Civil Liberties and National Security on the Internet
by Kate Martin

"If the resources available on the Internet—the deeper sense of other people's lives and kinds of information that will be theirs to examine and explore—if these things work on and strengthen the imaginations of those who use them—well, then you have something that can have great significance for the cause of world peace. Because, you see, a key to compassion and the urge to moral action is the ability to imagine someone else's life and circumstances and how it feels to be that person in those circumstances of war, famine, or imprisonment or political oppression."
--Father Andrew Greeley
"Leading security experts predict that it is only several years before a terrorist or rogue nation is capable of an on-line, hacker-style attack against the United States, causing massive failure of such crucial elements as banking or the financial markets, transportation systems, the power grid or telecommunications."
--USA Today


The vast power of modern computer networks presents an extraordinary opportunity to advance core civil liberties principles of freedom of expression and privacy in the United States and throughout the world. The free trade in ideas that results from expansive connectivity promotes freedom, democracy, and a more global sense of community.

At the same time, the advent of global communications and technology also raises new and serious national security concerns. Little attention has yet been given to the intersection of these issues.

National security claims have always been one of the major threats to and justifications for restricting civil liberties. But in this post-Cold War world, human rights and democracy are now key components of foreign policy and matters of concern to the Defense Department and the CIA. All this comes together in clashing or complementary ways on the Internet.

This paper proposes that we need a new examination and analysis of the relation between national security interests and individual liberties on the Internet. Does the Internet require that we rethink historic accommodations between civil liberties and national security interests? Will the protection of freedom of expression and privacy on the Internet actually advance national security interests? Or will national security concerns lead to restriction of civil liberties on the Internet?

This paper is offered to catalyze critical consideration of these issues. To that end, it outlines a variety of questions, both specific legal ones and broader policy ones and proposes some possible new analyses, all of which merit further study. I begin by suggesting a new understanding of national security to be used in looking at these issues on the Internet; one that recognizes that promotion of civil liberties actually serves national security interests. Such an understanding informs consideration of more specific questions: What are the implications of the growth of the Internet for the historic wall between law enforcement and national security, erected to safeguard civil liberties? How are First Amendment interests in free speech even for bomb-makers and in access to government information affected by national security concerns about the Internet? Next, the paper outlines a series of questions relating to privacy and surveillance on the Internet, for example, pointing out the gaping holes in existing legal regimes protecting overseas electronic communications. In that section, I propose a new analysis of the Fourth Amendment, one which would not confine its operation to the evaluation of the reasonableness of particular searches and seizures but to a broader consideration of the existence of enormous technological power by the government to invade individual privacy. The paper then outlines some civil liberties/national security questions regarding international law and the development of the rule of law in other countries. For perhaps the first time, the question of whether the CIA or other agencies should be allowed to use the Internet to conduct covert actions abroad is then posed. The paper ends by outlining a sociological question: what are the non-legal ways the Internet can be used to promote civil liberties and discourage violence that might threaten the national security?

Must National Security Always Oppose Civil Liberties?

The breakdown of international barriers including the rise of the Internet is challenging historic notions of geography, legal jurisdiction, and even sovereignty. That, coupled with greater cooperation between governments, calls for a reexamination of traditional conceptions of national security. Historically, national security interests and civil liberties interests have been understood as being quite separate, usually opposed to one another or at best uneasily existing side-by-side.

But post-Cold war changes in U.S. foreign policy together with the rise in connectivity suggest that such a view is incomplete and outmoded. The President and the Congress have decided that U.S. national security objectives now include promotion of democracy, human rights and the rule of law in many places around the world. And as much as the national security apparatus has pointed to the tangible dangers of networking criminals, drug dealers, and terrorists together, this same connective power also promotes civil liberties and democratic principles.

