"The Constitution in Cyberspace: Law and Liberty Beyond the Electronic Frontier"
by Laurence H. Tribe
Copyright 1991 Laurence H. Tribe, Tyler Professor of Constitutional Law,
Harvard Law School.
Professor Tribe is the author, most recently, of "On Reading the Constitution"
(Harvard University Press, Cambridge, MA, 1991).Introduction My topic is how to "map" the text and structure of our Constitution onto the texture and topology of "cyberspace". That's the term coined by cyberpunk novelist William Gibson, which many now use to describe the "place" -- a place without physical walls or even physical dimensions -- where ordinary telephone conversations "happen," where voice-mail and e-mail messages are stored and sent back and forth, and where computer-generated graphics are transmitted and transformed, all in the form of interactions, some real-time and some delayed, among countless users, and between users and the computer itself Some use the "cyberspace" concept to designate fantasy worlds or "virtual realities" of the sort Gibson described in his novel *Neuromancer*, in which people can essentially turn their minds into computer peripherals capable of perceiving and exploring the data matrix. The whole idea of "virtual reality," of course, strikes a slightly odd note. As one of Lily Tomlin's most memorable characters once asked, "What's reality, anyway, but a collective hunch?" Work in this field tends to be done largely by people who share the famous observation that reality is overrated! However that may be, "cyberspace" connotes to some users the sorts of technologies that people in Silicon Valley (like Jaron Lanier at VPL Research, for instance) work on when they try to develop "virtual racquetball" for the disabled, computer-aided design systems that allow architects to walk through "virtual buildings" and remodel them *before* they are built, "virtual conferencing" for business meetings, or maybe someday even "virtual day care centers" for latchkey children. The user snaps on a pair of goggles hooked up to a high-powered computer terminal, puts on a special set of gloves (and perhaps other gear) wired into the same computer system, and, looking a little bit like Darth Vader, pretty much steps into a computer-driven, drug-free, 3-dimensional, interactive, infinitely expandable hallucination complete with sight, sound and touch -- allowing the user literally to move through, and experience, information. I'm using the term "cyberspace" much more broadly, as many have lately. I'm using it to encompass the full array of computer-mediated audio and/or video interactions that are already widely dispersed in modern societies -- from things as ubiquitous as the ordinary telephone, to things that are still coming on-line like computer bulletin boards and networks like Prodigy, or like the WELL ("Whole Earth 'Lectronic Link"), based here in San Francisco. My topic, broadly put, is the implications of that rapidly expanding array for our constitutional order. It is a cyberspace, either get bent out of shape or fade out altogether. The question, then, becomes: when the lines along which our Constitution is drawn warp or vanish, what happens to the Constitution itself? Setting the Stage To set the stage with a perhaps unfamiliar example, consider a decision handed down nine months ago, *Maryland v. Craig*, where the U.S. Supreme Court upheld the power of a state to put an alleged child abuser on trial with the defendant's accuser testifying not in the defendant's presence but by one-way, closed-circuit television. The Sixth Amendment, which of course antedated television by a century and a half, says: "In all criminal prosecutions, the accused shall enjoy the right . . . to be confronted with the witnesses against him." Justice O'Connor wrote for a bare majority of five Justices that the state's procedures nonetheless struck a fair balance between costs to the accused and benefits to the victim and to society as a whole. Justice Scalia, joined by the three "liberals" then on the Court (Justices Brennan, Marshall and Stevens), dissented from that cost-benefit approach to interpreting the Sixth Amendment. He wrote: The Court has convincingly proved that the Maryland procedure serves a valid interest, and gives the defendant virtually everything the Confrontation Clause guarantees (everything, that is, except confrontation). I am persuaded, therefore, that the Maryland procedure is virtually constitutional. Since it is not, however, actually constitutional I [dissent]. Could it be that the high-tech, closed-circuit TV context, almost as familiar to the Court's youngest Justice as to his even younger law clerks, might've had some bearing on Justice Scalia's sly invocation of "virtual" constitutional reality? Even if Justice Scalia wasn't making a pun on "virtual reality," and I suspect he wasn't, his dissenting opinion about the Confrontation Clause requires *us* to "confront" the recurring puzzle of how constitutional provisions written two centuries ago should be construed and applied in ever-changing circumstances. Should contemporary society's technology-driven cost-benefit fixation be allowed to water down the old-fashioned value of direct confrontation that the Constitution seemingly enshrined as basic? I would hope not. In that respect, I find myself in complete agreement with Justice Scalia. But new technological possibilities for seeing your accuser clearly without having your accuser see you at all -- possibilities for sparing the accuser any discomfort in ways that the accuser couldn't be spared before one-way mirrors or closed-circuit