45 Emory L.J. Number 3(1996) READING THE CONSTITUTION IN CYBERSPACE Lawrence Lessig We might distinguish between two types of constitutional re- gimes, one codifying, the other transformative. A codifying consti- tutional regime aims at preserving something essential from the then- current constitutional or legal culture--to protect it against change in the future; a transformative constitutional regime aims at changing something essential in the then-current constitutional or legal cul ture--to make it different in the future. The picture of the codifying regime is Ulysses tied to the mast; the picture of the transformative is revolutionary In our constitutional tradition the Constitution of 1791 was codifying constitution-the Bill of Rights, that is, was a constitu tional regime that sought to entrench certain practices and values against change. I The Civil War Amendments, on the other hand were transformative, aiming to remake something of what the American social and legal culture had become to tear out from the American soul its tradition of inequality, and replace it with a prac- tice of equalit Professor of Law, University of Chicago School of Law. Thanks to the ever- present Chicago roundtable, as well as Bruce Ackerman, Tracey Meares, and Judge Richard Posner, for helpful advice. Funding for this project was provided by the russell Baker Scholars Fund and the Sarah Scaife Foundation. Thanks also to Ashley Parrish for the exceptional work that is his norm See,e. g, JACK N. RAKOVE, ORIGINAL MEANINGS: POLITICS AND IDEAS IN THE MAKING OF THE CONSTITUTION 302(1996). See also Akhil Reed Amar, he Bill of Rights as a Constitution, 100 YALE L.J. 1131(1991), for another such understanding of the Bill of Rights 2 This is not to deny that there were aspects of the equality in the Civil War Amendments that echoed in our constitutional past. The abolitionists, of made great weight of the Declaration of Independences claims to equality. See, e.g,Trisha Olson, The Natural Law Foundation of the Privileges or Immunities lause of the Fourteenth Amendment, 48 ARK. L. REv. 347, 364(1995). But
45 Emory L.J. Number 3 (1996) READING THE CONSTITUTION IN CYBERSPACE Lawrence Lessig† We might distinguish between two types of constitutional regimes, one codifying, the other transformative. A codifying constitutional regime aims at preserving something essential from the thencurrent constitutional or legal culture—to protect it against change in the future; a transformative constitutional regime aims at changing something essential in the then-current constitutional or legal culture—to make it different in the future. The picture of the codifying regime is Ulysses tied to the mast; the picture of the transformative is revolutionary France. In our constitutional tradition, the Constitution of 1791 was a codifying constitution—the Bill of Rights, that is, was a constitutional regime that sought to entrench certain practices and values against change.1 The Civil War Amendments, on the other hand, were transformative, aiming to remake something of what the American social and legal culture had become, to tear out from the American soul its tradition of inequality, and replace it with a practice of equality.2 † Professor of Law, University of Chicago School of Law. Thanks to the everpresent Chicago roundtable, as well as Bruce Ackerman, Tracey Meares, and Judge Richard Posner, for helpful advice. Funding for this project was provided by the Russell Baker Scholars Fund and the Sarah Scaife Foundation. Thanks also to Ashley Parrish for the exceptional work that is his norm. 1 See, e.g., JACK N. RAKOVE, ORIGINAL MEANINGS: POLITICS AND IDEAS IN THE MAKING OF THE CONSTITUTION 302 (1996). See also Akhil Reed Amar, The Bill of Rights as a Constitution, 100 YALE L.J. 1131 (1991), for another such understanding of the Bill of Rights. 2 This is not to deny that there were aspects of the equality in the Civil War Amendments that echoed in our constitutional past. The abolitionists, of course, made great weight of the Declaration of Independence’s claims to equality. See, e.g., Trisha Olson, The Natural Law Foundation of the Privileges or Immunities Clause of the Fourteenth Amendment, 48 ARK. L. REV. 347, 364 (1995). But
45 EMORY L J. NO. 3(1996) One might well be skeptical of transformative constitutionalism generally. One could be skeptical enough about our own transfor- mative past, let alone of the prospects for some of the more promi- nent transformations in post-communist Europe. In this essay, hoy ever, I will be skeptical of codifying constitutionalism. And I am skeptical here at a conference about cyberspace because cyberspace will make plain just how difficult a practice codifying constitution- alism really is We might think of the problem like this: A codify ing constitution enacts a set of legal constraints on (in our tradition) governmental action.3 But these constraints are just one kind of constraint on gov- ernmental action. The other we might call the constraints of technol- ogy. The warrant requirement is a legal constraint on police action; that the police, unlike Superman, don' t have x-ray vision is a tech nological constraint. We dont think much about technological con- straints when thinking of the constraints of law. We usually just take them for granted. But we should. For what draws into doubt codi fying constitutionalism is just what happens when these constraints of technology change, or more importantly, when we have the power to change them The Fourth Amendment is a ready example. At the time of the founding, the technologies for invading an individual's private space were few. There were spies, eavesdropping and listening through windows or doors, but all these are fairly costly technologies meaning that the ability to live free of review within the confines of one's home was fairly strong. This is not to say that life at the framing was more private-certainly neighbors were more nosy and certainly most of one's life was more public. But within one's home, or one s own papers, the ability of the state to monitor what was going on was quite slight and crude The common law of trespass and the protections of the Fourth Amendment rested upon this fairly crude technology of surveillance ey supplemented these technological constraints with legal con straints. The common law of trespass made it an offense for anyone to cross over into my property; the Fourth Amendment made it an an amendment can be transformative even if it is simply recalling a part of the past, and reestablishing it. This, for example, is what Germany did after World War ll 3 Other constitutional regimes, of course, have been understood to restrain more han governmental action. Germany is a prominent example. See DAVid CURRIE THE CONSTITUTION OF THE FEDERAL REPUBLIC OF GERMANY 182-83 A See, e.g., RICHARD A POSNER, THE ECONOMICS OF JUSTICE 268-71(1981)
