were residents of Singapore and the wrongful acts of copyright infringement occurred there. [FN116] Circuit Judge Ferguson dissented, noting that he was "astounded when I read that it is not convenient to try an American copyright case in an American court for copyright infringement that takes place solely in America . . . [and] by a decision that the convenient place to hold the trial is in Singapore, particularly when the majority have not the slightest idea that a court in that nation would even recognize an American copyright." [FN117]
Greater certainty with respect to the place of litigation can be obtained through the use of forum-selection clauses, which are an important determinant of venue. Historically, such clauses were not favored, but in a series of cases beginning with The Bremen v. Zapata Off-Shore Co., [FN118] the Supreme Court has been increasingly hospitable to them. In Stewart Organization, Inc. v. RICOH Corp., [FN119] the Supreme Court held that federal law rather than state law governs the enforceability of a forum-selection clause in a contract within the diversity jurisdiction of a federal district court. [FN120] Justices Kennedy and O'Connor wrote a concurring opinion explaining that enforcement of forum-selection clauses protects party expectations and furthers vital interests of the justice system, thus extending the rationale of The Bremen to federal courts sitting in *30 diversity. [FN121]
In Carnival Cruise Lines, Inc. v. Shute, [FN122] the Supreme Court made it more difficult to avoid forum-selection clauses even when they are contained in form contracts and not actually negotiated. [FN123] The Court acknowledged that forum-selection clauses are subject to judicial scrutiny for fundamental fairness and suggested that they are unenforceable if there is evidence of their inclusion for the purpose of discouraging parties from pursuing legitimate claims by assigning litigation to a remote and alien forum, by proof that resisting parties lacked notice of the clause or evidence of fraud or other overreaching. [FN124] Nevertheless, it extolled the virtues of such clauses as ways of dispelling confusion about where suits for breach of contract might be brought and defended "sparing litigants the time and expense of pretrial motions to determine the correct forum and conserving judicial resources that otherwise would be devoted to deciding those motions." [FN125] In addition, the facts of the case made the forum designated by the clause--Florida--entirely appropriate and did not support the court of appeals's conclusion of inconvenience to the consumer. [FN126] Although Carnival Cruise Lines was an admiralty case, there is no suggestion in the opinion that diversity or federal question cases would be treated any differently.
The presumption in favor of enforcing forum-selection clauses regularly is extended to clauses in international contracts that provide for litigation in the fora of other countries. [FN127] Further, the forum- selection clause cases have important implications for electronic commerce. They permit reduction of the uncertainty with respect to the place of litigation, and they also reduce uncertainty *31 about the methods of binding the user of an electronic service to a forum-selection clause. Forum- selection clauses not only resolve venue issues; they also resolve personal jurisdiction issues because a valid clause constitutes consent to personal jurisdiction in the forum selected. The Carnival Cruise Lines case reduces the likelihood that a participant in electronic commerce could avoid a forum- selection clause on the grounds that there was no negotiation over it. It should be enough to show that the resisting party had notice of the clause, and that the forum selected by the clause was reasonable in light of the nature of the contract.