TVs were developed -- *should* lead us at least to ask ourselves whether *two*-way confrontation, in which your accuser is supposed to be made uncomfortable, and thus less likely to lie, really *is* the core value of the Confrontation Clause. If so, "virtual" confrontation should be held constitutionally insufficient. If not -- if the core value served by the Confrontation Clause is just the ability to *watch* your accuser say that you did it -- then "virtual" confrontation should suffice. New technologies should lead us to look more closely at just *what values* the Constitution seeks to preserve. New technologies should *not* lead us to react reflexively *either way* -- either by assuming that technologies the Framers didn't know about make their concerns and values obsolete, or by assuming that those new technologies couldn't possibly provide new ways out of old dilemmas and therefore should be ignored altogether. The one-way mirror yields a fitting metaphor for the task we confront. As the Supreme Court said in a different context several years ago, "The mirror image presented [here] requires us to step through an analytical looking glass to resolve it." (*NCAA v. Tarkanian*, 109 S. Ct. at 462.) The world in which the Sixth Amendment's Confrontation Clause was written and ratified was a world in which "being confronted with" your accuser *necessarily* meant a simultaneous physical confrontation so that your accuser had to *perceive* you being accused by him. Closed-circuit television and one-way mirrors changed all that by *decoupling* those two dimensions of confrontation, marking a shift in the conditions of information-transfer that is in many ways typical of cyberspace. What does that sort of shift mean for constitutional analysis? A common way to react is to treat the pattern as it existed *prior* to the new technology (the pattern in which doing "A" necessarily *included* doing "B") as essentially arbitrary or accidental. Taking this approach, once the technological change makes it possible to do "A" *without* "B" -- to see your accuser without having him or her see you, or to read someone's mail without her knowing it, to switch examples -- one concludes that the "old" Constitution's inclusion of "B" is irrelevant; one concludes that it is enough for the government to guarantee "A" alone. Sometimes that will be the case; but it's vital to understand that, sometimes, it won't be. A characteristic feature of modernity is the subordination of purpose to accident -- an acute appreciation of just how contingent and coincidental the connections we are taught to make often are. We understand, as moderns, that many of the ways we carve up and organize the world reflect what our social history and cultural heritage, and perhaps our neurological wiring, bring to the world, and not some irreducible "way things are." A wonderful example comes from a 1966 essay by Jorge Louis Borges, "Other Inquisitions." There, the essayist describes the following taxonomy of the animal kingdom, which he purports to trace to an ancient Chinese encyclopedia entitled *The Celestial Emporium of Benevolent Knowledge*: On those remote pages it is written that animals are divided into: (a) those belonging to the Emperor (b) those that are embalmed (c) those that are trained (d) suckling pigs (e) mermaids (f) fabulous ones (g) stray dogs (h) those that are included in this classification (i) those that tremble as if they were mad (j) innumerable ones (k) those drawn with a very fine camel's hair brush (l) others (m) those that have just broken a water pitcher (n) those that, from a great distance, resemble flies Contemporary writers from Michel Foucault, in *The Archaeology of Knowledge*, through George Lakoff, in *Women, Fire, and Dangerous Things*, use Borges' Chinese encyclopedia to illustrate a range of different propositions, but the *core* proposition is the supposed arbitrariness -- the political character, in a sense -- of all culturally imposed categories. At one level, that proposition expresses a profound truth and may encourage humility by combating cultural imperialism. At another level, though, the proposition tells a dangerous lie: it suggests that we have descended into the nihilism that so obsessed Nietzsche and other thinkers -- a world where *everything* is relative, all lines are up for grabs, all principles and connections are just matters of purely subjective preference or, worse still, arbitrary convention. Whether we believe that killing animals for food is wrong, for example, becomes a question indistinguishable from whether we happen to enjoy eating beans, rice and tofu. This is a particularly pernicious notion in a era when we pass more and more of our lives in cyberspace, a place where, almost by definition, our most familiar landmarks are rearranged or disappear altogether -- because there is a pervasive tendency, even (and perhaps especially) among the most enlightened, to forget that the human values and ideals to which we commit ourselves may indeed be universal and need not depend on how our particular cultures, or our latest technologies, carve up the universe we inhabit. It was my very wise colleague from Yale, the late Art Leff, who once observed that, even in a world without an agreed-upon God, we can still agree -- even if we can't "prove" mathematically -- that "napalming babies is wrong." The Constitution's core values, I'm convinced, need not be transmogrified, or metamorphosed into oblivion, in the dim recesses of cyberspace. But to say that they *need* not be lost there is hardly to predict that they *will* not be. On the