45 EMORY L. J. NO. 3 (1996) — 2 — One might well be skeptical of transformative constitutionalism generally. One could be skeptical enough about our own transformative past, let alone of the prospects for some of the more prominent transformations in post-communist Europe. In this essay, however, I will be skeptical of codifying constitutionalism. And I am skeptical here at a conference about cyberspace because cyberspace will make plain just how difficult a practice codifying constitutionalism really is. We might think of the problem like this: A codifying constitution enacts a set of legal constraints on (in our tradition) governmental action.3 But these constraints are just one kind of constraint on governmental action. The other we might call the constraints of technology. The warrant requirement is a legal constraint on police action; that the police, unlike Superman, don’t have x-ray vision is a technological constraint. We don’t think much about technological constraints when thinking of the constraints of law. We usually just take them for granted. But we should. For what draws into doubt codifying constitutionalism is just what happens when these constraints of technology change, or more importantly, when we have the power to change them. The Fourth Amendment is a ready example. At the time of the founding, the technologies for invading an individual’s private space were few. There were spies, eavesdropping and listening through windows or doors; but all these are fairly costly technologies, meaning that the ability to live free of review within the confines of one’s home was fairly strong. This is not to say that life at the framing was more private—certainly neighbors were more nosy, and certainly most of one’s life was more public.4 But within one’s home, or one’s own papers, the ability of the state to monitor what was going on was quite slight and crude. The common law of trespass and the protections of the Fourth Amendment rested upon this fairly crude technology of surveillance. They supplemented these technological constraints with legal constraints. The common law of trespass made it an offense for anyone to cross over into my property; the Fourth Amendment made it an an amendment can be transformative even if it is simply recalling a part of the past, and reestablishing it. This, for example, is what Germany did after World War II. 3 Other constitutional regimes, of course, have been understood to restrain more than governmental action. Germany is a prominent example. See DAVID CURRIE, THE CONSTITUTION OF THE FEDERAL REPUBLIC OF GERMANY 182-83 (1995). 4 See, e.g., RICHARD A. POSNER, THE ECONOMICS OF JUSTICE 268-71 (1981)
READING THE CONSTITUTION IN CYBERSPACE nse for a(federal) officer to escape the proscriptions of the mon law. unless that officer had either a warrant or the search was under the circumstances. reasonable 5 Without a warrant. or without sufficient cause an officer of the state was liable in damage for the trespass. This liability strengthened the protection of privacy beyond protections of the technological constraints These technological and these legal constraints combined to de- fine the constraints that confronted the state as it desired if it de- sired to intrude into a citizen's private domain The sum of these constraints, both legal and technological, might be said to define the domain of security that the individual had against such intrusion Now it is a commonplace that a constitution, at least a codifying constitution, is to preserve these legal constraints against the changes of time. It is a commonplace, that is, that a courts task is to assure that framing values of dignity and liberty are maintained passions for"law and order" notwithstanding. It is a commonplace though, in the context of the Fourth Amendment, that is increasingly ignored. 7 But whether respected in practice or in the breach, at least our ideals are clear-at least we are clear about what we are sup posed to do. about legal values that the Framers constitutionalized we are to be firm: preserving them against the changes in passions that later generations might bring But what should we do about changes in technology? What is a court to do when technologies make it easier for police to monitor what happens inside the home? Or when technologies make it easier for citizens to hide? Here the question is more difficult, and we have at least one clear example of two very different responses we he case is Olmstead v United States, 8 and the question was ther wiretapping was within the se ment. The Court held it was not. when the Constitution was en- acted. said Chief Justice Taft the Fourth amendment was intended to limit trespass on property; that was the common law origin of the 5 TELFORD TAYLOR. TWO STUDIES IN CONSTITUTIONAL INTERPRETATION 21 50(1969): Amar, supra note 1, at 1178-80 6 Minnesota v. Dickerson, 508 U.S. 366, 380-83(1993)(Scalia,J,concur 7 See, e. g, Tracey Maclin, When the Cure for the Fourth Amendment Is Worse than the disease, 68 S CAL L. REV. 1(1994) 8 277 U.S. 438(1928). See also JAMES BOYD WHITE, JUSTICE AS TRANS LATION149-57(1989)