Democracy and human rights are certainly aided, for instance, by citizens having unfettered access to the World Wide Web. David Halperin, in a essay on the Internet and national security, suggests that the vast power of the Web in this regard is implied, for example, by actions taken by Chinese officials to prevent its citizens from accessing news information and activist materials.(1) Just last year, in fact, China adopted regulations making it a crime to use the Internet to promote independence movements. Elsewhere, and also illustrative of the Web's power to spread democratic ideas, Serbian President Slobodan Milosevic began jamming the signal of Radio B92, the mouthpiece of the Serbian pro-democracy movement, after the radio station began transmiting its broadcasts to the world via the Internet.(2) Mr. Halperin writes that "the Serbian experience suggests the Internet's power to affect politics even in a country where Internet-connected computers are few and far between. As such computers become cheaper and cheaper, and telephone service and Internet providers become more prevalent, the capability of the Net to transform societies grows."(3)

By advancing core civil liberties and human rights principles through the free exchange of information, free expression by individuals, and the basic liberty of having a private life, the Internet can also advance U.S. national security interests. The Internet mandates a new understanding of national security interests as not being opposed to civil liberties but in many cases served by civil liberties protections. In many instances, the promotion of civil liberties on the Internet advances, rather than conflicts with, national security interests.

This understanding should have a concrete effect on public debate and bureaucratic decision-making. For example, current debate regarding the restrictions on encryption programs usually pits privacy and free speech advocates against those arguing national security interests. However, it is not readily understood that protecting the civil liberties interests at stake may, in fact, also contribute to national security goals, for, as in the case of encryption, the advancement of human rights that the new technology brings about also eases U.S. national security concerns.

Cybercriminals and terrorists: What was a crime is now a national security threat.

Many have pointed to the Internet's blurring of the distinction between the foreign and the domestic as a way in which national security threats are brought closer by connectivity. In warning about the dangers of terrorists on the Internet, rarely is any distinction made between home-grown threats and foreign ones. Similarly, national security threats have been expanded to include what are traditionally criminal, not national security, problems. Now included in the litany of national security dangers is, for example, the Russian Mafia's operating in New York and using the Internet. Erosion of the lines between the foreign and the domestic and between criminal activity and national security threats will have profound effects on basic institutional safeguards against civil liberties abuses. While these lines are already being eroded, their blurring is likely to be escalated by the growth of the Internet.

In response to widespread political spying and other abuses in the 1950s and 1960s, an institutional and conceptual wall between domestic law enforcement and foreign intelligence was built in the 1970's. This wall has been one of the most crucial safeguards against civil liberties abuses. This distinction between the foreign and the domestic underlies the legal regimes governing the Federal Bureau of Investigation, the Central Intelligence Agency, and other intelligence agencies. While the CIA is charged with foreign intelligence gathering and covert actions overseas, for example, the FBI concentrates on domestic matters. Similarly, domestic wiretapping is governed by Title III, while eavesdropping on foreign powers and their agents is governed by the Foreign Intelligence Surveillance Act.

This wall between domestic law enforcement and foreign intelligence is now being dismantled in response to government claims that such dismantling is required to deal with new threats from the Internet and elsewhere. Close examination of how these developments will affect civil liberties is needed.

As a recent report of the President's Commission on Critical Infrastructure Protection(4) ("Infrastructure Report") outlines, attacks on cyber-networks may be either the work of common criminals or part of a terrorist attack, and the government will probably be unable, initially at least, to identify the perpetrators.(5) Will the threat of such incidents and preparations therefor be treated as law enforcement matters or national security threats?

How these questions are answered will determine not only which U.S. government agencies respond but, more significantly from the civil liberties standpoint, what the response will be. For example, how much intelligence gathering on the activities of Americans should be allowed in order to prevent such attacks, and what are the limits on such collection? Do we need new rules? In the national security realm, the law has tolerated greater government secrecy, less judicial review, and lesser Fourth Amendment protection than in other realms, including law enforcement. Does it make sense to simply import such national security restrictions on civil liberties crafted for the Cold War to this brand new medium in a quite different world? Do we need a new evaluation of their appropriateness? What is most striking about all the recent government reports, like the Infrastructure Report, is the complete silence about these issues of constitutional rights.

Following are some specific problems regarding the Internet, national security, and civil liberties that merit extensive discussion and require further study.

Freedom of Speech and Access to Information Issues.