contrary, without further thought and awareness of the kind this conference might provide, the danger is clear and present that they *will* be. The "event horizon" against which this transformation might occur is already plainly visible: Electronic trespassers like Kevin Mitnik don't stop with cracking pay phones, but break into NORAD -- the North American Defense Command computer in Colorado Springs -- not in a *WarGames* movie, but in real life. Less challenging to national security but more ubiquitously threatening, computer crackers download everyman's credit history >from institutions like TRW; start charging phone calls (and more) to everyman's number; set loose "worm" programs that shut down thousands of linked computers; and spread "computer viruses" through everyman's work or home PC. It is not only the government that feels threatened by "computer crime"; both the owners and the users of private information services, computer bulletin boards, gateways, and networks feel equally vulnerable to this new breed of invisible trespasser. The response from the many who sense danger has been swift, and often brutal, as a few examples illustrate. Last March, U.S. Secret Service agents staged a surprise raid on Steve Jackson Games, a small games manufacturer in Austin, Texas, and seized all paper and electronic drafts of its newest fantasy role-playing game, *GURPS[reg.t.m.] Cyberpunk*, calling the game a "handbook for computer crime." By last Spring, up to one quarter of the U.S. Treasury Department's investigators had become involved in a project of eavesdropping on computer bulletin boards, apparently tracking notorious hackers like "Acid Phreak" and "Phiber Optik" through what one journalist dubbed "the dark canyons of cyberspace." Last May, in the now famous (or infamous) "Operation Sun Devil," more than 150 secret service agents teamed up with state and local law enforcement agencies, and with security personnel >from AT&T, American Express, U.S. Sprint, and a number of the regional Bell telephone companies, armed themselves with over two dozen search warrants and more than a few guns, and seized 42 computers and 23,000 floppy discs in 14 cities from New York to Texas. Their target: a loose-knit group of people in their teens and twenties, dubbed the "Legion of Doom." I am not describing an Indiana Jones movie. I'm talking about America in the 1990s. The Problem The Constitution's architecture can too easily come to seem quaintly irrelevant, or at least impossible to take very seriously, in the world as reconstituted by the microchip. I propose today to canvass five axioms of our constitutional law -- five basic assumptions that I believe shape the way American constitutional scholars and judges view legal issues -- and to examine how they can adapt to the cyberspace age. My conclusion (and I will try not to give away too much of the punch line here) is that the Framers of our Constitution were very wise indeed. They bequeathed us a framework for all seasons, a truly astonishing document whose principles are suitable for all times and all technological landscapes. Axiom 1: There is a Vital Difference *Between Government and Private Action* The first axiom I will discuss is the proposition that the Constitution, with the sole exception of the Thirteenth Amendment prohibiting slavery, regulates action by the *government* rather than the conduct of *private* individuals and groups. In an article I wrote in the Harvard Law Review in November 1989 on "The Curvature of Constitutional Space," I discussed the Constitution's metaphor-morphosis from a Newtonian to an Einsteinian and Heisenbergian paradigm. It was common, early in our history, to see the Constitution as "Newtonian in design with its carefully counterpoised forces and counterforces, its [geographical and institutional] checks and balances." (103 *Harv. L. Rev.* at 3.) Indeed, in many ways contemporary constitutional law is still trapped within and stunted by that paradigm. But today at least some post-modern constitutionalists tend to think and talk in the language of relativity, quantum mechanics, and chaos theory. This may quite naturally suggest to some observers that the Constitution's basic strategy of decentralizing and diffusing power by constraining and fragmenting governmental authority in particular has been rendered obsolete. The institutional separation of powers among the three federal branches of government, the geographical division of authority between the federal government and the fifty state governments, the recognition of national boundaries, and, above all, the sharp distinction between the public and private spheres, become easy to deride as relics of a simpler, pre-computer age. Thus Eli Noam, in the First Ithiel de Sola Pool Memorial Lecture, delivered last October at MIT, notes that computer networks and network associations acquire quasi-governmental powers as they necessarily take on such tasks as mediating their members' conflicting interests, establishing cost shares, creating their own rules of admission and access and expulsion, even establishing their own *de facto* taxing mechanisms. In Professor Noam's words, "networks become political entities," global nets that respect no state or local boundaries. Restrictions on the use of information in one country (to protect privacy, for example) tend to lead to export of that information to other countries, where it can be analyzed and then used on a selective basis in the country attempting to restrict it. "Data havens" reminiscent of the role played by the Swiss in banking may emerge, with few