READING THE CONSTITUTION IN CYBERSPACE — 3 — offense for a (federal) officer to escape the proscriptions of the common law, unless that officer had either a warrant, or the search was, under the circumstances, reasonable.5 Without a warrant, or without sufficient cause, an officer of the state was liable in damages for the trespass. This liability strengthened the protection of privacy beyond protections of the technological constraints. These technological and these legal constraints combined to define the constraints that confronted the state as it desired, if it desired, to intrude into a citizen’s private domain. The sum of these constraints, both legal and technological, might be said to define the domain of security that the individual had against such intrusion. Now it is a commonplace that a constitution, at least a codifying constitution, is to preserve these legal constraints against the changes of time. It is a commonplace, that is, that a court’s task is to assure that framing values of dignity6 and liberty are maintained, passions for “law and order” notwithstanding. It is a commonplace, though, in the context of the Fourth Amendment, that is increasingly ignored.7 But whether respected in practice or in the breach, at least our ideals are clear—at least we are clear about what we are supposed to do. About legal values that the Framers constitutionalized, we are to be firm: preserving them against the changes in passions that later generations might bring. But what should we do about changes in technology? What is a court to do when technologies make it easier for police to monitor what happens inside the home? Or when technologies make it easier for citizens to hide? Here the question is more difficult, and we have at least one clear example of two very different responses. The case is Olmstead v. United States, 8 and the question was whether wiretapping was within the scope of the Fourth Amendment. The Court held it was not. When the Constitution was enacted, said Chief Justice Taft, the Fourth Amendment was intended to limit trespass on property; that was the common law origin of the 5 TELFORD TAYLOR, TWO STUDIES IN CONSTITUTIONAL INTERPRETATION 21- 50 (1969); Amar, supra note 1, at 1178-80. 6 Minnesota v. Dickerson, 508 U.S. 366, 380-83 (1993) (Scalia, J., concurring). 7 See, e.g., Tracey Maclin, When the Cure for the Fourth Amendment Is Worse than the Disease, 68 S. CAL. L. REV. 1 (1994). 8 277 U.S. 438 (1928). See also JAMES BOYD WHITE, JUSTICE AS TRANSLATION 149-57 (1989)
45 EMORY L J. NO. 3(1996) Amendment. Wiretapping a person's phone is not a trespass; there fore, concluded Taft, wiretapping did not invade the Fourth Amendment's interests Justice Brandeis saw the case differently. Of course the Fourth Amendment originally protected against trespass, but this was be- cause trespass was the only effective way that the state could invade privacy interests. Sure, it could eavesdrop without trespassing, so it could in some sense intrude without constitutional violation: but eavesdropping was of little importance at the founding since police were nonexistent, and eavesdropping quite public. And in any case eavesdropping is not as significant an invasion as the invasion that would be permitted if the government could tap phones without limit. For even in 1928. much of life had moved onto the wires and in those first steps into cyberspace, Brandeis argued, the Constitu tion should not leave citizens exposed o What had changed, he ar gued, was a technology of surveillance and a technology of commu nication II Life existed now in cyberspace, and the Constitution should be read to protect the same interests of privacy in cyberspace that the Framers had protected in real space. Technology had changed. but Brandeis argued that change should not be allowed to change the meaning of the Constitution. 2 The Constitution should protect now what it protected then If there is a justice who deserves c-world's praise, if there is an opinion of the Supreme Court that should be the model for cy be activists, if there is a first chapter in the fight to protect cyberspace it is this Justice, this opinion and this case. Here, in as clear an ex ample as any, is a method that will be central to cyberspaces sur vival as a place where values of individual liberty are sustained. The 9 As Brandeis wrote, "When the Fourth and Fifth Amendments were adopted, the form that evil had theretofore taken had been necessarily simple. 277 U.S.at 473 10 The an iae from the telephone companies in the Olmstead case pre ented quite effectively the place the telephone had taken in the ordinary life of most citizens. See Brief in Support of Petitioners'Contention, on behalf of Pa cific Telephone and Telegraphy Co., et al., Olmstead, 277 U.S. 438 1l277Us.at473 12 Brandeis's fears were well stated: "Ways may some day be developed by which the Government, without removing papers from secret drawers, can repro- duce them in court, and by which it will be enabled to expose to a jury the most intimate occurrences of the home. ld at 474
45 EMORY L. J. NO. 3 (1996) — 4 — Amendment. Wiretapping a person’s phone is not a trespass; therefore, concluded Taft, wiretapping did not invade the Fourth Amendment’s interests. Justice Brandeis saw the case differently. Of course the Fourth Amendment originally protected against trespass, but this was because trespass was the only effective way that the state could invade privacy interests.9 Sure, it could eavesdrop without trespassing, so it could in some sense intrude without constitutional violation; but eavesdropping was of little importance at the founding since police were nonexistent, and eavesdropping quite public. And in any case, eavesdropping is not as significant an invasion as the invasion that would be permitted if the government could tap phones without limit. For even in 1928, much of life had moved onto the wires; and in those first steps into cyberspace, Brandeis argued, the Constitution should not leave citizens exposed.10 What had changed, he argued, was a technology of surveillance and a technology of communication.11 Life existed now in cyberspace, and the Constitution should be read to protect the same interests of privacy in cyberspace that the Framers had protected in real space. Technology had changed, but, Brandeis argued, that change should not be allowed to change the meaning of the Constitution.12 The Constitution should protect now what it protected then. If there is a Justice who deserves c-world’s praise, if there is an opinion of the Supreme Court that should be the model for cyberactivists, if there is a first chapter in the fight to protect cyberspace, it is this Justice, this opinion and this case. Here, in as clear an example as any, is a method that will be central to cyberspace’s survival as a place where values of individual liberty are sustained. The 9 As Brandeis wrote, “When the Fourth and Fifth Amendments were adopted, the form that evil had theretofore taken had been necessarily simple.” 277 U.S. at 473. 10 The amici curiae from the telephone companies in the Olmstead case presented quite effectively the place the telephone had taken in the ordinary life of most citizens. See Brief in Support of Petitioners’ Contention, on behalf of Pacific Telephone and Telegraphy Co., et al., Olmstead, 277 U.S. 438. 11 277 U.S. at 473. 12 Brandeis’s fears were well stated: “Ways may some day be developed by which the Government, without removing papers from secret drawers, can reproduce them in court, and by which it will be enabled to expose to a jury the most intimate occurrences of the home.” Id. at 474