What are the First Amendment implications of extensive and detailed bomb-making information becoming widely available on the Internet? There are already political efforts to censor such information, although it is relatively clear the First Amendment outlaws such censorship.(6) Will there be a new First Amendment analysis offered in support of such efforts? Will it be that new categories of content should be carved out as outside the protection of the First Amendment given the new connectivity, interdependence of the world, and the increased access by millions to such information? Will these factors lead to the dilution of existing rules for protecting speech? Scholars are already asking whether the fact that the Internet makes bomb instructions so much more easily available to so many more people changes the traditional First Amendment calculus by lessening traditional protections.(7)

At the same time, the government has characterized as a national security threat -- rather than simply criminal violence -- isolated bombings by individuals regardless of any articulated political objective (such as the Oklahoma City bombing). Such characterization of course, weighs heavily in any First Amendment balancing.

Thus, any consideration of censoring bomb-making instructions that relies upon the new and unique nature of the Internet as a factor weighing towards less First Amendment protection is incomplete without, at the same time, a new analysis of the "national security interests" on the other side. Are all such interests equally weighty? Isn't the threat of nuclear annihilation in fact much different from the threat of another Oklahoma City bombing? And don't we need to consider whether acceptance of a principle that says content on the Internet may be censored by virtue of being on the Internet may itself have adverse national security consequences? Would adoption of such a principle make Internet censorship worldwide much more likely and make the Internet much less useful in promoting democracy and human rights, thereby having a negative impact on U.S. national security?

We also need to examine how the growth of the Internet is likely to further the First Amendment value of access to government information. While in general, of course, the Internet vastly increases the public's access to all kinds of information, specific characteristics of the Internet may at the same time be used as grounds to decrease access to government information in particular. Such access is now, in large part, guaranteed by the Freedom of Information Act. But the Infrastructure Report(8), in the course of recommending much greater sharing of information between the government and the private sector with no discussion then recommended changes to the Freedom of Information Act to exempt such information from public disclosure.

Will the increased points of connection between government and private parties in building the Internet in fact result in a new -- and more restricted -- definition of what information concerns government activities and, therefore, is subject to disclosure under the Freedom of Information Act? Will the increase in technological partnerships between public and private parties increase transparency by making publicly available more information on the non-governmental sector? Or, will increasing public-private partnerships, in fact, decrease transparency by making more government information secret?

The answers to these questions will depend upon whether one begins with a presumption of disclosure.(9) I would argue that such information should be subject to the FOIA because it is functionally like other information in the possession of the government -- it is relevant to democratic decision-making. Then it must be asked whether there are good reasons to exempt narrow and specific categories of such information from disclosure? What are the costs of doing so? This proposal of course, simply calls for application of traditional FOIA analysis to such information, where the presumption is that all government information belongs to the people and must be disclosed unless Congress legislates narrow and specific categories of exemption after a cost-benefit examination.

Privacy, Surveillance and National Security on the Internet.

In the past, debates over Fourth Amendment protections against government intrusions on privacy have largely occurred in the context of criminal law enforcement and the government's need to obtain information to identify and convict criminals. But the Fourth Amendment issues relating to privacy on the Internet are seen by the government through the prism not only of law enforcement interests, but also of national security interests.(10) Once again, such characterization includes the danger of according too much weight to such interests and calls for specific identification of the national security harms being referred to and a skeptical look at their imminence and probability.

At the same time, it has been difficult to fit Internet privacy concerns neatly into traditional Fourth Amendment analysis: Was there a search or seizure by the government where the targeted individual had a reasonable expectation of privacy? Was there probable cause? A judicial warrant? Was the search or seizure reasonable? Such analysis has proven less than fully satisfactory for privacy advocates, particularly on structural issues like encryption or the design of the telephone network, where the government argues that its proposals always recognize that it will not seize, decrypt or read the contents of any communication without meeting traditional Fourth Amendment requirements of probable cause, a warrant, etc. Similarly, it has been difficult to articulate a Fourth Amendment objection to the government's encouraging or even mandating the design of wireless telephones and modems so that they automatically transmit the user's location, so long as the government does not access such information without probable cause and a warrant.