restrictions on the storage and manipulation of information. A tempting conclusion is that, to protect the free speech and other rights of *users* in such private networks, judges must treat these networks not as associations that have rights of their own *against* the government but as virtual "governments" in themselves -- as entities against which individual rights must be defended in the Constitution's name. Such a conclusion would be misleadingly simplistic. There are circumstances, of course, when non-governmental bodies like privately owned "company towns" or even huge shopping malls should be subjected to legislative and administrative controls by democratically accountable entities, or even to judicial controls as though they were arms of the state -- but that may be as true (or as false) of multinational corporations or foundations, or transnational religious organizations, or even small-town communities, as it is of computer-mediated networks. It's a fallacy to suppose that, just because a computer bulletin board or network or gateway is *something like* a shopping mall, government has as much constitutional duty -- or even authority -- to guarantee open public access to such a network as it has to guarantee open public access to a privately owned shopping center like the one involved in the U.S. Supreme Court's famous *PruneYard Shopping Center* decision of 1980, arising from nearby San Jose. The rules of law, both statutory and judge-made, through which each state *allocates* private powers and responsibilities themselves represent characteristic forms of government action. That's why a state's rules for imposing liability on private publishers, or for deciding which private contracts to enforce and which ones to invalidate, are all subject to scrutiny for their consistency with the federal Constitution. But as a general proposition it is only what *governments* do, either through such rules or through the actions of public officials, that the United States Constitution constrains. And nothing about any new technology suddenly erases the Constitution's enduring value of restraining *government* above all else, and of protecting all private groups, large and small, >from government. It's true that certain technologies may become socially indispensable -- so that equal or at least minimal access to basic computer power, for example, might be as significant a constitutional goal as equal or at least minimal access to the franchise, or to dispute resolution through the judicial system, or to elementary and secondary education. But all this means (or should mean) is that the Constitution's constraints on government must at times take the form of imposing *affirmative duties* to assure access rather than merely enforcing *negative prohibitions* against designated sorts of invasion or intrusion. Today, for example, the government is under an affirmative obligation to open up criminal trials to the press and the public, at least where there has not been a particularized finding that such openness would disrupt the proceedings. The government is also under an affirmative obligation to provide free legal assistance for indigent criminal defendants, to assure speedy trials, to underwrite the cost of counting ballots at election time, and to desegregate previously segregated school systems. But these occasional affirmative obligations don't, or shouldn't, mean that the Constitution's axiomatic division between the realm of public power and the realm of private life should be jettisoned. Nor would the "indispensability" of information technologies provide a license for government to impose strict content, access, pricing, and other types of regulation. *Books* are indispensable to most of us, for example -- but it doesn't follow that government should therefore be able to regulate the content of what goes onto the shelves of *bookstores*. The right of a private bookstore owner to decide which books to stock and which to discard, which books to display openly and which to store in limited access areas, should remain inviolate. And note, incidentally, that this needn't make the bookstore owner a "publisher" who is liable for the words printed in the books on her shelves. It's a common fallacy to imagine that the moment a computer gateway or bulletin board begins to exercise powers of selection to control who may be on line, it must automatically assume the responsibilities of a newscaster, a broadcaster, or an author. For computer gateways and bulletin boards are really the "bookstores" of cyberspace; most of them organize and present information in a computer format, rather than generating more information content of their own. Axiom 2: The Constitutional Boundaries of Private Property and Personality Depend on Variables Deeper Than *Social Utility and Technological Feasibility* The second constitutional axiom, one closely related to the private-public distinction of the first axiom, is that a person's mind, body, and property belong *to that person* and not to the public as a whole. Some believe that cyberspace challenges that axiom because its entire premise lies in the existence of computers tied to electronic transmission networks that process digital information. Because such information can be easily replicated in series of "1"s and "0"s, anything that anyone has come up with in virtual reality can be infinitely reproduced. I can log on to a computer library, copy a "virtual book" to my computer disk, and send a copy to your computer without creating a gap on anyone's bookshelf. The same is true of valuable computer programs, costing hundreds