READING THE CONSTITUTION IN CYBERSPACE method is translation 13 Brandeis first identifies values from the original fourth amendment, and then translates these values into the context of cyberspace. He read beyond the specific applications that the Framers had in mind, to find the meaning they intended to con- stitutionalize. He found a way to read the Constitution in 1928 to preserve the meaning it had in 1791. Brandeis's opinion shows how to carry the Framers values into our interpretive context, in a way that has an extremely strong claim to constitutional fidelity. The ar gument was as simple as it was compelling: By doing what the Framers would have done, Brandeis argued, the Court would change the Constitution's meaning(since now a large part of inti mate life was open to governmental intrusion without the protections of the Fourth Amendment and by doing something other than what the Framers would have done(by protecting something more than trespass), the Court could preserve the Constitutions meaning In the almost seventy years since Brandeis's opinion, we ha vent any better example of the translators craft. Unlike originalist who believe that fidelity requires doing just what the Framers would have done, 4 the translator understands that to preserve meaning across contexts. one must change readings across context 15 This is what the linguistic translator does: In Germany, she says" danke In France, she says"merci. In the two places she has said different things, but in both places, she has meant the very same thing. Con- texts are different, so meaning is preserved by saying something different The most cyberspace could hope for, from judges at least, is the practice that Brandeis offered the most we could aspire to, in inter- preting our Constitution, is the sensitivity that he displayed. But in this article, I want to point to the limits of even this method of inter- pretive fidelity. Translation may well be fidelity's best method; but 13 Paul Brest, The Misconceived Quest for the Original Understanding, 60 B.U. L. REV. 204, 218(1980). For a recent application, see William Michael Treanor, The Original Understanding of the Takings Clause and the politico Process, 95 COLUM. L. REV. 782 (1995). See generally Lawrence Lessig, Fidel- ity in Tr 1165(1993); Larry Alexander, All or Noth ing at All? The Intentions of Authorities and the Authority of Intentions,in LAW AND INTERPRETATION: ESSAYS IN LEGAL PHILOSOPHY 371-75(1995); and RICHARD A POSNER, OVERCOMING LAW 494-97(1995)for useful criticism of the method 14 This we might call"one-step"originalism. See Lessig, supra note 13,at 15 And this we might call"two-step"originalism. Id
READING THE CONSTITUTION IN CYBERSPACE — 5 — method is translation: 13 Brandeis first identifies values from the original Fourth Amendment, and then translates these values into the context of cyberspace. He read beyond the specific applications that the Framers had in mind, to find the meaning they intended to constitutionalize. He found a way to read the Constitution in 1928 to preserve the meaning it had in 1791. Brandeis’s opinion shows how to carry the Framers’ values into our interpretive context, in a way that has an extremely strong claim to constitutional fidelity. The argument was as simple as it was compelling: By doing what the Framers would have done, Brandeis argued, the Court would change the Constitution’s meaning (since now a large part of intimate life was open to governmental intrusion without the protections of the Fourth Amendment); and by doing something other than what the Framers would have done (by protecting something more than trespass), the Court could preserve the Constitution’s meaning. In the almost seventy years since Brandeis’s opinion, we haven’t any better example of the translator’s craft. Unlike originalists who believe that fidelity requires doing just what the Framers would have done,14 the translator understands that to preserve meaning across contexts, one must change readings across context.15 This is what the linguistic translator does: In Germany, she says “danke.” In France, she says “merci.” In the two places she has said different things, but in both places, she has meant the very same thing. Contexts are different, so meaning is preserved by saying something different. The most cyberspace could hope for, from judges at least, is the practice that Brandeis offered; the most we could aspire to, in interpreting our Constitution, is the sensitivity that he displayed. But in this Article, I want to point to the limits of even this method of interpretive fidelity. Translation may well be fidelity’s best method; but 13 Paul Brest, The Misconceived Quest for the Original Understanding, 60 B.U. L. REV. 204, 218 (1980). For a recent application, see William Michael Treanor, The Original Understanding of the Takings Clause and the Political Process, 95 COLUM. L. REV. 782 (1995). See generally Lawrence Lessig, Fidelity in Translation, 71 TEX. L. REV. 1165 (1993); Larry Alexander, All or Nothing at All? The Intentions of Authorities and the Authority of Intentions, in LAW AND INTERPRETATION: ESSAYS IN LEGAL PHILOSOPHY 371-75 (1995); and RICHARD A. POSNER, OVERCOMING LAW 494-97 (1995) for useful criticism of the method. 14 This we might call “one-step” originalism. See Lessig, supra note 13, at 1183-85. 15 And this we might call “two-step” originalism. Id
