The courts' early attempts to grapple with the jurisdictional implications of the Internet were greeted widely with academic despair. The technology seemed to defy our sense that governmental authority could continue to be territorially allocated. Ordinary Internet activity such as the hosting of a web page, participating in an eBay auction, or posting to an electronic discussion had potentially global impact even though the actors may have perceived their behavior as essentially local. The technology, seemingly overnight, transformed all of us into global actors, potentially subject to the jurisdictional and regulatory authority of many governments. Numerous commentators asserted that existing jurisdictional doctrine was not up to the challenge of sorting it all out.¹1

       My position is that the Internet does not pose unique jurisdictional challenges. ²2 People have been inflicting injury on each other from afar for a long time. Although the Internet may have increased the quantity of those remote consequences, it has not created problems that are qualitatively more difficult. Recent decisions have borne out my sense that courts would eventually work through the technological befuddlement evident in some early cases and develop a more or less consistent and predictable body of law. ³3

       My doctrinal optimism is fueled, at least in part, by what might fairly be deemed an unconventional account of the Court's due process jurisprudence. The Internet has made clear that a given assertion of jurisdiction must be evaluated in relation to the regulatory interests of other states. Due process does not simply look to the bilateral relationship between a defendant and a forum. Rather, it tests jurisdictional authority in relation to the nature of defendant's activity, and the regulatory claims of all states with an interest in that activity. It is multilateral, not bilateral. It is claim-specific, not trans-substantive. In the time allotted, I can do little more than sketch-out the broad parameters of my argument. In particular, I want to focus on a tension in due process doctrine between the jurisdiction-constraining principles of World-Wide Volkswagen, ⁴4 and the jurisdiction enabling principles of Calder v. Jones⁵.5 The key to untangling the jurisdictional challenge of cyberspace is in understanding the relationship between those poles. A. Situating the Internet Between Calder and Volkswagen

       World-Wide Volkswagen poses the biggest constraint on jurisdiction over defendants who have inflicted injury over the Internet. The Court's holding that Seaway, a car dealer in New York, was not subject to jurisdiction in Oklahoma, the place where the allegedly defective car caused injury, could be seen as analogous to Internet-based claims. Local behavior of a defendant that causes remote injury will not subject the defendant to jurisdiction in the place of injury unless the defendant "purposefully availed" itself of the connection with the forum state. Even a foreseeable interstate consequence of local behavior will not suffice unless that consequence is deemed "purposeful." Seaway's indifference to the post-sale location of the car was fatal to the argument that it acted purposefully toward Oklahoma.

       The central problem with applying this principle to Internet-based claims is figuring out how to characterize Internet behavior: is it local conduct, analogous to selling a car in one place that causes injury in another; or is it interstate behavior, more akin to purposely sending a defective product into the forum state? ⁶6 Has the defendant by virtue of posting information to the Internet sought out the resulting connection with the forum, or is the connection simply a foreseeable, but not desired consequence?

       Courts framing the issue in this way initially differed in their characterizations in cases where defendant had caused injury (typically an intellectual property claim) through its maintenance of a web site. Following the lead of an early exposition of the problem, ⁷7 most courts have now settled on a particularly arbitrary test of whether the offending web site is "interactive" or "passive." Thus, if a web site simply conveys information, but does not have the programming mechanism to consummate a sale, the courts will not find purposeful availment. Recent decisions have overwhelmingly characterized most web pages in question as insufficiently interactive, and have denied jurisdiction. ⁸8 But the passive/active doctrine has not eliminated all inconsistency, since there is plenty of room to argue about the definitions of the categories. Some courts have even found the provision on a web site to send an email to the operator to constitute an interactive web site that will subject the operator to personal jurisdiction. ⁹9

       This passive/interactive test represents an egregious failure of legal imagination. Lacking an adequate conceptual account of why purposeful availment matters, the courts have reverted to the thinking about jurisdiction in Pennoyer-like physical terms: the interactive web site looks like the defendant is really operating a branch store in the forum. If defendant is simply conveying information, he's not "really" here.

       As Justice Stone did in International Shoe, it clearly is time once more for the courts to see through a jurisdictional fiction. ¹⁰10 The level of interactivity of a web site has no obvious connection to the reasons why it might be unreasonable to subject a defendant to jurisdiction. Whether a sale is consummated through an automated web site, or through a subsequent, non-cyber contact with the web operator does not change the nature of the web operator's connection with the forum.