of dollars, creating serious piracy problems. This feature leads some, like Richard Stallman of the Free Software Foundation, to argue that in cyberspace everything should be free -- that information can't be owned. Others, of course, argue that copyright and patent protections of various kinds are needed in order for there to be incentives to create "cyberspace property" in the first place. Needless to say, there are lively debates about what the optimal incentive package should be as a matter of legislative and social policy. But the only *constitutional* issue, at bottom, isn't the utilitarian or instrumental selection of an optimal policy. Social judgments about what ought to be subject to individual appropriation, in the sense used by John Locke and Robert Nozick, and what ought to remain in the open public domain, are first and foremost *political* decisions. To be sure, there are some constitutional constraints on these political decisions. The Constitution does not permit anything and everything to be made into a *private commodity*. Votes, for example, theoretically cannot be bought and sold. Whether the Constitution itself should be read (or amended) so as to permit all basic medical care, shelter, nutrition, legal assistance and, indeed, computerized information services, to be treated as mere commodities, available only to the highest bidder, are all terribly hard questions -- as the Eastern Europeans are now discovering as they attempt to draft their own constitutions. But these are not questions that should ever be confused with issues of what is technologically possible, about what is realistically enforceable, or about what is socially desirable. Similarly, the Constitution does not permit anything and everything to be *socialized* and made into a public good available to whoever needs or "deserves" it most. I would hope, for example, that the government could not use its powers of eminent domain to "take" live body parts like eyes or kidneys or brain tissue for those who need transplants and would be expected to lead particularly productive lives. In any event, I feel certain that whatever constitutional right each of us has to inhabit his or her own body and to hold onto his or her own thoughts and creations should not depend solely on cost-benefit calculations, or on the availability of technological methods for painlessly effecting transfers or for creating good artificial substitutes. Axiom 3: *Government May Not Control Information Content* A third constitutional axiom, like the first two, reflects a deep respect for the integrity of each individual and a healthy skepticism toward government. The axiom is that, although information and ideas have real effects in the social world, it's not up to government to pick and choose for us in terms of the *content* of that information or the *value* of those ideas. This notion is sometimes mistakenly reduced to the naive child's ditty that "sticks and stones may break my bones, but words can never hurt me." Anybody who's ever been called something awful by children in a schoolyard knows better than to believe any such thing. The real basis for First Amendment values isn't the false premise that information and ideas have no real impact, but the belief that information and ideas are *too important* to entrust to any government censor or overseer. If we keep that in mind, and *only* if we keep that in mind, will we be able to see through the tempting argument that, in the Information Age, free speech is a luxury we can no longer afford. That argument becomes especially tempting in the context of cyberspace, where sequences of "0"s and "1"s may become virtual life forms. Computer "viruses" roam the information nets, attaching themselves to various programs and screwing up computer facilities. Creation of a computer virus involves writing a program; the program then replicates itself and mutates. The electronic code involved is very much like DNA. If information content is "speech," and if the First Amendment is to apply in cyberspace, then mustn't these viruses be "speech" -- and mustn't their writing and dissemination be constitutionally protected? To avoid that nightmarish outcome, mustn't we say that the First Amendment is *inapplicable* to cyberspace? The answer is no. Speech is protected, but deliberately yelling "Boo!" at a cardiac patient may still be prosecuted as murder. Free speech is a constitutional right, but handing a bank teller a hold-up note that says, "Your money or your life," may still be punished as robbery. Stealing someone's diary may be punished as theft -- even if you intend to publish it in book form. And the Supreme Court, over the past fifteen years, has gradually brought advertising within the ambit of protected expression without preventing the government from protecting consumers from deceptive advertising. The lesson, in short, is that constitutional principles are subtle enough to bend to such concerns. They needn't be broken or tossed out. Axiom 4: The Constitution is Founded on Normative Conceptions of Humanity That Advances *in Science and Technology Cannot "Disprove"* A fourth constitutional axiom is that the human spirit is something beyond a physical information processor. That axiom, which regards human thought processes as not fully reducible to the operations of a computer program, however complex, must not be confused with the silly view that, because computer operations involve nothing more than the manipulation of "on" and "off" states of myriad microchips, it somehow follows that government control or outright seizure of computers