D. Service of Process and Attachment
Service of process performs two functions in Anglo-American civil procedure: it represents assertion of judicial power of the forum state over the person of the defendant, and it is the formal means of providing notice to the defendant so that he or she may defend the lawsuit. [FN128] Sections A through C above considered the power dimension of personal jurisdiction. This section considers the notice requirement of procedural due process in the civil litigation context. Service of process is a more important concept for civil procedure than criminal procedure because of the nearly universal prohibition in criminal justice systems against trials in absentia--the analog of a civil default judgment. [FN129]
In early Anglo-American procedure, a defendant subjected to personal jurisdiction was entitled to pre-seizure or pre-arrest notice except for claims involving breach of the peace or malice, in which case personal jurisdiction could be asserted by arresting the defendant under a writ of capias ad respondendum. Even when no breach of the peace or malice was involved, capias would issue if the defendant did not respond to the initial notice. Real claims--those involving what later came to be known as in rem jurisdiction-- always could be commenced simply by attaching or otherwise levying on the property itself, regardless of whether separate notice was given to the person entitled to possession. Now, constitutional due process entitlements prohibit significant deprivation of liberty or property interests until notice is given. Thus, capias and commencement *32 of an action by attachment are reserved for exigent circumstances, such as those which involve a risk that the defendant will abscond or dispose of property before notice can be given. [FN130]
The Supreme Court has noted that "[p]ersonal service of written notice within the jurisdiction is the classic form of notice always adequate in any type of proceeding." [FN131] Other forms of notice may be constitutionally adequate when reasonably calculated under the circumstances to give actual notice. [FN132] Of course, constitutional adequacy is not enough; the method of notice also must be authorized by some affirmative source of law representing the service of process rules of the forum. Additionally, service of process must occur within the time limits set by civil procedural rules, although extensions can be granted by the court. [FN133]
The growing use of digital information technologies introduces the possibility of electronic service of process, including telex, fax and e-mail, [FN134] although these methods of service are not yet countenanced widely. [FN135] Despite the practical feasibility and attractiveness of electronic service of process, authorization for electronic service is thin. On occasion, this form of service has been authorized in particular cases. [FN136] Some commentators have urged *33 authorization of service by facsimile, [FN137] but most jurisdictions only authorize the use of facsimile or service of process after the initial service of summons and complaint. [FN138] Some older state rules authorize telegraphic service, [FN139] but fax service has the advantage that fax machines automatically produce an acknowledgment thus providing a form of proof of service.
Although the federal rules authorize service of process pursuant to the law of the forum state or the law of the state in which service is effected, [FN140] and by the law of foreign countries in which service is to be made or under international treaty, [FN141] the explicit provisions of the federal rule provide only modest possibilities for electronic service. [FN142] The appellate rules permit individual courts of appeals to permit papers to be filed "by facsimile or other electronic means." [FN143] Few state rules permit electronic service. The Federal Rules Advisory Committee, supported by courts in which electronic service experiments were conducted, has dicussed amending the Federal Rules of Civil and Appellate Procedure to permit electronic service. [FN144]
Service of process is not the only way to commence an action against one within the personal jurisdiction of the court if a foreign defendant has minimum contacts with the foreign state. [FN145] An action may be commenced by attaching property within the jurisdiction *34 belonging to the defendant. [FN146] This might seem a remote basis for commencing an action growing out of an electronic communication, until one realizes that the property belonging to the defendant might take the form of obligations under a service contract between the defendant and a service provider within the jurisdiction. [FN147] When jurisdiction is obtained in this way, any judgment must be limited to the value of the property attached, although the rationale for this restriction is unclear, the suggestion that the nature of the jurisdiction effected by valid attachment is personal was made by Shaffer v. Heitner. [FN148]
Thus, in the unlikely event that a cyberspace defendant owned hardware that was physically located within the jurisdiction, and assuming that the defendant satisfied minimum contacts, a plaintiff could commence action by attaching the hardware. Far more likely is a situation in which the defendant has intangible rights to service provided by in-jurisdiction entities. For example, a foreign defendant might be a subscriber to America Online. In such a situation, the obligation by America Online to provide service could be subject to attachment or writ of garnishment. The problem, of course, is that the value of such a service contract is likely small. Of greater potential significance is the possibility that execution of the attachment could effectively cut off the defendant from some or all access in the forum jurisdiction. [FN149]