       I want to suggest, however, an implicit element common to all the cases that seems to do more work: a concern with the appropriate allocation of regulatory authority over the actions of the defendant. In Volkwagen, the Court was concerned that Oklahoma's assertion of jurisdiction would deprive the New York defendant of its ability to control its amenability to jurisdiction; the automobiles it sold would become its "agent" for service of process. ¹¹11 Imbedded in this notion of "control" is the problem of Oklahoma exceeding its appropriate regulatory authority; it was, as the Court suggested earlier in the opinion, stepping on New York's toes. ¹²12 Seaway could have indeed controlled its jurisdictional exposure: it did not have to sell cars in New York, or anywhere else within automotive proximity of Oklahoma. The problem was not that Seaway could not prevent its exposure to Oklahoma jurisdiction; rather that Seaway could not control its jurisdictional exposure without unduly constraining its behavior in New York. Oklahoma had no right to affect New York behavior to that extent. This "spill-over" effect is the central difficulty in basing jurisdiction on territorial effects rather than territorial behavior. ¹³13

       This concern over the spill-over effects is clearly reflected in the Internet jurisdiction cases. The courts consistently recognize that the jurisdictional claim of a forum based solely on the undifferentiated consequence of the defendant's action in the forum (compared with its effects on other places) will impair the freedom of people to use the Internet. ¹⁴14 It takes more than a harmful effect within the forum to give that state authority to regulate extraterritorial behavior.

       This jurisdictional constraining principle is tempered by two jurisdiction-enabling factors. First, the spill-over effect is implicated only when a forum's jurisdictional claim is undifferentiated. That is to say, if the defendant's conduct did not ex ante endanger forum residents more directly than it threatened residents of any other state, the forum will normally lack regulatory authority over defendant's conduct. However, a forum's assertion of regulatory authority over forum-directed behavior will not inhibit the defendant's non-forum directed activity and will therefore be permitted. The jurisdictional authority over forum-directed activity is clearly recognized by Calder v. Jones. In Calder, the Court permitted California to assert jurisdiction over a Florida writer who could not be said to have purposefully availed himself of California law, but who defamed a California resident. The Court held that defendant's action in targeting his tortious conduct toward a California domiciliary gave California the right to assert jurisdiction over him, just as California would have jurisdiction over someone who shot a bullet into the state.

       The only way to harmonize Volkswagen and Calder is to recognize that they both deal with the problem of regulatory precision. Both require forum directed behavior as a predicate to asserting long-arm jurisdiction. Purposeful availment serves the same function as an intentional tort; both enable effects-based jurisdiction without risking regulatory spill-over.

       The use of the "targeted conduct" rationale for asserting jurisdiction over defamation claims is not without problems. While treated as an "intentional tort" (at least for public-figure plaintiffs), defendants frequently assert that their speech was not actionable. Defendant's behavior can be considered "targeted" only if we assume the validity of plaintiff's complaint. It may well turn out after trial that defendant's speech was not actionable - and thus not "targeted"--, yet defendant will have incurred the cost of defending the lawsuit in plaintiff's chosen forum. Assertions of jurisdiction can thus inhibit lawful conduct outside of the forum - the same spill-over problem the Court was concerned about in Volkswagen. The courts at one time protected against this chilling effect by raising the jurisdictional threshold for defamation claims, ¹⁵15 but it abandoned that approach in Calder. Calder thus represents a potentially problematic weapon in the context of Internet defamation.

       It is therefore not surprising that most courts dealing with defamation published over the Internet have, in effect, declawed Calder by concluding that the mere fact that plaintiff would feel the impact of the defamation in his home state does not make defendant's conduct "targeted." Defendant must affirmatively direct his speech at defendant's domicile. ¹⁶16 The distinction between merely knowing that defamation will affect plaintiff in his home, and "targeting" that defamation at the defendant's home state is not obvious to me, particularly given the fact that the author of the defamation in Calder had no hand in disseminating the defamation in California. ¹⁷17 Nonetheless, the courts' reluctance to extend Calder to Internet-based speech is consistent with concerns about regulatory spill-over and are a positive development.

       Although less evident in the cases, I want to suggest a second limiting principle on Volkswagen: regulatory spill-over is only a problem when the underlying activity is of value to other states. Suppose a defendant designed a web site with the sole purpose of spreading a computer virus to any computer that logged onto the site. The defendant could not be said per Volkswagen to have purposefully availed itself of any particular forum (outside of his home state). Nor could his behavior be deemed "targeted" in Calder terms. Yet our intuition is that any victim would be permitted to sue in the state in which injury was suffered, and the defendant would not be heard to complain that he could not control his amenability to jurisdiction. That intuition is consistent with a model of due process keyed to regulatory precision. We don't care about regulatory spill-over where no state would be troubled by inhibiting defendant's conduct.