and computer programs threatens no First Amendment rights because human thought processes are not directly involved. To say that would be like saying that government confiscation of a newspaper's printing press and tomorrow morning's copy has nothing to do with speech but involves only a taking of metal, paper, and ink. Particularly if the seizure or the regulation is triggered by the content of the information being processed or transmitted, the First Amendment is of course fully involved. Yet this recognition that information processing by computer entails something far beyond the mere sequencing of mechanical or chemical steps still leaves a potential gap between what computers can do internally and in communication with one another -- and what goes on within and between human minds. It is that gap to which this fourth axiom is addressed; the very existence of any such gap is, as I'm sure you know, a matter of considerable controversy. What if people like the mathematician and physicist Roger Penrose, author of *The Emperor's New Mind*, are wrong about human minds? In that provocative recent book, Penrose disagrees with those Artificial Intelligence, or AI, gurus who insist that it's only a matter of time until human thought and feeling can be perfectly simulated or even replicated by a series of purely physical operations -- that it's all just neurons firing and neurotransmitters flowing, all subject to perfect modeling in suitable computer systems. Would an adherent of that AI orthodoxy, someone whom Penrose fails to persuade, have to reject as irrelevant for cyberspace those constitutional protections that rest on the anti-AI premise that minds are *not* reducible to really fancy computers? Consider, for example, the Fifth Amendment, which provides that "no person shall be . . . compelled in any criminal case to be a witness against himself." The Supreme Court has long held that suspects may be required, despite this protection, to provide evidence that is not "testimonial" in nature -- blood samples, for instance, or even exemplars of one's handwriting or voice. Last year, in a case called *Pennsylvania v. Muniz*, the Supreme Court held that answers to even simple questions like "When was your sixth birthday?" are testimonial because such a question, however straightforward, nevertheless calls for the product of mental activity and therefore uses the suspect's mind against him. But what if science could eventually describe thinking as a process no more complex than, say, riding a bike or digesting a meal? Might the progress of neurobiology and computer science eventually overthrow the premises of the *Muniz* decision? I would hope not. For the Constitution's premises, properly understood, are *normative* rather than *descriptive*. The philosopher David Hume was right in teaching that no "ought" can ever be logically derived from an "is." If we should ever abandon the Constitution's protection for the distinctively and universally human, it won't be because robotics or genetic engineering or computer science have led us to deeper truths, but rather because they have seduced us into more profound confusions. Science and technology open options, create possibilities, suggest incompatibilities, generate threats. They do not alter what is "right" or what is "wrong." The fact that those notions are elusive and subject to endless debate need not make them totally contingent on contemporary technology. Axiom 5: Constitutional Principles Should Not *Vary With Accidents of Technology* In a sense, that's the fifth and final constitutional axiom I would urge upon this gathering: that the Constitution's norms, at their deepest level, must be invariant under merely *technological* transformations. Our constitutional law evolves through judicial interpretation, case by case, in a process of reasoning by analogy >from precedent. At its best, that process is ideally suited to seeing beneath the surface and extracting deeper principles from prior decisions. At its worst, though, the same process can get bogged down in superficial aspects of preexisting examples, fixating upon unessential features while overlooking underlying principles and values. When the Supreme Court in 1928 first confronted wiretapping and held in *Olmstead v. United States* that such wiretapping involved no "search" or "seizure" within the meaning of the Fourth Amendment's prohibition of "unreasonable searches and seizures," the majority of the Court reasoned that the Fourth Amendment "itself shows that the search is to be of material things -- the person, the house, his papers or his effects," and said that "there was no searching" when a suspect's phone was tapped because the Constitution's language "cannot be extended and expanded to include telephone wires reaching to the whole world from the defendant's house or office." After all, said the Court, the intervening wires "are not part of his house or office any more than are the highways along which they are stretched." Even to a law student in the 1960s, as you might imagine, that "reasoning" seemed amazingly artificial. Yet the *Olmstead* doctrine still survived. It would be illuminating at this point to compare the Supreme Court's initial reaction to new technology in *Olmstead* with its initial reaction to new technology in *Maryland v. Craig*, the 1990 closed-circuit television case with which we began this discussion. In *Craig*, a majority of the Justices assumed that, when the 18th- century Framers of the Confrontation Clause included a guarantee of two-way *physical* confrontation, they did so solely because it