45 EMORY L J. NO. 3(1996) argument here is about how difficult fidelity in cyberspace is ng to be The point is this: We, as a legal culture, want the Constitution to resolve the questions of rights, we don't want them to be in the cen ter of our interpretive struggles. We therefore seek, through various forms of originalism, ways to link our constitutional practice today to the practice of the Framers. We seek the authority of originalism as a trump in present legal disputes, and the best way to locate that authority is Brandeis's-translation But even the best translations at some point give out. Even the most careful translators must at some stage concede there is not enough left from the framing regime to guide any more 16 Transla tion will have its limits, and these limits will be of great and I fear terrible significance for us today. For what these limits will yield is a relatively passive judiciary, and a relatively deferential attitude to- ward governmental intrusion. My sense is that, knowing nothing, or at least not very much, terrified by the threats of that which they dont know, these judges will defer to those with democratic authority; without clear rules to limit the democrats, the juricrats will step aside One can hardly blame them for this. Indeed in some cases this deference should be encouraged. 17 But i want to move that counsel of prudence along a bit--to place it in context and limit its reach. We should understand just why the judge's position will be so difficult and we should isolate the source of the difficulty. In some places this difficulty should counsel deference a certain hesitation before resolving the questions of the Constitution in cyberspace finally, or firmly, or with any pretense to permanence. But in others, I argue that judges-especially lower court judges-should be stronger. In cases of simple translation, judges should advance quite firmly ar guments that seek to preserve original values of liberty in a very d ferent context. In these places, there is a space for activism. But in cases where translation is not so simple, judges should kvetch. In these places, they should talk about the questions these changes 16 Felix Cohen, Field Theory and Judicial Logic, 59 YALE LJ.239,272 ( 1950)("Only in mathematics do we find perfect translations--the sort of thing chat enables us to translate any proposition about a straight line in Euclidean geometry into an equivalent proposition about a curve in riemannian geometry But outside mathematics, though we live in a world of imperfections, some im perfections are worse than others. 17 One could well have argued that during the crisis of the Depression, defer ence by the Court to the Congress was well advised. See, e.g., CASS R SUNSTEIN, DEMOCRACY AND THE PROBLEM OF FREE SPEECH 39(1993)
45 EMORY L. J. NO. 3 (1996) — 6 — my argument here is about how difficult fidelity in cyberspace is going to be. The point is this: We, as a legal culture, want the Constitution to resolve the questions of rights; we don’t want them to be in the center of our interpretive struggles. We therefore seek, through various forms of originalism, ways to link our constitutional practice today to the practice of the Framers. We seek the authority of originalism, as a trump in present legal disputes, and the best way to locate that authority is Brandeis’s—translation. But even the best translations at some point give out. Even the most careful translators must at some stage concede there is not enough left from the framing regime to guide any more.16 Translation will have its limits, and these limits will be of great and, I fear, terrible significance for us today. For what these limits will yield is a relatively passive judiciary, and a relatively deferential attitude toward governmental intrusion. My sense is that, knowing nothing, or at least not very much, terrified by the threats of that which they don’t know, these judges will defer to those with democratic authority; without clear rules to limit the democrats, the juricrats will step aside. One can hardly blame them for this. Indeed, in some cases this deference should be encouraged.17 But I want to move that counsel of prudence along a bit—to place it in context and limit its reach. We should understand just why the judge’s position will be so difficult and we should isolate the source of the difficulty. In some places, this difficulty should counsel deference, a certain hesitation before resolving the questions of the Constitution in cyberspace finally, or firmly, or with any pretense to permanence. But in others, I argue that judges—especially lower court judges—should be stronger. In cases of simple translation, judges should advance quite firmly arguments that seek to preserve original values of liberty in a very different context. In these places, there is a space for activism. But in cases where translation is not so simple, judges should kvetch. In these places, they should talk about the questions these changes 16 Felix Cohen, Field Theory and Judicial Logic, 59 YALE L.J. 239, 272 (1950) (“Only in mathematics do we find perfect translations—the sort of thing that enables us to translate any proposition about a straight line in Euclidean geometry into an equivalent proposition about a curve in Riemannian geometry. But outside mathematics, though we live in a world of imperfections, some imperfections are worse than others.”). 17 One could well have argued that during the crisis of the Depression, deference by the Court to the Congress was well advised. See, e.g., CASS R. SUNSTEIN, DEMOCRACY AND THE PROBLEM OF FREE SPEECH 39 (1993)