So we must ask whether new technology and Internet connectivity so alters the world that a new Fourth Amendment analysis is called for. A new and broader understanding of how the constitutional guarantee is meant to work seems required.

The Fourth Amendment can be understood as a restriction in law on the power of the government vis-à-vis the individual. But the other restriction on government power, is of course the practical one of lack of capacity. In 1791, governmental power to invade an individual's privacy was constrained by technological limitations, the Fourth Amendment added the constraint of law to the government's limited capabilities. It was not possible to eavesdrop on an individual's conversations inside his or her bedroom, unless perhaps someone stood outside with her ear to the wall. Before the invention of the telephone, when individuals had to communicate either in person or by writing, the government lacked the means of surreptitiously overhearing a conversation between two non-consenting individuals.(11) And even when government agents did seize papers, the limited number of government agents and the size of the government budget made any wholesale seizure and reading of large numbers of individuals' papers virtually impossible. Thus, individual privacy was protected in two equally important ways: by the requirements of the law and by the technological incapacity of the government to effect widespread surveillance of private communications.(12)

Now it seems that the second leg of that protection is lost and in its absence, we must ask whether the thin reed of the law, especially as whittled away by recent Supreme Court decisions, will in fact be enough to protect privacy. Will the requirement that government searches and seizures be reasonable really be sufficient to defeat the temptations of vast power?

Beginning with the telephone, the government's surveillance capabilities have exponentially increased in recent years. It can cheaply and easily capture tens of thousands of electronic communications and then use computers to scan them for interesting content. Computer databases can easily collate and organize mountains of information. And with the growth of the Internet and globalization of communications, it is likely the case that many more private communications are conducted electronically rather than face to face, at least in many parts of the world. At the same time, the government can also now surreptitiously capture the most private communications held face to face. And many of the private papers that were kept in one's house in 1791 are now stored on the servers and computers of third party corporations.

Thus, the state's power to conduct surveillance and invade privacy has fundamentally and drastically increased at the expense of the individual. Simultaneously, crabbed readings by the Supreme Court have also had the effect of steadily diminishing the scope and protections of the Fourth Amendment.

The genius of the framers was to recognize that despite the good intentions of good men, it is a sure road to tyranny to embody too much power in any group of them. The Constitution does not rely simply on legal prohibitions to protect against abuse of liberty; the framers also sought to limit the power of the governors. And with this historical understanding and the current technological context, we can find the seeds for a new analysis of the Fourth Amendment: one that is not limited to judging the propriety of any particular search as reasonable, but also requires maintenance of an appropriate balance of power between government surveillance capabilities and individuals so that protection of individual privacy against government abuse does not depend solely upon a legal requirement of reasonableness in a particular instance.

From this point of view, it is an interesting theoretical question whether the mere development of technological capabilities by the government could in itself so skew the balance between state and individual power as to threaten the constitutional scheme and make such technological developments subject to direct challenge under the Fourth Amendment. But for now, we face the easier question of how to evaluate the impact on Fourth Amendment values of legislated restrictions on private sector development of technologies that counteract and restore the technological balance of power between government and the individual. I am suggesting that, in order to assure that constitutional protections against abuse of individual privacy remain meaningful in this new world, such evaluation should encompass consideration of the technological balance of power between state and individual, in addition to whether illegal searches or seizures are being authorized. Such analysis should prove useful in answering many of the new privacy and national security questions raised by the growth of the Internet.

Following are some of those questions suggested for further consideration.

1. The most discussed of such questions is the current debate over controls on the availability of strong encryption. Encryption technology allows Internet users to easily encrypt their messages, and the FBI claims that it will be unable to decipher messages sent by terrorists and other criminals. The government is seeking the adoption of a key-escrow system whereby users of encryption would deposit the keys to their codes with a designated third party. When the government has met the applicable legal requirements, such as a judicial warrant, it could seize the keys and read the encrypted information. The government claims national security interests will be compromised in the absence of such a key-escrow system, which would guarantee its ability to decipher information. Internet users answer that only strong encryption will adequately protect the privacy of electronic communications not only from government intrusion, but from criminals and hackers. These "clashing imperatives," as Richard Epstein calls them, have framed the current controversy over encryption technology.