had not yet become technologically feasible for the accused to look his accuser in the eye without having the accuser simultaneously watch the accused. Given that this technological obstacle has been removed, the majority assumed, one-way confrontation is now sufficient. It is enough that the accused not be subject to criminal conviction on the basis of statements made outside his presence. In *Olmstead*, a majority of the Justices assumed that, when the 18th-century authors of the Fourth Amendment used language that sounded "physical" in guaranteeing against invasions of a person's dwelling or possessions, they did so not solely because *physical* invasions were at that time the only serious threats to personal privacy, but for the separate and distinct reason that *intangible* invasions simply would not threaten any relevant dimension of Fourth Amendment privacy. In a sense, *Olmstead* mindlessly read a new technology *out* of the Constitution, while *Craig* absent-mindedly read a new technology *into* the Constitution. But both decisions -- *Olmstead* and *Craig* -- had the structural effect of withholding the protections of the Bill of Rights from threats made possible by new information technologies. *Olmstead* did so by implausibly reading the Constitution's text as though it represented a deliberate decision not to extend protection to threats that 18th-century thinkers simply had not foreseen. *Craig* did so by somewhat more plausibly -- but still unthinkingly -- treating the Constitution's seemingly explicit coupling of two analytically distinct protections as reflecting a failure of technological foresight and imagination, rather than a deliberate value choice. The *Craig* majority's approach appears to have been driven in part by an understandable sense of how a new information technology could directly protect a particularly sympathetic group, abused children, from a traumatic trial experience. The *Olmstead* majority's approach probably reflected both an exaggerated estimate of how difficult it would be to obtain wiretapping warrants even where fully justified, and an insufficient sense of how a new information technology could directly threaten all of us. Although both *Craig* and *Olmstead* reveal an inadequate consciousness about how new technologies interact with old values, *Craig* at least seems defensible even if misguided, while *Olmstead* seems just plain wrong. Around 23 years ago, as a then-recent law school graduate serving as law clerk to Supreme Court Justice Potter Stewart, I found myself working on a case involving the government's electronic surveillance of a suspected criminal -- in the form of a tiny device attached to the outside of a public telephone booth. Because the invasion of the suspect's privacy was accomplished without physical trespass into a "constitutionally protected area," the Federal Government argued, relying on *Olmstead*, that there had been no "search" or "seizure," and therefore that the Fourth Amendment "right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures," simply did not apply. At first, there were only four votes to overrule *Olmstead* and to hold the Fourth Amendment applicable to wiretapping and electronic eavesdropping. I'm proud to say that, as a 26-year-old kid, I had at least a little bit to do with changing that number >from four to seven -- and with the argument, formally adopted by a seven-Justice majority in December 1967, that the Fourth Amendment "protects people, not places." (389 U.S. at 351.) In that decision, *Katz v. United States*, the Supreme Court finally repudiated *Olmstead* and the many decisions that had relied upon it and reasoned that, given the role of electronic telecommunications in modern life, the First Amendment purposes of protecting *free speech* as well as the Fourth Amendment purposes of protecting *privacy* require treating as a "search" any invasion of a person's confidential telephone communications, with or without physical trespass. Sadly, nine years later, in *Smith v. Maryland*, the Supreme Court retreated from the *Katz* principle by holding that no search occurs and therefore no warrant is needed when police, with the assistance of the telephone company, make use of a "pen register", a mechanical device placed on someone's phone line that records all numbers dialed from the phone and the times of dialing. The Supreme Court, over the dissents of Justices Stewart, Brennan, and Marshall, found no legitimate expectation of privacy in the numbers dialed, reasoning that the digits one dials are routinely recorded by the phone company for billing purposes. As Justice Stewart, the author of *Katz*, aptly pointed out, "that observation no more than describes the basic nature of telephone calls . . . . It is simply not enough to say, after *Katz*, that there is no legitimate expectation of privacy in the numbers dialed because the caller assumes the risk that the telephone company will expose them to the police." (442 U.S. at 746-747.) Today, the logic of *Smith* is being used to say that people have no expectation of privacy when they use their cordless telephones since they know or should know that radio waves can be easily monitored! It is easy to be pessimistic about the way in which the Supreme Court has reacted to technological change. In many respects, *Smith* is unfortunately more typical than *Katz* of the way the Court has behaved. For example, when movies were invented, and for several decades thereafter, the Court held that movie exhibitions were not entitled to First Amendment protection. When community access cable TV was