*35 E. Personal Jurisdiction in Criminal Cases
Despite the infrequency of use of the term, at least some commentators have described jurisdiction over the person in criminal prosecutions as "personal jurisdiction." [FN150] Personal jurisdiction in criminal cases universally is based on physical presence, usually obtained through arrest. Rule 43 of the Federal Rules of Criminal Procedure requires the defendant's presence at the trial. [FN151] Many countries, in addition to the United States, guarantee defendants the right to be tried in their presence. [FN152] In addition, the International Covenant on Civil and Political Rights specifically prohibits trials in absentia. [FN153] Commentators report that the defendant's presence is a requisite for a criminal trial under the national laws of many industrial countries. [FN154] Once the trial commences, however, the continued presence of the defendant is not always required, [FN155] as when the defendant flees during the trial [FN156] or is removed from the courtroom due to disruptive conduct. [FN157] Furthermore, under American law, the defendant's presence is not required in four other situations: [FN158] (1) if the defendant is a corporation, it may be represented solely by counsel; [FN159] (2) if the offenses are punishable by a fine or imprisonment of not more than one year; [FN160] (3) "at a conference or argument upon a question of law"; [FN161] and (4) for a reduction of sentence. [FN162] Accordingly, exercising jurisdiction over international computer crimes requires obtaining custody of the individual defendant which can be obtained through the formal process of extradition or by extralegal means. [FN163]
F. Pre-Trial Extradition
International extradition is a means of obtaining criminal jurisdiction over a defendant not found within the territorial limits of the forum state. [FN164] Extradition arose in the mid-nineteenth century *37 in order to impose legal obligations on signatories to extradite fugitives from other signatories, thus limiting unfettered discretion to grant asylum to any fugitive. [FN165] Reluctance to enter into extradition treaties, however, stemmed from a reluctance to extradite persons charged with political crimes in the requesting state (or to extradite victims of ethnic or religious persecution). Because of the difficulty in defining political crimes, the result was treaties drafted to require extradition only for enumerated crimes, reinforced by the doctrine of specialty, which prohibits prosecution in the requesting state for crimes other than those used as the basis for extradition. [FN166]
The United States has extradition treaties with 100 nations. [FN167] Historically, they were limited to certain crimes, but more recently include any crime subject to the principle of dual criminality: one is not extraditable under the treaty unless the crime for which his extradition is requested violates the criminal laws of both the requesting state and the requested state. [FN168] U.S. policy is to replace enumerated crimes and extradition treaties with the dual criminality principle, subject to a political offense exception. [FN169]
Procedurally, the designated authority of the requesting state makes an official request for the extradition of a particular person. The courts of the requested state, acting upon a formal document issued by the justice or foreign affairs ministry, issue an arrest warrant. Once the target is arrested, he or she is entitled to a judicial hearing in which two findings must be made before extradition can proceed: first, that the arrested person is the person named in the extradition warrant; and second, that there is probable cause to believe that the person named committed the crime identified. [FN170]
*38 The process under international extradition statutes resembles interstate extradition within the United States. Virtually all states within the United States have adopted the Uniform Criminal Extradition Act ("Act"). [FN171] Under the Act, the governor of one state has a duty to arrest and to deliver to the executive authority of any other state any person charged with a crime in the other state who is found in the first state. [FN172] Demands for extradition must be in writing and allege that the accused was present in the demanding state at the time the alleged crime was committed and thereafter fled from the state. [FN173] The writing requirement apparently has not been construed to include or exclude a computer message.
Once the demand has been made, "[i]f the Governor decides that the demand should be complied with, he shall sign a warrant of arrest . . . directed to any peace officer or other person whom he may think fit to entrust with the execution thereof." [FN174] Although the governor has a nondiscretionary duty to honor an extradition request, apparently there are no means of compelling him or her to honor it. [FN175] No person may be returned after an arrest on an extradition warrant without being afforded the opportunity to challenge the legality of the arrest under a writ of habeas corpus or otherwise. [FN176] The statute also allows for arrest without a warrant upon reasonable information that the accused is charged in the courts of *39 any state with a capital crime or a serious felony. [FN177] In the event of such a warrantless arrest, the arrestee may be confined for a period not to exceed thirty days while the governor is given time to issue a warrant under section 6 of the Act. [FN178] The warrantless arrest authority adequately covers situations in which the request for extradition or a less formal request to arrest a fugitive is communicated electronically.