As mentioned above, this is an instance where traditional Fourth Amendment analysis does not fit very well, and yet it is clear that serious privacy interests are threatened by government attempts to ban encryption.(13) The government's easy response to privacy concerns has been to promise that it will not obtain the escrowed keys and use them to read any communications or information without complying with applicable Fourth Amendment requirements.

Here, the suggested Fourth Amendment analysis outlined above could be useful. Applying this analysis, legislated restrictions on the development of technology that enables individuals to make it harder for the government to listen to their communications (in an era when the government's ability to eavesdrop has exponentially increased) must be evaluated in terms of their effect on the balance of technological power between government and individual. The effect of such efforts to impose a key-escrow scheme in order to weaken encryption capabilities is to skew further the balance of power in favor of the government, and such restrictions would therefore be prohibited under this view of Fourth Amendment protections. When the government corals both technology and the law to give it overwhelming power vis-à-vis individuals, real dangers of abuse arise. Today, those dangers can be met through technological self-defense if the government is restricted to relying on technological advantages and does not, at the same time, attempt to make new laws to outlaw such technological self defense.

In addition, closer examination and analysis of the government's national security claims is required. Specifically, isn't the government's listing of the national security effects of strong encryption incomplete? Assuming that the availability of strong encryption is likely to cause some harm to national security interests -- or, at a minimum, make it more difficult to solve or prevent some crimes -- isn't it also the case that the availability of strong encryption at the same time advances other national security interests by helping to spread democratic ideas and bolstering respect for the rule of law and fundamental human rights? The mantra of national security -- especially when tied to gruesome possibilities -- should not substitute for careful and complete analysis.

2. Does the growth of the Internet require a re-evaluation of Fourth Amendment precedent concerning the lack of constitutional protection for information held by third parties? Does it make sense to afford less Fourth Amendment protection to e-mail messages stored in America Online's servers than to messages printed out and kept in a drawer in one's house?

Since the Supreme Court ruled that the Fourth Amendment did not protect information voluntarily left with third parties, like financial information contained in bank records, many statutory regimes have been enacted providing various levels of protection for different kinds of information held by third parties. As long as this process is conducted on an ad hoc statutory basis, privacy advocates will find themselves having to argue in every case that such information should be protected in the way it would be if the Fourth Amendment were held to apply.(14) It is time to reconsider whether constitutional protections do, in fact, apply.

3. Given the increased surveillance and information-gathering capabilities of the government, do we need new rules for protection against government collection of publicly available Internet information about individuals? Current precedent recognizes no Fourth Amendment-protectable interest against government seizure of publicly available Internet information about individuals when done for national security reasons. While the Privacy Act on its face prohibits national security collection of such information regarding individuals' First Amendment-protected activities, the CIA and FBI do so anyway and the courts have failed to stop them.(15)

Do current laws and guidelines about intelligence agency collection of open source information need to be re-evaluated in light of the massive amounts of personal information becoming available on the Internet? In what circumstances, if any, should such information be permitted? Can we reopen the question of Fourth Amendment protection against such activity? At the least, doesn't the Privacy Act need to be reconsidered?

4. What legal regime should govern FBI or CIA agents' participating in Usenet discussions without disclosing their identities? Should they be permitted to post items on the Internet without identifying the government as the source? If the government is permitted such activities in the course of a law enforcement investigation, why should it be restricted when doing so for national security reasons? Aren't there additional First Amendment concerns especially about allowing the government anonymously and deceptively to engage in political discussions, for example with U.S.-based solidarity groups for foreign political organizations?

5. Do we need new legal protections to protect the privacy of overseas Internet communications by Americans in light of the increased number of such communications, the growth in technological surveillance capabilities, and increased cooperation and sharing between governments of surveillance intercepts?

* There is no statutory regime regulating electronic surveillance of Americans' communications originating overseas.