       It may feel circular for courts to consider the value of defendant's activity in deciding which state has authority to regulate the activity, especially since different states may assess that value differently. But as the Court made clear in Volkswagen, the problem of jurisdiction involves enforcing "the shared interest of the several States in furthering fundamental substantive social policies" as much as it does allocating authority. ¹⁸18 B. Why Redish is Wrong

       My disagreement with Professor Redish on jurisdictional matters goes well beyond the Internet context. He rejects the notion that the "process" due under the 14th Amendment has anything to do with the "authority" of court in question. Rather, he asserts, due process should be used only as a guarantee of a litigant's participatory opportunities in the adjudication. Jurisdiction under this measure would be constitutionally excessive only when it imposed such an enormous burden on the defendant that it effectively denied defendant the opportunity to fully participate in the adjudication. ¹⁹19 Redish would thus collapse the right guaranteed under International Shoe with the "procedural" due process protections of Mathews v. Eldridge²⁰.20 To the extent that Supreme Court has any role to play in allocating the regulatory authority of different states, it should do so exclusively by way of controlling choice of law under the Full Faith and Credit clause. ²¹ 21

       I have elsewhere discussed some of the problems with this approach. These include the fact that virtually all of the Court's precedents going back to Pennoyer v. Neff are at odds with his understanding; that interjurisdictional disputes are not adequately resolved through constraints on choice of law; and that the marginal "burden" imposed by even the most exorbitant assertion of jurisdiction is trivial in comparison with burdens imposed without constitutional difficulty on intrastate litigants every day. ²²22

       I want to focus here on the particular problem of applying his approach to the Internet. Redish concludes that the Court will be forced to reevaluate the wisdom of purposeful availment because the "test fails to provide states with the ability to effectively protect their citizens by providing a convenient forum in which to enforce claims against out-of-staters who have caused them harm." ²³23 From my perspective, if there is anything uniquely problematic about the Internet, it is that existing jurisdictional doctrine may not confer adequate protection on defendants. I know of no case in which the Internet makes it harder to assert jurisdiction than if the comparable injury were inflicted through a non-electronic means. In both cyber and non-cyber cases, courts permit jurisdiction over actions targeted at the forum, and otherwise resist mere effects-based authority. The Internet makes it potentially easier for a state to assert jurisdiction over an out-of-state defendant on the theory that by virtue of participating in a global medium, the defendant might be deemed to have targeted its conduct everywhere and purposefully availed itself of the benefits of the forum.

       If anything, the technologically evolving nature of the Internet promises to minimize the regulatory spill-over problems inherent in effects-based assertions of jurisdiction in the "molecular world." Technology offers the promise of a regulatory precision that would not otherwise be possible. Consider again Volkswagen. The case poses a stark choice between subjecting Seaway to jurisdiction anywhere the car happened to explode, or insulating defendant from the out-of-state jurisdictional consequences of its behavior. There was no way Oklahoma could exercise regulatory authority without affecting Seaway's behavior in New York generally. It was not possible to have one rule for cars headed for Oklahoma and another for cars staying in New York because defendant had no control over where the cars were going.