G. Extralegal Seizure
The United States Supreme Court has held that extradition pursuant to a treaty is not the only way in which criminal jurisdiction of someone located in a foreign country can be obtained. As an alternative, the person may be seized and delivered physically into forum-state territory by foreign officers, by private citizens, [FN179] or by forum-state law enforcement officials or military personnel. [FN180] While these cases involve the assertion of U.S. criminal law, their reasoning, based on international law, would support the assertion of criminal jurisdiction by another country over a person seized and removed from the United States. In United States v. Alverez-Machain, [FN181] the Supreme Court held that a federal court did not lose jurisdiction over a criminal prosecution because the defendant had been abducted in Mexico and brought to the United States without resort to the extradition treaty then in effect between the two countries. [FN182] In so holding, it noted that historical views of international law permit jurisdiction to be exercised even though a person or thing is seized abroad in violation of international law, despite the general principle of international law that one government may not exercise its police power in the territory of another state. [FN183] Significantly, the treaty with Mexico did not prohibit abduction nor did it adopt a suggestion of a group of Harvard law *40 professors, advanced as early as 1935, that "[i]n exercising jurisdiction under this Convention, no State shall prosecute or punish any person who has been brought within its territory or a place subject to its authority by recourse to measures in violation of international law or international convention without first obtaining the consent of the State or States whose rights have been violated by such measures." [FN184]
In other words, the remedies for extralegal obtaining of custody do not include dismissal of the prosecution. [FN185] The U.S. Criminal Code [FN186] provides that one who commits an offense against the United States may be arrested and imprisoned for that offense, [FN187] contemplating the need to reach beyond the territorial boundaries of the United States to obtain the presence of some defendants:
Section 3041 [providing the power of courts and magistrates] of this title shall apply in any country where the United States exercises extraterritorial jurisdiction for the arrest and removal therefrom to the United States of any citizen or national of the United States who is a fugitive from justice charged with or convicted of the commission of any offense against the United States, and shall also apply throughout the United States for the arrest and removal therefrom to the jurisdiction of any officer or representative of the United States vested with judicial authority in any country in which the United States exercises extraterritorial jurisdiction, of any citizen or national of the United States who is a fugitive from justice charged with or convicted of the commission of any offense against *41 the United States in any country where it exercises extraterritorial jurisdiction.
Such fugitive first mentioned may, by any officer or representative of the United States vested with judicial authority in any country in which the United States exercises extraterritorial jurisdiction and agreeably to the usual mode of process against offenders subject to such jurisdiction, be arrested and detained or conditionally released pursuant to section 3142 of this title, as the case may be, pending the issuance of a warrant for his removal, which warrant the principal officer or representative of the United States vested with judicial authority in the country where the fugitive shall be found shall seasonably issue, and the United States marshal or corresponding officer shall execute.
Such marshal or other officer, or the deputies of such marshal or officer, when engaged in executing such warrant without the jurisdiction of the court to which they are attached, shall have all the powers of a marshal of the United States so far as such powers are requisite for the prisoner's safekeeping and the execution of the warrant. [FN188]
The most straightforward and powerful remedy for an extralegal acquisition of jurisdiction over the person of a criminal defendant is to dismiss the prosecution. It was this remedy that was rejected in Alverez-Machain. In 1993, however, the English House of Lords reached a contrary conclusion in Ex parte Bennett, [FN189] which mirrored the New Zealand Court of Appeals conclusion in R. v. Hartley. [FN190] The basis for these approaches is the British abuse-of-process concept, in which a court refuses to allow its power to be *42 used to further wrongdoing by the police. [FN191] The rationale for declining prosecution over an unlawfully arrested defendant is thus the same as the rationale for excluding unlawfully obtained evidence.
Although Alverez-Machain says the prosecution could proceed if one were to be kidnapped and brought to court for a crime in the forum state, the forum court may prosecute only for acts which are crimes in the forum. Thus, the court cannot prosecute for a crime under the law of a foreign country. The same limitation would prevent prosecution by a court in the foreign country. It is illuminating to compare this criminal law limitation with long arm service of process in the civil law context. Suppose one were served in country X for suit in a Virginia court for a Virginia tort or breach of contract. The Virginia court would have jurisdiction only if there were purposeful contact with Virginia, and then could apply either Virginia or country X law, depending on the respective interests of Virginia in country X. The same result would be obtained if the tort or the breach of contract were committed in country X. The only difference would be that choice of law would more likely point to the application of the law of country X, and forum non conveniens doctrines would be more likely to make the transfer infeasible, or more likely, the dismissal of the suit in the Virginia courts. If the tort or breach of contract suit were maintained in country X, there would be no difficulty in applying Virginia or country X law.