* National Security Agency regulations drafted in the late 1970's have not been revised since then and explicitly contemplate that Americans' overseas communications will be captured only in rare instances and for foreign intelligence purposes.(16)

6. Important questions also exist regarding Internet communications by non-Americans. For example, what are the implications for Internet privacy of the holding in United States v. Verdugo-Urquidez, 494 U.S. 259 (1990), that the Fourth Amendment does not protect foreigners overseas against U.S. government searches and seizures even when conducted to gather evidence to be used against the foreigner in a U.S. court? What are the implications of the current U.S. government position that it may eavesdrop on any overseas Internet communications by non-Americans without violating any U.S. law? As such communications increase, will we see wholesale introduction of intercepts obtained by the United States without a warrant or even probable cause as evidence in U.S. courts?

7. What legal regime should govern agreements between governments to share intelligence information consisting of Internet intercepts? While Mutual Legal Assistance treaties spell out the circumstances for formal assistance in the gathering of evidence, intelligence-sharing arrangements between spy agencies are cloaked in secrecy and go virtually unregulated.

8. What will be the effect on the development of international law for the United States to advance the position that it is unrestrained in eavesdropping on Internet communications by non-Americans? And if the development of international law is hindered, won't that have a detrimental effect on U.S. national security?

International Law and Development of the Rule of Law in Other Countries.

When United States national security objectives include the promotion of human rights and the rule of law, the development of both international law and national laws respecting human rights become matters of national security concern. In this context, the Internet will play an important role in influencing the development of international law and of national laws on national security, free expression and privacy.

First, other countries increasingly look to legal approaches to the Internet taken in the United States as a model for their own policies. U.S. government regulation of the Internet, or lack thereof, serves as a vivid demonstration to the world of what one democratic constitutional regime of surveillance and free speech looks like. This regime includes tolerance of extremist speech, using the Internet to counter disinformation by more information, and respecting the privacy of individual communications. At the same time, there is no doubt but that restrictive U.S. actions regarding the Internet will be looked to around the world in order to justify such restrictions in other countries. For example, Ukraine, which is barely emerging from the shadow of totalitarianism and recognized to be of strategic importance to the United States, has recently decreed that Internet connections by all state institutions must be arranged through the three state-owned servers. While the precise scope of this arrangement is unclear, many more Internet users are affected by this requirement than would be the case in other countries because many institutions in Ukraine, such as universities, are still defined as state institutions. Human rights activists in Ukraine are quite concerned that the intent and effect of this edict is to allow censorship of Internet content, both entering and leaving the country, and interception of individual messages. This is a concrete example of how the national security interests of the United States in promoting democracy and stability in Ukraine may be affected by developments on the Internet.

Second, the Internet makes information about U.S. legal models -- including information about restrictions on individual liberties approved in the name of national security -- much more easily, inexpensively and widely available. Should we be concerned about exporting our models without a full consideration of how those models will work in countries without the historical experience of and a legal culture based on respect for individual rights?

Finally, by transcending national borders, the very nature of the Internet seems to call for a multi-national, rather than unilateral, approach to many of these issues. Martin Bangemann of the European Commission recently called for an international charter regulating the Internet rather than country-by-country laws.(17) Whether such a charter will be developed and what aspects of the Internet would be covered are all questions of paramount importance for worldwide human rights and civil liberties. What restrictions on human rights principles for national security reasons, if any, should be included in any such charter?

If such a charter, or even some multi-national agreement between some countries on some issues is signed, what effect in turn will it have on the development of national laws in emerging democracies, on democratization of other countries, and thus, on long-term U.S. national security interests?

Such issues have already arisen, for example, regarding variations from country to country in free speech protections. While all international human rights treaties recognize the right of free expression, even among established democracies generally respectful of individual rights, there exist wide variations as to when such rights may be restricted, particularly for national security reasons.

For example, Germany outlaws pro-Nazi speech, but that same speech is entitled to protection in the United States. How might the Internet alter what has been, to date, the peaceful co-existence of these fundamentally incompatible views?(18) Specifically, should the rules most protective of free speech and the right of access to information be applicable to the Internet world-wide? What would be the ramifications of that? Isn't it right that "[b]ecause the Internet knows no national boundaries, on-line censorship laws, in addition to trampling on the free expression rights of a nation's own citizens, threaten to chill expression globally and to impede the development of the Global Information infrastructure ... before it becomes a truly global phenomenon"?(19)

CIA Covert Actions on the Internet.