       Much has been made about the anonymity of transactions over the Internet. ²⁴ But in fact, the Internet makes it possible for a defendant, in at least some cases, to tailor its conduct in regard to the specific location of the plaintiff with relatively little spill-over on defendant's conduct toward others. ²⁵ An Internet seller of exploding cars - or, more realistically, defective software - can control where his actions have consequences with relative ease. He has the ability to limit access to his web site. He can choose to exclude people from Oklahoma. All he has to do is ask for a credit card or digital certificate. Oklahoma can thus exercise regulatory authority over defendant's transactions with citizens of Oklahoma without substantially affecting defendant's other business. Once the courts recognize that due process, at bottom, is a constraint on regulatory spill-over, states will be far more empowered to protect their citizens than they are today. 1See, e.g., Robert W. Hamilton & Gregory A. Castanias, Tangled Web: Personal Jurisdiction and the Internet, 24 Litig. 27 (1998); Martin H. Redish, Of New Wine and Old Bottles: Personal Jurisdiction, The Internet, and the Nature of Constitutional Evolution, 38 Jurimetrics 575 (1998). 1Allan R. Stein, The Unexceptional Problem of Jurisdiction in Cyberspace, 32 Int'l Lawyer1167 (1998); Allan R. Stein, Frontiers of Jurisdiction: From Isolation to Connectedness, 2001 U. Chi. Legal F. 373. 1Accord, Suzanna Sherry, Haste Makes Waste: Congress and the Common Law in Cyberspace, 55 Vand. L. Rev. 309, 364-74 (2002). 1World-Wide Volkswagen v. Woodson, 444 U.S. 286 (1980). 1465 U.S. 783 (1984). 1See, e.g., Gray v. American Radiator & Standard Sanitary Corp., 22 Ill. 2d 432, 176 N.E.2d 761 (1961). 1Zippo Manuf. Co. v. Zippo Dot Com, Inc., 952 F. Supp. 1119 (W.D. Pa. 1997). 1See, e.g., Fix My PC, L.L.C. v. N.F.N. Assoc., Inc., 48 F. Supp. 2d. 640 (N.D. Tex. 1999); Mid City Bowling Lanes & Sports Palace, Inc. v. Ivercrest, Inc., 35 F. Supp. 2d 507 (E.D. La. 1999); Riviera Operating Corp. v. Riviera Hotel & Casino, 29 S.W. 3d 905 (Tex. App. 2000); Efford v. The Jockey Club, 796 A.2d 370 (Pa. Super. Ct. 2001). But see Hsin Ten Enterprises USA, Inc. v. Clark Enterprises, 138 F. Supp. 2d 449 (2000). 1See, e.g, Hasbro, Inc. v. Clue Computing, Inc., 994 F. Supp. 34 (D. Mass. 1997); Blumenthal v. Drudge, 992 F. Supp. 44 (D.D.C. 1998). 1International Shoe Co. v. Washington, 326 U.S. 310 (1945) ("corporate "presence" test is legal fiction) 1444 U.S. at 297. 1444 U.S. at 294. 1Accord, Jack L. Goldsmith, Against Cyberanarchy, 65 U. Chi. L. Rev. 1199 (1998). 1See, e.g., Digital Equipment Corp. v. Altavista Techn., Inc., 960 F. Supp. 456, 463 (D. Mass. 1997). Accord, American Libraries Assn. v. Pataki, 969 F. Supp. 160 (S.D.N.Y. 1997) (invalidating on Commerce Clause grounds New York statute criminalizing digital dissemination of sexual material). 1See, e.g., Curtis Pubg. Co. v. Birdsong, 360 F. 2d 344 (5th Cir. 1966). 1See, e.g., Griffis v. Luban, 646 N.W. 2d 527 (Minn. 2002); English Sports Betting, Inc. v. Tostigan, 2002 U.S. Dist. Lexis 4985 (E.D. Pa. March 15, 2002). Accord, Cynthia L. Counts & C. Amanda Martin, Libel in Cyberspace: A Framework for Addressing Liability and Jurisdictional Issues in This New Frontier, 59 Alb. L. Rev. 1083, 1124 (1996) 1Calder might be distinguished from many subsequent defamation cases to the extent that the contacts of the writer's employer, the National Enquirer, might be imputable to the writer, just as the downstream contacts of a manufacturer are sometimes imputed to a componenet manufacturer. See Asahi Metal Indus. Co. v. Superior Court, 480 U.S. 102 (1987). 1444 U.S. at 292. 1Martin H. Redish, New Wine, supra n..1 at 607. 1424 U.S. 319 (1976). 1Martin H. Redish, Due Process, Federalism and Personal Jurisdiction: A Theoretical Evaluation, 75 N.W.U.L. Rev. 1112 (1981). See Allan R. Stein, Styles of Argument and Interstate Federalism in the Law of Personal Jurisdiction, 65 Tex. L. Rev. 698 (1987); Allan R. Stein, Burnham and the Death of Theory in The Law of Personal Jurisdiction, 22 Rutgers L. J. 597, 599 n. 13 (1991). Redish, New Wine, supra. At 579. See, e.g., David R. Johnson & David G. Post, The Rise of Law on the Global Network, in BORDERS IN CYBERSPACE 13 (Kahin and Nesson eds., 1997). Accord Lawrence Lessig, Code and Other Laws of Cyberspace 38-39, 49-53 (Basic 1999); Jack Goldsmith.& Alan O. Sykes, The Internet and the Dormant Commerce Clause, 110 Yale L. J. 785, 809-13 (2001).