READING THE CONSTITUTION IN CYBERSPACE raise. And if the result is deferential or passive, it should be so in protest. Here there may well be a place for prudence in deed, but to earn this right to be passive, to compensate for allowing rights claims to fail, judges should not be deferential in word. They should raise before the legal culture the conflict that these new cases pres- ent. Hard cases need not make bad law, but neither should they be treated as if they were easy This Article moves in two stages. I begin with some examples of the changes that cyberspace will present, and some of the questions of constitutionalism that these changes will invite. Following bran- deis. in each case i hazard a translation that reflects what i believe is the best guess at the most we could expect a court in such a case to do. That is Section I In Section Il, I will isolate what about these changes makes them so significant, and what about them makes the change of cyberspace itself so significant, at least to the practice of constitutional fidelity In a line that will no doubt seem far too abstract, we might describe the problem of cyberspace for constitutional law like this: That it leaves us without constraint enough; that we are vis-a-vis the laws of nature in this new space, gods; and that the problem with bein gods is that we must choose. These choices will be choices of great moment; they will raise contested values; they will be of great con titutional significance but they will be made by an institution that is, as it were, allergic to such choice. They will be made, by a Court, pretending that in making its decisions, it is following the choice of others-of the people, of"we the people, "who in truth have not yet confronted the constitutional choices that must be made L THREE PROBLEMS FROM CYBERSPACE I begin with three puzzles presented by cyberspace, and some- thing of the constitutional issues that they raise. I then attempt to solve these puzzles using the technique of translation. My treatment of each, of course, will be far less extensive than a complete account would have to be. But my point in presenting the three is not finally to answer them, but rather to draw out from the collection something important and general about cyberspace Anonymity
READING THE CONSTITUTION IN CYBERSPACE — 7 — raise. And if the result is deferential or passive, it should be so in protest. Here there may well be a place for prudence in deed, but to earn this right to be passive, to compensate for allowing rights claims to fail, judges should not be deferential in word. They should raise before the legal culture the conflict that these new cases present. Hard cases need not make bad law; but neither should they be treated as if they were easy. This Article moves in two stages. I begin with some examples of the changes that cyberspace will present, and some of the questions of constitutionalism that these changes will invite. Following Brandeis, in each case I hazard a translation that reflects what I believe is the best guess at the most we could expect a court in such a case to do. That is Section I. In Section II, I will isolate what about these changes makes them so significant, and what about them makes the change of cyberspace itself so significant, at least to the practice of constitutional fidelity. In a line that will no doubt seem far too abstract, we might describe the problem of cyberspace for constitutional law like this: That it leaves us without constraint enough; that we are, vis-à-vis the laws of nature in this new space, gods; and that the problem with being gods is that we must choose. These choices will be choices of great moment; they will raise contested values; they will be of great constitutional significance; but they will be made by an institution that is, as it were, allergic to such choice. They will be made, by a Court, pretending that in making its decisions, it is following the choice of others—of the people, of “we the people,” who in truth have not yet confronted the constitutional choices that must be made. I. THREE PROBLEMS FROM CYBERSPACE I begin with three puzzles presented by cyberspace, and something of the constitutional issues that they raise. I then attempt to solve these puzzles using the technique of translation. My treatment of each, of course, will be far less extensive than a complete account would have to be. But my point in presenting the three is not finally to answer them, but rather to draw out from the collection something important and general about cyberspace. Anonymity
45 EMORY L J. NO. 3(1996) The architecture of cyberspace--as it is just now--is open 18 One enters cyberspace as one wants. One can enter identifying who one is, or one can hide who one is One can enter speaking a lan guage that anyone can understand, or one can encrypt the language one speaks, so only the intended listeners can understand what one says. 19 What others see of you is within your control; whether oth ers understand of you is within your control as well We can call this general power of control a power of privacy 20 It is the power to determine what others will know about you; the power to determine whether they will kn now your name or who you are, the power to determine whether they will know what you say, or even what language you speak. Anonymity refers to the power to control whether people know who you are; it is a tool of privacy Encryption is the power to control whether people know what lan guage you speak; it too is a tool of privacy real life. If one wants to hide who one is, one can don a mask To some degree, technologies of privacy exist everywhere use a fake name. One can speak in code, or in a language that very few understand. Not many people do, of course. The streets are not filled with masked men, and it is actually quite difficult to use a fake name. But nonetheless, since espionage began, there have been techniques for appearing as other than who one is, and people have used these techniques to construct a certain personal anonymity But anonymity in cyberspace is not just different in degree from anonymity in real space. As cyberspace presently is, it gives an in dividual a kind of power that doesn't exist in real space. This is not just the ability to put on a mask; it is the ability to hide absolutely 18 I am following Monroe Price in this distinction. See Monroe Price,Free Expression and Digital Dreams, 22 CRITICAL INQ. 64, 67(1995). It is this ar hitecture of openness that makes possible the world Tim May describes as crypto-anarchy See Timothy C May, Crypto-Anarchy and Virtual Communi ties, Posted to Cypherpunks List, December 1994 19 On the technologies of encryption, see A. Michael Froomkin, The Meta phor Is the Key: Cryptograph, the Clipper Chip, and the Constitution, 143 U PA L REv. 709(1995); U.S. Cong Off. Techn. Assessment, Information Se- curity and Privacy in Network Environments, 111-32(1994). On the constitu- tionality of government control of encryption, see Froomkin, supra 20 See ANNE WELLS BRANSCOMB, WHO OWNS INFORMATION? 44-45(1994) OSCAR GANDY, THE PANOPTIC SORT 190(1993); M. ETHAN KATSH, LAW IN A DIGITAL WORLD 228(1995); M. ETHAN KATSH, THE ELECTRONIC MEDIA AND THE TRANSFORMATION OF LAW 189-97(1989). See also JEFFREY ROTHFEDER, PRIVACY FOR SALE(1992): ALAN WESTIN, PRIVACY AND FREEDOM(1967)