Another area insufficiently studied has been the potential for the Internet to be used by the United States government as an instrument of covert action abroad. Such covert action, activities carried out by the CIA or other intelligence agencies designed to influence events abroad without the role of the U.S. becoming known, have a long and sorry history of ending in bloodshed and injustice overseas and in lies and crimes by U.S. government officials at home. They have ranged from the CIA-sponsored Bay of Pigs invasion of Cuba in 1963 to more low-key efforts to spread propaganda overseas in order to influence foreign elections or other political events.

The Internet offers enormous possibilities for covert action and does so cheaply and perhaps with much less risk of detection than has been the case with more traditional methods. While there has been almost no public discussion of such possibilities, they have not gone unnoticed by the professionals, who have suggested, for example, that the Internet offers an easy way to spread propaganda.(20) Even more dangerously, it has been suggested that the Internet could be employed to conduct "psychological operations" whereby for example, false communications are posted on the Internet to confuse or deceive an adversary. The implications of the Internet's carrying false information planted by the U.S. government in order to covertly affect events in other countries are harrowing to contemplate. While it is beyond the scope of this paper to outline the dangers of covert actions in general, conducting covert actions on the Internet could have disastrous consequences on its usefulness in building international trust and understanding. Public examination and debate is crucial before any such venture is undertaken.

Sociological-Political Questions.

Millions of Americans spend hours daily "surfing the Web"—gathering information, talking to others in real-time conversations (known on the Web as "Internet Relay Chat" or IRC), or posting their own information and reading that of others on Usenet and other information bulletin boards. Some of these discussion groups are dedicated to discussing alleged government cover-ups, including knowledge of unidentified flying objects and conspiracies in assassinations. Other sites—from Matt Drudge's gossip sheet to thousands of personal homepages around the world—show that everyone can now be their own publishers and instantaneously relay their messages to the world, be they ones of paranoia, information, or politics.

The role of the Internet as an unfiltered megaphone raises important sociocultural and political questions. Is the Internet fueling the historical strain of paranoia in the United States? More than ever in the past, conspiracy theories can circle the globe in days and reach millions of individuals: Will the enormous and quick circulation of conspiracy theories—and false facts—lead to pressures to restrict such circulation, in the way the debate over indecency on the Net blossomed? (In Mexico, for example, the government introduced controls on disseminating information over the Net, as part of its efforts to quash the Chiapas rebellion.)(21) Alternatively, might the Internet become a viable medium for combating these trends?

The U.S. government has designated home-grown extremist violence as a threat to the national security. Can the Internet play a helpful role in persuading Americans to become politically involved other than through acts of violence? Alternatively, can it stir up socio-political angst, leading to more violence? Put differently, is there a connective mob mentality?

Finally, are there ways actively to use the Internet to promote civil liberties the United States, by promoting access to information and decreasing the appeal of terrorist violence?

1. "The Internet and National Security: Emerging Issues," David Halperin.

2. President Milosevic eventually stopped jamming the signal but only after Western governments pressured him to do so.

3. "The Internet and National Security: Emerging Issues," supra at 27.

4. Critical Foundations: Protecting America's Infrastructures, the Report of the President's Commission on Critical Infrastructure Protection, Oct. 1997. [hereinafter "Infrastructure Report"].

5. A recent example of this is found in newspaper reports suggesting that hacker intrusions into computers at the Department of Defense at the height of the military build-up in the Iraqi crisis might be the work of foreign powers when it now appears that tenth graders in California were responsible. "11 U.S. Military Computer Systems Breached by Hackers this Month" Washington Post, Feb. 26, 1998; "Two Calif. Teens Suspected of Breaking Into Government Computers" Washington Post, Feb. 28, 1998.

6. See amendment proposed by Sen. Diane Feinstein, S. 735, 104th Cong. § 901 (1995).

7. See, e.g., Cass R. Sunstein, Is Violent Speech a Right?, 22 Am. Prospect 34, 36 (1995); "Report on the availability of bombmaking information, the extent to which its dissemination is controlled by federal law, and the extent to which such dissemination may be subject to regulation consistent with the First Amendment to the United States Constitution," prepared by the United States Department of Justice and submitted to the U.S. Congress, April 1997.

