Courts channel the coercive power of states, and therefore have whatever personal jurisdiction is conferred on them by the states for which they act, subject to some superseding limitations on the legislative power of states, imposed by international law and American federalism concepts. International lawyers call this legislative power "jurisdiction to prescribe."

Typically, courts are given personal jurisdiction over anyone served with process while physically within the territory of the state, and also are given jurisdiction over any property physically located within the territorial limits of the state when that property is attached. Service of process and attachment are formal mechanisms for asserting personal jurisdiction over persons or property and also serve the purpose of giving constitutionally required notice of the pendency of a judicial proceeding that may result in deprivation of property. It makes no difference whether a person thus within the personal jurisdiction of a court wants to participate or not. The power is coercive. Similarly, it makes no difference whether the owners of interests in property thus subjected to personal jurisdiction want to participate or not. The power over the property is coercive.

The challenge is to translate these personal jurisdiction concepts into the less coercive regimes of dispute resolution institutions in cyberspace. In some modes of dispute resolution, personal jurisdiction does not matter because coercion plays no role whatsoever in the effort to resolve disputes. In mediation, an ombudsman processes for example at no point are participants or interested parties obligated to do anything against their will; these processes are voluntary from beginning to end.

Arbitration, even the limited form used in the Virtual Magistrate pilot, is different. The arbitration award is meant to be enforceable through coercive power if necessary. For example, a valid arbitration award can be registered with a court and thereafter enforced like a court judgment. In other words, at the end stage, arbitration is almost indistinguishable from litigation in the regular courts. But coercion is not available at the first stage. It is at the first stage where personal jurisdiction is at issue in arbitration systems. Because arbitration is a private, not a governmental, process, the state does not compel people to participate and does not confer personal jurisdiction on arbitrators in the absence of party consent. In other words, the source of personal jurisdiction in all arbitration systems (outside court annexed and certain regulatory arbitration systems) is consent.

The traditional means of expressing consent to arbitrate is through a written arbitration agreement or an arbitration provision in a broader contractual document. Insisting on that degree of formality and fixation on paper would cripple the Virtual Magistrate system and its expected progeny.

There is little doubt that consent to arbitrate, if sufficiently explicit and voluntary, can be expressed entirely through electronic means. This is the necessary logical result of the proposition that contracts can be made electronically. For example, if the Virtual Magistrate complaint form concluded with the following statement, "Please type your first and last name in the space provided if you intend by this signature to submit your dispute to final and binding arbitration," and the complainant types her name in the space provided that would manifest complainant consent to arbitration. A similar statement and response in a submission responding to complaint would signify the respondent's consent to arbitrate.

The more interesting questions involve extension of personal jurisdiction beyond such explicit agreements to arbitrate. Such personal jurisdiction can exist if consent is implied, either implied in law or implied in fact. The law does not imply consent to arbitrate outside a few limited circumstances, mostly involving collective bargaining agreements, a circumstance in which employees within the bargaining unit statutorily relinquish their power to contract for themselves to the lawfully designated exclusive representative for the unit (usually a trade union).

So the interesting questions involve implied in fact consent. There is a wide range of circumstances under which one acting after notice that certain consequences will result from the act is deemed to have impliedly consented to whatever the notice says. Thus, one entering premises prominently posted with signs saying, "If you enter here you are consenting to have your hand luggage and purses searched," would impliedly consent to such searches. How this mechanism for imposing implied consent on persons who enter into certain physical spaces be used with respect to virtual spaces? No doubt, America Online and CompuServe take the legal position that anyone who participates as subscribers in their services impliedly consents to all the terms of their subscriber agreements and operating rules, even though they are changed from time to time and may not ever actually be read by the subscribers thus bound.

It is important to recall that this kind of implied in fact consent probably depends on the existence of a choice. If the choice is absent, the purported consent cannot be said to be voluntary, which is the core idea of contract obligation. Determining the voluntariness of consent is the centerpiece of debates over contracts of adhesion. Choice regularly is litigated in the employment context with respect to disclaimers of employer obligation found in handbooks and waivers of statutory and common law rights when employees are terminated.

Another interesting possibility is to extend the in rem jurisdiction idea to cyberspace. Personal jurisdiction over property, asserted by attachment, is a form of in rem jurisdiction. Suppose the Virtual Magistrate system's virtual Attorney General Tierney files a complaint against a particular item carried by a service provider. The purpose of the proceeding is not to impose any liability on the service provider or on the originator of the message directly, in an institutional sense, but only to "condemn" or to declare forfeit the accused item. To be sure, this is somewhat different from the usual civil in rem proceeding. Complaining party Tierney does not seek to have title to the accused item transferred to him; nor did he seek to have the item sold to generate proceeds to pay him. Rather, he simply seeks to have it destroyed. But these factual differences do not have any apparent bearing on the legality of the in rem proceeding.

Quite recently, the Supreme Court recognized the legitimacy of in rem proceedings even when they extinguish the rights of innocent parties. In characterizing the historical context for this holding, one of the dissenting justices observed that in rem forfeiture had an old and undoubted pedigree as a mechanism of providing relief to victims when the person with ownership interest in the property was beyond the effective reach of legal institutions available to the victim. So also is in rem jurisdiction a useful idea for a victim or a representative of the public interest who can bring a thing (typically an electronic file) within the power of the arbitration tribunal. Then, it is unnecessary to inquire whether the owner, author, or placer of the thing is within the personal jurisdiction of the arbitrators.

The weakness with the in rem analysis is that it does nothing to affect the rights of the owner of the thing vis a vis the person or entity holding it. Suppose the Virtual Magistrate says that the accused thing (posting) should be condemned or forfeited, and the entity with custody over it -- typically an online service -- accordingly removes it and erases it. Then suppose the author of the accused posting sues the intermediary for removing it and erasing it, on a contract or trespass theory. The intermediary will want to use the Virtual Magistrate decision as a defense, but it is not altogether clear how such a defense could be legitimated at common law. One possibility is the common law privilege of "general justification." It would be useful to think more about theories for defending such claims based on a Virtual Magistrate decision supported by this type of in rem jurisdiction.