45 EMORY L. J. NO. 3 (1996) — 8 — The architecture of cyberspace—as it is just now—is open.18 One enters cyberspace as one wants. One can enter identifying who one is, or one can hide who one is. One can enter speaking a language that anyone can understand, or one can encrypt the language one speaks, so only the intended listeners can understand what one says.19 What others see of you is within your control; whether others understand of you is within your control as well. We can call this general power of control a power of privacy.20 It is the power to determine what others will know about you; the power to determine whether they will know your name, or who you are; the power to determine whether they will know what you say, or even what language you speak. Anonymity refers to the power to control whether people know who you are; it is a tool of privacy. Encryption is the power to control whether people know what language you speak; it too is a tool of privacy. To some degree, technologies of privacy exist everywhere in real life. If one wants to hide who one is, one can don a mask, or use a fake name. One can speak in code, or in a language that very few understand. Not many people do, of course. The streets are not filled with masked men, and it is actually quite difficult to use a fake name. But nonetheless, since espionage began, there have been techniques for appearing as other than who one is, and people have used these techniques to construct a certain personal anonymity. But anonymity in cyberspace is not just different in degree from anonymity in real space. As cyberspace presently is, it gives an individual a kind of power that doesn’t exist in real space. This is not just the ability to put on a mask; it is the ability to hide absolutely 18 I am following Monroe Price in this distinction. See Monroe Price, Free Expression and Digital Dreams, 22 CRITICAL INQ. 64, 67 (1995). It is this architecture of openness that makes possible the world Tim May describes as crypto-anarchy. See Timothy C. May, Crypto-Anarchy and Virtual Communities, Posted to Cypherpunks List, December 1994. 19 On the technologies of encryption, see A. Michael Froomkin, The Metaphor Is the Key: Cryptograph, the Clipper Chip, and the Constitution, 143 U. PA. L. REV. 709 (1995); U.S. Cong. Off. Techn. Assessment, Information Security and Privacy in Network Environments, 111-32 (1994). On the constitutionality of government control of encryption, see Froomkin, supra. 20 See ANNE WELLS BRANSCOMB, WHO OWNS INFORMATION? 44-45 (1994); OSCAR GANDY, THE PANOPTIC SORT 190 (1993); M. ETHAN KATSH, LAW IN A DIGITAL WORLD 228 (1995); M. ETHAN KATSH, THE ELECTRONIC MEDIA AND THE TRANSFORMATION OF LAW 189-97 (1989). See also JEFFREY ROTHFEDER, PRIVACY FOR SALE (1992); ALAN WESTIN, PRIVACY AND FREEDOM (1967)
READING THE CONSTITUTION IN CYBERSPACE who one is. It is not just the ability to speak a different, or encoded, language; it is the ability to speak a language that is(practically)im- possible to crack. 21 Cyberspace is a place that maximizes both social and individual plasticity, which means it is a place that determines very little about what others must know about you This power of privacy is not inherently bad. 22 Indeed, one way to understand its salience is as a reaction to just the opposite trend he powers of surveillance have increased(through sophisticated transaction tracking devices that can collect profiles of individual behavior and predict future behavior on the basis of these profiles) one might well understand the anonymity as a effort at reclaiming private space lost. As the number of eyes that watch increase, ano- nymity becomes a more effective technology to block the vision of any one set. So understood, these tools of privacy are accommoda tions to the increased power of surveillance that the changing tech nologies of surveillance have created 23 The most important place for this technology will be com merce24 There was a time when we relied upon another tool of anonymity to a very large degree: cash. Cash is an al most perfect tool of anonym Nothing in its nature reveals anything ab person using it; it is self-authenticating, depending upon individual credentials not at all. One could use cash. and unless identification were independently given, the identity of the one who used the casl would be lost to the system. Cash flows without a trace 21 There is in fact an ongoing question about whether these algorithms can be cracked. See Froomkin, supra note 19, at 735-42. Levy describes an example where the Internet facilitated a conspiracy to crack a code, with great success. See Steven Levy, Wisecrackers, WIRED, Mar. 1996, at 128. Where this struggle end 3 See sTEvEN L NocK. THE COSTS OF PRIVACY and extremely helpful discussion of privacy in cyber see anne meredith Fulton, Cyberspace and the Internet: Who will Privacy Police? COMMLAW CONSPECTUS 63(1995) with Anonymity, Digital Cash, and Distributed databases, 15 J. L& COM. 39
READING THE CONSTITUTION IN CYBERSPACE — 9 — who one is. It is not just the ability to speak a different, or encoded, language; it is the ability to speak a language that is (practically) impossible to crack.21 Cyberspace is a place that maximizes both social and individual plasticity, which means it is a place that determines very little about what others must know about you. This power of privacy is not inherently bad.22 Indeed, one way to understand its salience is as a reaction to just the opposite trend. As the powers of surveillance have increased (through sophisticated transaction tracking devices that can collect profiles of individual behavior and predict future behavior on the basis of these profiles), one might well understand the anonymity as a effort at reclaiming private space lost. As the number of eyes that watch increase, anonymity becomes a more effective technology to block the vision of any one set. So understood, these tools of privacy are accommodations to the increased power of surveillance that the changing technologies of surveillance have created.23 The most important place for this technology will be commerce.24 There was a time when we relied upon another tool of anonymity to a very large degree: cash. Cash is an almost perfect tool of anonymity. Nothing in its nature reveals anything about the person using it; it is self-authenticating, depending upon individual credentials not at all. One could use cash, and unless identification were independently given, the identity of the one who used the cash would be lost to the system. Cash flows without a trace. 