8. Infrastructure Report, supra.

9. The government and industry may well argue that only wholesale exclusion of such information from the FOIA will adequately assure companies of confidentiality and that, without such assurance, they will be unwilling to enter into such public-private partnerships. This argument is analogous to the argument made about information obtained from foreign governments, that without an absolute assurance of perpetual confidentiality, foreign governments will withhold key information from U.S. officials. But this is demonstrably not true because foreign governments have many reasons and incentives to share information with the United States even when they cannot be sure that their requests for confidentiality will always be honored.

10. These privacy issues include not only the well-known encryption controversy, but more specific issues about what legal process should be required for the government to obtain access to many different kinds of electronic communications and information. They also include digital telephony issues regarding what requirements the government may impose on the building of communications networks as well as issues regarding legal protections for the privacy of overseas communications. They do not however, include privacy issues relating to private individuals or corporations obtaining access to personal information on the Internet because the absence of government action fails to trigger Fourth Amendment concerns.

11. Indeed, the Fourth Amendment's reference to "papers" rather than communications may well reflect this technological incapacity.

12. This is not to suggest that somehow the violation of one individual's right to privacy is of less importance than the violations of many, only that the potential number and scope of such violations was limited.

13. Perhaps the most successful Fourth Amendment challenge to the government's key escrow schemes has been the argument that requiring the deposit of encryption keys with parties who act as the government's agents is a general search and seizure, which admittedly lacks probable cause or specificity. See, e.g., testimony of Prof. Kathleen M. Sullivan, on behalf of Americans for Computer Privacy, before the Senate Judiciary Subcommittee on the Constitution, Federalism and Property Rights, March 17, 1998. See also, testimony of Prof. Richard A. Epstein, also on behalf of Americans for Computer Privacy, from same date and same subcommittee. Both of these are available on-line at <>.

14. One example of this is the current controversy over whether and when the FBI should have the right to access location-identifying information from cellular phones. Another controversy involves the FBI's continuing access to conference calls after the target of the eavesdropping has left the call.

15. E.g., Privacy Act, 5 U.S.C. sec. 552a(e)(7); see J. Roderick MacArthur Foundation v. FBI, 102 F.3d 600 (D.C. Cir. 1996); cert. denied sub nom. Lindblom v. FBI, 118 S. Ct. 296 (1997).

16. The 1996 amendment to the National Security Act, authorizing intelligence agencies to collect information overseas for law enforcement (rather than for foreign intelligence) purposes may also require revision of these regulations.

17. "A New World Order for Global Communication: The Need for an International Charter," Speech by Dr. Martin Bangemann, September 8, 1997. This speech is available on-line at <>.

18. In fact, these thorny intellectual problems about the continued side-by-side existence of different national regimes of speech protection are more than purely theoretical: Germany recently prosecuted individuals who posted electronic messages to the Internet from outside of Germany; those messages violated the free speech laws in Germany, but were posted from elsewhere and thus protected under the First Amendment.

19. "Silencing the Net: The Threat to Freedom of Expression On-line," A Human Rights Watch Report, May 1996, Vol. 8, No. 2, at 2. [hereinafter "Silencing the Net"]. This publication is not on-line, but a brief description of it may be found at <>.

20. Strategic Assessment: The Internet, paper prepared by Charles Swett, Assistant for Strategic Assessment, Office of the Assistant Secretary of Defense for Special Operations and Low-Intensity Conflict (Policy Planning), July 17, 1995. This report can be found on-line at <>.

21. "Asia-Communication: Bumps Lie Ahead in Information Superhighway," Inter Press Service, December 14, 1995, as quoted in Silencing the Net.

Kate Martin, a civil liberties lawyer, is director of the Center for National Security Studies, a Washington, D.C.-based research and advocacy group working to ensure that civil liberties and human rights are not eroded in the name of national security. This paper was written with the support of the Harvard Law School Center for Internet & Society.