H. Criminal Venue
The location of a criminal prosecution is determined by the location of the crime and physical custody of the person accused of the crime. A vexing procedural problem arises from the difficulty in determining where a computer crime was committed. A criminal defendant is entitled constitutionally and statutorily to be tried in the place where he allegedly committed the crime. For instance, Article III, section 2 of the United States Constitution guarantees a defendant a trial in the state where the crimes were committed. [FN192] The Sixth Amendment entitles criminal defendants to a trial in "the State and district wherein the crime shall have been committed." [FN193] Federal Rule of Criminal Procedure 18 states that "except as otherwise permitted by statute or by these rules, the prosecution shall be had in a district in which the offense was committed." [FN194] These criminal venue guarantees do not apply to the states [FN195] nor to foreign countries. [FN196]
Many federal criminal statutes provide for venue, but when they donot, the prevailing practice is to apply the "verb test" to determine the nature of the crime alleged and the location of the act or acts constituting it. Under this test, the court looks to the verbs defining the proscribed act and determines where the conduct specified by those verbs actually occurred. That is the place of the crime and the place for prosecution. [FN197] As mentioned above, the United States Constitution entitles a criminal defendant to prosecution in the state and district in which the crime was committed. This does not mean, however, that there is a single proper situs for trial in the case of a crime that affects more than one district. [FN198] The verb test properly takes into account a number of factors: "the site of the defendant's acts, the elements and nature of *44 the crime, the locus of the effect of the criminal conduct, and the suitability of each district for accurate factfinding." [FN199] Usually, the site of the defendant's acts provides proper venue, but other factors may give another venue equal standing. [FN200] In particular, the place where the effects of the crime are felt may be an appropriate venue. [FN201] Unfortunately, the verb test for criminal venue creates as much ambiguity as the minimum contacts test for personal jurisdiction. For instance, suppose a criminal statute uses the verb "to send, transmit, or distribute." [FN202] It is far from certain in Figure 3 above whether any of the actors, the content originator, server X or server Y can be said "to send, transmit, or distribute" the item into jurisdiction B.
V. Choice of Law
Choice of law, application of foreign criminal law and extraterritorial application of criminal law all are aspects of jurisdiction to prescribe; [FN203] the application of a nation's law to a dispute presupposes that the nation has jurisdiction to prescribe rules applicable to that dispute. The term "choice of law" usually is confined to the universe of civil disputes because of the basic rule that criminal jurisdiction to prescribe is local. Actually, however, criminal jurisdiction to prescribe is not strictly local. [FN204] It is thus appropriate for *45 purposes of this Article to consider choice of law or jurisdiction to prescribe in a broad sense, as including both civil and criminal matters. The following sections work through the analytical framework for choice of law in civil matters and then consider three bases for applying the substantive criminal law outside the territory in which the offense actually occurred.
A. Choice of Law in Civil Cases
As explained above, choice of law is related to personal jurisdiction, venue and enforcement of judgments. Nevertheless, it is a distinct inquiry; a forum may have personal jurisdiction and venue, and nevertheless be obligated by its choice of law rules, perhaps reinforced by the Constitution, to apply the substantive law of another jurisdiction. Historically, choice of law was driven by formal rules such as lex locus contractu, which required the forum to apply the substantive law of the place of contracting to a contracts dispute, and lex locus delicti, which obligated the forum to apply the substantive law of the place of the wrong to a tort claim.