21 There is in fact an ongoing question about whether these algorithms can be cracked. See Froomkin, supra note 19, at 735-42. Levy describes an example where the Internet facilitated a conspiracy to crack a code, with great success. See Steven Levy, Wisecrackers, WIRED, Mar. 1996, at 128. Where this struggle ends is yet undetermined. 22 But see RICHARD A. POSNER, OVERCOMING LAW ch. 25 (1995). 23 See STEVEN L. NOCK, THE COSTS OF PRIVACY 1-14 (1993). For a general and extremely helpful discussion of privacy in cyberspace, see Anne Meredith Fulton, Cyberspace and the Internet: Who Will Be the Privacy Police?, 3 COMMLAW CONSPECTUS 63 (1995). 24 See A. Michael Froomkin, Flood Control on the Information Ocean: Living with Anonymity, Digital Cash, and Distributed Databases, 15 J.L. & COM. 395 (1996)
45 EMORY L J. NO. 3(1996) Digital cash25 is the next great cash station. 26 Digital cash, like regular cash, can be traded without traceability 27 Giving digital cash to another creates a credit in their account without any traceable liability to the giver. More importantly, such transfers can occur without the most common burden of cash getting in the way-the risk of loss. One can move $1,000,000 in digital cash with just the touch of a button; one can move $1, 000, 000 in regular cash only with a wheelbarrow But when one begins to think of digital cash, and then encrypted conversation, and when one puts the two together, one begins to see the fear this very same power of anonymity creates 28 For anonym- ity becomes a technology for the perfect crime. 29 And the question then becomes whether there is an ability-let' s start and stop with constitutional permissibility--to regulate this anonymity What kind of regulation would this be? It would certainly be too broad to ban all encrypted conversation. 30 Indeed, there may well be a constitutional right to encryption, as a constitutional accommoda tion to the paucity of constitutional protection for personal informa 5 Private money has a long history in America; see VIVIANA A. ZELIZER, THE SOCIAL MEANING OF MONEY (1994). One might wonder to what extent the government will allow competing currency to return 6 See, e. g, KEVIN KELLY, OUT OF CONTROL 203-30(1994) cally necessary. Indeed, my central point is that there is very little that is necessary and too much that is possible 28 For an excellent discussion of legal dilemmas related to these questions, see Michael Rustad Lori E. Eisenschmidt, The Commercial Law of Internet Secu rity, 10 HIGH TECH. L.J. 213(1995) 29 See Froomkin, supra note 19, at 727; BUILDING IN BIG BROTHER: THE CRYPTOGRAPHIC POLICY DEBATE 323-91 (Lance J. Hoffman ed, 1994). For a more immediate concern about the negative consequences of anonymity,see Mclntyre v Ohio Elections Comm'n, 115 S Ct 1511, 1537(1995)(Scalia, J 30 This has not stopped some states from trying. See 18 PA. CONS. STAT 8910(Supp. 1996)(Pennsylvania statute making it a crime to possess a pro- gram or device which can be used to" conceal or to assist another to conceal the or place of origin or of destination of any tele Ilso Lawrence Lessig, The Path of Cyberlaw, 104 YALE L J. 1743, 1750 n 20 95)(Connecticut bill)
45 EMORY L. J. NO. 3 (1996) — 10 — Digital cash25 is the next great cash station.26 Digital cash, like regular cash, can be traded without traceability.27 Giving digital cash to another creates a credit in their account without any traceable liability to the giver. More importantly, such transfers can occur without the most common burden of cash getting in the way—the risk of loss. One can move $1,000,000 in digital cash with just the touch of a button; one can move $1,000,000 in regular cash only with a wheelbarrow. But when one begins to think of digital cash, and then encrypted conversation, and when one puts the two together, one begins to see the fear this very same power of anonymity creates.28 For anonymity becomes a technology for the perfect crime.29 And the question then becomes whether there is an ability—let’s start and stop with constitutional permissibility—to regulate this anonymity. What kind of regulation would this be? It would certainly be too broad to ban all encrypted conversation.30 Indeed, there may well be a constitutional right to encryption, as a constitutional accommodation to the paucity of constitutional protection for personal informa- 25 Private money has a long history in America; see VIVIANA A. ZELIZER, THE SOCIAL MEANING OF MONEY (1994). One might wonder to what extent the government will allow competing currency to return. 26 See, e.g., KEVIN KELLY, OUT OF CONTROL 203-30 (1994). 27 I am talking about what is technologically possible, not what is technologically necessary. Indeed, my central point is that there is very little that is “necessary” and too much that is possible. 28 For an excellent discussion of legal dilemmas related to these questions, see Michael Rustad & Lori E. Eisenschmidt, The Commercial Law of Internet Security, 10 HIGH TECH. L.J. 213 (1995). 29 See Froomkin, supra note 19, at 727; BUILDING IN BIG BROTHER: THE CRYPTOGRAPHIC POLICY DEBATE 323-91 (Lance J. Hoffman ed., 1994). For a more immediate concern about the negative consequences of anonymity, see McIntyre v. Ohio Elections Comm’n, 115 S. Ct. 1511, 1537 (1995) (Scalia, J., dissenting). 30 This has not stopped some states from trying. See 18 PA. CONS. STAT. § 910 (Supp. 1996) (Pennsylvania statute making it a crime to possess a program or device which can be used to “conceal or to assist another to conceal the existence or place of origin or of destination of any telecommunication”). See also Lawrence Lessig, The Path of Cyberlaw, 104 YALE L.J. 1743, 1750 n.20 (1995) (Connecticut bill)