born, the Court hindered municipal attempts to provide it at low cost by holding that rules requiring landlords to install small cable boxes on their apartment buildings amounted to a compensable taking of property. And in *Red Lion v. FCC*, decided twenty-two years ago but still not repudiated today, the Court ratified government control of TV and radio broadcast content with the dubious logic that the scarcity of the electromagnetic spectrum justified not merely government policies to auction off, randomly allocate, or otherwise ration the spectrum according to neutral rules, but also much more intrusive and content-based government regulation in the form of the so-called "fairness doctrine." Although the Supreme Court and the lower federal courts have taken a somewhat more enlightened approach in dealing with cable television, these decisions for the most part reveal a curious judicial blindness, as if the Constitution had to be reinvented with the birth of each new technology. Judges interpreting a late 18th century Bill of Rights tend to forget that, unless its *terms* are read in an evolving and dynamic way, its *values* will lose even the *static* protection they once enjoyed. Ironically, *fidelity* to original values requires *flexibility* of textual interpretation. It was Judge Robert Bork, not famous for his flexibility, who once urged this enlightened view upon then Judge (now Justice) Scalia, when the two of them sat as colleagues on the U.S. Court of Appeals for the D.C. Circuit. Judicial error in this field tends to take the form of saying that, by using modern technology ranging from the telephone to the television to computers, we "assume the risk." But that typically begs the question. Justice Harlan, in a dissent penned two decades ago, wrote: "Since it is the task of the law to form and project, as well as mirror and reflect, we should not . . . merely recite . . . risks without examining the *desirability* of saddling them upon society." (*United States v. White*, 401 U.S. at 786). And, I would add, we should not merely recite risks without examining how imposing those risks comports with the Constitution's fundamental values of *freedom*, *privacy*, and *equality*. Failing to examine just that issue is the basic error I believe federal courts and Congress have made: * in regulating radio and TV broadcasting without adequate sensitivity to First Amendment values; * in supposing that the selection and editing of video programs by cable operators might be less than a form of expression; * in excluding telephone companies from cable and other information markets; * in assuming that the processing of "O"s and "1"s by computers as they exchange data with one another is something less than "speech"; and * in generally treating information processed electronically as though it were somehow less entitled to protection for that reason. The lesson to be learned is that these choices and these mistakes are not dictated by the Constitution. They are decisions for us to make in interpreting that majestic charter, and in implementing the principles that the Constitution establishes. *Conclusion* If my own life as a lawyer and legal scholar could leave just one legacy, I'd like it to be the recognition that the Constitution *as a whole* "protects people, not places." If that is to come about, the Constitution as a whole must be read through a technologically transparent lens. That is, we must embrace, as a rule of construction or interpretation, a principle one might call the "cyberspace corollary." It would make a suitable Twenty-seventh Amendment to the Constitution, one befitting the 200th anniversary of the Bill of Rights. Whether adopted all at once as a constitutional amendment, or accepted gradually as a principle of interpretation that I believe should obtain even without any formal change in the Constitution's language, the corollary I would propose would do for *technology* in 1991 what I believe the Constitution's Ninth Amendment, adopted in 1791, was meant to do for *text*. The Ninth Amendment says: "The enumeration in the Constitution, of certain rights, shall not be construed to deny or disparage others retained by the people." That amendment provides added support for the long-debated, but now largely accepted, "right of privacy" that the Supreme Court recognized in such decisions as the famous birth control case of 1965, *Griswold v. Connecticut*. The Ninth Amendment's simple message is: The *text* used by the Constitution's authors and ratifiers does not exhaust the values our Constitution recognizes. Perhaps a Twenty-seventh Amendment could convey a parallel and equally simple message: The *technologies* familiar to the Constitution's authors and ratifiers similarly do not exhaust the *threats* against which the Constitution's core values must be protected. The most recent amendment, the twenty-sixth, adopted in 1971, extended the vote to 18-year-olds. It would be fitting, in a world where youth has been enfranchised, for a twenty-seventh amendment to spell a kind of "childhood's end" for constitutional law. The Twenty-seventh Amendment, to be proposed for at least serious debate in 1991, would read simply: "This Constitution's protections for the freedoms of speech, press, petition, and assembly, and its protections against unreasonable searches and seizures and the deprivation of life, liberty, or property without due process of law, shall be construed as fully applicable without regard to the technological method or medium through which information content is generated, stored, altered, transmitted, or controlled."