Although these rules were applied by American courts virtually universally, they had their origin in seventeenth and eighteenth century private international law, which adapted itself well to the American federal system with the states as independent sovereigns. [FN205] Joseph Story's 1834 treatise onconflict of laws was the first effort to synthesize standard rules in this area, and his preface noted the importance of the subject for trade between foreign states and between the different states of the American union; he also noted that much of the learning on the subject was derived from the work of continental European civil law commentators. [FN206]
Despite their unambiguous expression, the traditional rules required considerable interpretation. For example, the place of contracting might be the place of making a contract, or it might be the place of performance. [FN207] The place of wrong in a tort claim might be the place where the defendant acted or the place where the injury *46 occurred. [FN208] By the latter third of the twentieth century, most courts and commentators agreed on a more flexible "interests analysis" approach to resolving choice of law questions. In many cases, the traditional rule produces the same results as the interest analysis. [FN209]
The analysis for tort claims--such as a GII defamation claim--requires determining the state with the most significant relationship to the occurrence and the parties, including consideration of the place where the injury occurred, the place where the conduct causing the injury occurred, the domicile, residence, nationality, place of incorporation and place of business of the parties, and the place where any relationship between the parties is centered. [FN210] Contract cases are adjudicated according to the law chosen by the parties or, in the absence of any such chosen law, by the law of the state which has the most significant relationship to the transactions and the parties with respect to a particular issue. [FN211] The most significant relationship is determined based on the place of contracting, the place of negotiation of the contract, the place of performance, the location of the subject matter of the contract, and the domicile, residence, nationality, place of incorporation and place of business of the parties. [FN212] "If the place of negotiating the contract and the place of performance are in the same state, the local law of this state will usually be applied . . . ." [FN213]
There are limitations on a party's choice of law. The general *47 rule is that their choice will govern "if the particular issue is one which the parties could have resolved by an explicit provision in their agreement directed to that issue." [FN214] In other words, if the parties could explicitly provide detailed ground rules for determining whether performance met the requirements of the contract, they could choose the law of a state to decide whether performance was sufficient. Even if they could not have explicitly governed the issue, they can choose a state's law unless the chosen state has no reasonable relationship to the parties or the transaction and there is no other reasonable basis for the party's choice, or if application of the law of the chosen state would contravene a fundamental public policy of a state which has a materially greater interest in determination of the particular issue. [FN215] The Restatement of Conflicts of Laws specifically contemplates the possibility of party choice of the law of a jurisdiction well known to them even though that state has no relationship to the contract:
When contracting in countries whose legal systems are strange to them as well as relatively immature, the parties should be able to choose a law on the ground that they know it well and that it is sufficiently developed. For only in this way can they be sure of knowing accurately the extent of their rights and duties under the contract. [FN216] The language of the Restatement apparently contemplates that the only law eligible for choosing is the law of a particular state, [FN217] thus excluding the possibility of the parties choosing general legal principles such as UNIDROIT or lex mercatoria.
Constitutional due process requirements constrain choice of law to some extent, but Supreme Court jurisprudence on the relationship between due process and choice of law is in an uncertain state. In Sun Oil Co. v. Wortman, [FN218] five Justices agreed on the soundness of the doctrine that the Constitution does not bar the application of a forum state statute of limitations to claims that in their substance are governed by the law of a different state. [FN219] The Court held that "[t]he Full Faith and Credit Clause does not compel 'a state to substitute the statutes of other states for its own statutes *48 dealing with a subject matter concerning which it is competent to legislate."' [FN220] The procedural rules of its courts are matters on which a state is competent to legislate. Thus, a state may apply its own procedural rules to actions litigated in its courts. A statute of limitations is procedural for this purpose, and therefore, a state court may apply its own statute of limitations. [FN221] The issue on which controversy existed, however, was whether a statute of limitations should be considered procedural for this purpose. [FN222] The early cases, looking to international law, uniformly held that the forum statute of limitations governed. [FN223] The majority rejected the argument that Guaranty Trust v. York, [FN224] which treated statutes of limitations as substantive for Erie purposes, should govern because the words "substantive" and "procedural" do not have precise content. [FN225] Because of the long-standing tradition under the Full Faith and Credit Clause, a due process challenge similarly must fail. [FN226] Justice Brennan reached the same result through interest analysis:
Given the complex of interests underlying statutes of limitations, I conclude that the contact a State has with a claim simply by virtue of being the forum creates a sufficient procedural interest to make the application of its limitations period to wholly out-of-state claims consistent with the Full Faith and Credit Clause. This is clearest when the forum State's limitations period is shorter than that of the claim State. [FN227]
On the other hand, when, as in the case before him, the forum state's limitation period was longer, the interests of the forum state were less clear. Because the various choice of law arguments were fairly balanced and somewhat ambiguous, Justice Brennan concluded that the Constitution permitted the choice the forum state made to apply its own limitations period. [FN228] The balance of the case involved disagreement over whether Kansas correctly interpreted *49 the substantive law of Texas, Oklahoma and Louisiana. [FN229]
Three years earlier, in Phillips Petroleum Co. v. Shutts, [FN230] the Supreme Court reviewed a case in which the Kansas courts applied Kansas law to a class action, notwithstanding that over ninety-nine percent of the gas leases involved in the case and ninety-seven percent of the plaintiffs had no connection with the state of Kansas except for the lawsuit. [FN231] The Court began by noting that in Allstate Insurance Co. v. Hague, [FN232] a plurality recognized that the Due Process Clause and the Full Faith and Credit Clause provide only modest restrictions on the application of forum law: " 'for a State's substantive law to be selected in a constitutionally permissible manner, that State must have a significant contact or significant aggregation of contacts, creating state interests, such that choice of its law is neither arbitrary nor fundamentally unfair."' [FN233] Applying this standard, the Court considered the fact that the defendant owned property and conducted substantial business in the state, and the fact that none of the plaintiffs had opted out of the class too tenuous to support application of Kansas law. Fairness, it held, depends importantly on the expectations of the parties, and "there is no indication that when the leases involving land and royalty owners outside of Kansas were executed, the parties had any idea that Kansas law would control." [FN234] It declined the Kansas Supreme Court's suggestion that more relaxed standards should apply to nationwide class actions. [FN235] Justice Stevens would have affirmed, however, because he saw no real conflict among the laws of the various states having significant contact. [FN236]
Because GII transactions frequently permit actor, intermediaries and victim to be widely separated geographically, interest analysis requires careful scrutiny of the technology and the facts of a particular transaction. Consider a hypothetical situation in which an Internet server in California makes available material that users in Tennessee find offensive. In a lawsuit filed by the Tennessee *50 users against the California server operator, say for intentional infliction of emotional distress, the choice of law question would be whether Tennessee or California law should apply. Assuming there is a difference between the substantive law of the intentional-infliction tort in the two states, Tennessee may have the most significant interests. It seeks to protect its residents against extreme emotional distress intentionally inflicted by others. California, of course, also has interests, such as in seeing its citizens free to publish material that would not under California standards be tortious. But under the Supreme Court's tests outlined above, there would be sufficient Tennessee interest to allow the application of Tennessee law and the refusal of a Tennessee court to apply California law.
Cases involving defamation claims against interstate publishers offer some analogies to GII activities. [FN237] In these cases, there is a rule that the most significant interest is where the "sting was felt," which usually leads to the domicile of the plaintiff in defamation cases. [FN238] Many courts apply a nine-factor test in defamation cases involving more than one state: (1) the state of plaintiff's domicile; (2) the state of plaintiff's principal activity to which the defamation relates; (3) the state where plaintiff suffered greatest harm; (4) the state of publisher's domicile or incorporation; (5) the state where defendant's main publishing office is located; (6) the state of principal circulation; (7) the state of emanation; (8) the state where *51 libel was first seen; and (9) the state of the forum. [FN239] Each factor, however, does not receive the same weight. For example, in Hoffman v. Roberto, [FN240] the plaintiff filed a defamation claim arising from union officials' statements in telex messages sent to various states claiming that the plaintiff diverted funds of an employer. Because of the multiple states of publication, the district court looked to the place of conduct, the residences or places of business of the parties and the place where the relationship was centered. [FN241] As the first two of these inquiries were indeterminative, the court concluded that Michigan law, where the employment relationship was centered, should be applied. [FN242]
Using the preceeding defamation choice of law cases as an analogy, domicile will likely govern defamation actions arising in the GII. Invasion of privacy claims, like defamation claims, are centered where the plaintiff lives and conducts his or her affairs, because that is where the privacy interest exists. [FN243] The same result is appropriate for intellectual property infringement actions [FN244] unless the party arguing for the choice of different law can show that adversely affected markets are located elsewhere. Intellectual property claims primarily involve injuries to markets, and the location of the markets should drive the interest analysis.
B. Extraterritorial Application of Criminal Statutes
International and American law recognize the legitimacy of giving extraterritorial effect to the criminal laws of a national sovereign. The extension of criminal jurisdiction is particularly likely in two circumstances: when the actor, and potential criminal defendant, is a citizen of the state whose law is to be applied, and when the "effects test" shows that a non-national has engaged in extraterritorial conduct with the intention or the likelihood that it will have effects in the country whose law is to be applied. [FN245] In addition to *52 international crimes, international law recognizes, with some dispute, the possibility of passive personal jurisdiction to apply to criminal law. This basis allows a nation to prosecute anyone committing a crime against one of that nation's citizens, regardless of where the crime was committed. [FN246]
Both branches of extraterritorial criminal jurisdiction have counterparts in civil personal jurisdiction and choice of law analysis. The nationality branch of extraterritorial criminal jurisdiction corresponds to extending personal jurisdiction in a civil case over one who is domiciled in the forum state. The effects test corresponds to minimum contacts analysis in civil personal jurisdiction, especially insofar as it authorizes the assertion of personal jurisdiction in a civil case over one who acts outside the jurisdiction intending that contact with the jurisdiction will result from that person's conduct. [FN247] The cases reviewed above involving the assertion of personal jurisdiction over publications that publish outside the state, but intending that their publication come into the state, are clear *53 examples. [FN248]
In general, there is no constitutional bar to the extraterritorial application of U.S. penal law. [FN249] Extraterritoriality is determined by looking to congressional intent, presuming that Congress does not want to violate international law. [FN250] Thus, unless Congress explicitly directs otherwise, extraterritoriality is valid to the extent permitted by international law. [FN251] Additionally, choice of law does not arise in criminal cases in the same way that it arises in civil cases. The basic reason for this is that crimes were not traditionally considered transitory [FN252] and thus a court either had jurisdiction or it did not. When a court had jurisdiction, it applied its own law, [FN253] but a wide range of extraterritorial conduct might still be criminal. [FN254] Therefore, one could be criminally liable in state A for computer-triggered conduct in state B that caused injury in state A as long as state A expressly prohibited extraterritorial conduct of that character and the actor knew or should have known of the adverse effect in state A. [FN255] Extraterritorial application of criminal law by multiple nations has some advantages, at least when the substantive *54 criminal law being applied extraterritorially is more or less the same. "Concurrent jurisdiction had the considerable virtue of permitting any nation catching an offender to act upon his wrongs--without resolving the fine points of a theory of exclusive jurisdiction, and without facing the political, moral and legal concerns of aiding a foreign system of justice." [FN256]
C. Enforcement of Foreign Criminal Law
As explained above, choice of law does not function in the criminal arena the way it does in the civil arena. Strict localization of criminal law [FN257] leads to the usual conclusion that the courts of one jurisdiction may not apply the criminal laws of another. Therefore, "[a] court in the United States may try a person only for violation of United States law, not for violation of the penal law of a foreign state." [FN258] The Restatement (Third) of Foreign Relations notes that some civil law countries try persons whom they cannot extradite for crimes committed in other countries. [FN259] This constitutes a kind of "transitory" criminal action.
In common law systems, such as at the state level in the United States, one can argue that a state court would have the power to try *55 an actor for the foreign crime because state courts are not limited by the separation of powers doctrine of the federal Constitution. There are at least some states in which common law crimes exist, albeit mostly in theory. In such a state, the courts are not bound only to those crimes declared expressly to be so by statute. Furthermore, they would have the power to apply foreign criminal definitions, unless some overriding constitutional privilege would be infringed by doing so. This would, of course, be a revolutionary idea; the linkage between territorial sovereigns and substantive criminal law is very strong. [FN260]
Even if the territorial problems described above are overcome, there still exists the need for an affirmative basis for the criminal proscription for which the defendant is prosecuted. It surely would violate procedural due process to prosecute for a crime that was not defined as such when the defendant acted. [FN261] Therefore, the prosecution would need to establish either a sufficient connection between the defendant's conduct and the foreign jurisdiction whose criminal law is being relied on, or else a basis in international law as might be the case with piracy or terrorism.