At the same time, new communication technologies empower citizens anywhere to broadcast charges that their own governments violate freedoms and human rights. The objective is to bring unified world pressure and public opinion to bear against repressive regimes unable to hide their misdeeds as successfully as before. That demonstrators in Tiananmen Square displayed signs written in English was not a coincidence. Cross-border communication in the defense of democracy and human rights is the activity on which citizen diplomacy groups like Amnesty International stake their success. The new technologies markedly enhance these capabilities.

As the "Dictator's Dilemma" formally suggests, governments that try to squelch the new information technologies to protect their monopoly on power do so at the peril of economic growth. It is the message that Shultz shared with Gorbachev and it is what leading analysts have begun predicting: "For nations to be economically competitive, they must allow individual citizens access to information networks and computer technology. In doing so, they cede significant control over economic, cultural, and eventually political events in their countries" (Builder, 1993: 160). A developing theory suggests that this reasonably ought to be so, that the new technologies, which combine for the first time both autonomy and influence in the same medium, couple decentralization of political power indivisibly with economic growth. While some communications media tend to assist authoritarian regimes in maintaining control, the characteristics of electronic networks are exceptional in their theoretic capability to sustain reciprocal communication on which democracy thrives.

Empirical evidence confirms the postulated correlation between democracy and network interconnectivity. Despite the inherent limitations of statistical analyses, several analytic perspectives, every model, set of statistical tests and functional form in this study is consistent with the hypothesis that interconnectivity is a powerful predictor of democracy, more than any of democracy's traditional correlates. In univariate analysis, the correlation coefficient for interconnectivity on democracy is larger than that of any other variable. As a variable in an ordinary least squares multiple linear regression, interconnectivity is exclusively the dominant predictor. As an interaction term in conjunction with regional categorical variables, the correlation of interconnectivity with democracy is everywhere positive, and has both the largest substantive value and greatest statistical significance in regions characterized by dynamic political transformations. Interconnectivity correlates strongly not only with the level of democracy, but also with the change of democracy over ten years. The highest statistical significance is apparent among countries labeled as "Not Free." While many of the least democratic states became more democratic with becoming interconnected, not a single one of them became even moderately interconnected without also becoming more free. Within cohorts of states of shared historical conditions, a progression of greater connectivity evidently associates with advanced democratic development. Tests of alternative causal explanations invariably fail. As an endogenous variable in systems of simultaneous equations, interconnectivity always proves to be a significant predictor of democracy and economic development, but never is the reverse true. And finally, among communication media, the positive correlation with democracy of electronic networks is unique, deflating the apparent importance of potentially confounding economic factors.

More specifically, despite the persistence of the unanswerable causal question, the consistent test results that identify interconnectivity as the most powerful predictor of democracy should impress the policy debate on global democratization for several fundamental reasons. First, to the extent that the United States and other Western democracies aim to encourage the development of democracy worldwide they do so primarily through programs to improve economic development, education, health, legal reform, etc. The causal connection supporting those programs is no stronger, and in most instances quite a bit weaker, than can be inferred in the case of networked communication technology.

Second, the backdrop behind the empirical results presented here is a crescendo of anecdotal and theoretical analyses that cast electronic communications networks to the front among factors that may contribute to the development of democracy. The statistical results reinforce the validity and credibility of more subjective analyses.

Third, the causal relationships in international affairs are rarely either simple or in one direction only. The most plausible relationship between democracy and networked communication (and perhaps economic development) is probably a virtuous circle with substantial positive feedback all around. Even if a causal argument positing an alternative explanation of the correlation, such as "greater democracy causes greater interconnectivity," could be established with completely irrefutability, the existence of such an argument does not negate or diminish the importance of a likely stream of causality in the opposite direction. Furthermore, the total weight of evidence makes denying the likelihood of the latter causal flow nearly untenable. In sum, if the United States seeks to influence international democratic development effectively and efficiently, for the purpose of its own national security interests, it ought not ignore the effects introduced by advanced capabilities to communicate and access information across borders.

Whether American and other democratic policymakers take advantage of it or not, they may able to exploit the "Dictator's Dilemma" to their own favor, for the dilemma posits choices not only for the autocratic ruler but also for the democratic strategist. While the former is forced to ponder what E can do to avoid being scissored by the dual blades of the global economy and personal media, the latter ought to be thinking about what E should do to sharpen the cutting edges and make sure they work well in unison. Several types of implements are available, even in a policy environment characterized by significant budget cuts and political constraints.

1. The United States government can identify an institutional home to delegate the responsibility for international interconnectivity policies. Until now, the primary federal organization that has concerned itself with the development of global networks, albeit specifically for collaboration in scientific research, has been the National Science Foundation. Current developments, however, lead in a discouraging direction. At the same time that world-wide information needs are ballooning, institutional support from the United States is deflating. NSF is phasing out its involvement because the "general purpose" (i.e. commercial) activity is assuming a dominant role over research (Goldstein, 1995).

   A cabinet department such as State or Commerce could assume and expand the limited tasks that the NSF is shedding and adapt them to the new international communications circumstances. Alternatively or additionally, a more narrowly focused mission-oriented agency like USAID, could pick up some of the slack specific to its own mandate. To this end, a statement by James Russo from the Internet Data Services Group of USAID at the NATO Networking Workshop in Moscow seemed promising, although as yet unresolved, "We are currently working on a policy to decide to what level we want to get involved into telecommunication" (North Atlantic Treaty Organization, 1995: 157).

2. The United States can support international networking by providing expertise, equipment or connectivity internationally. There are numerous substantial low-cost high-payoff opportunities. A small scale, possibly covert, example that captures the imagination is the "Japanese guy" who delivered unsolicited modems to Czechoslovakian student activists "trying to coordinate the uprising across the nation." The gift-bearer summarily disappeared but several months later, "By mid-December, the Civic Forum was in power" (Sterling, 1995: 102).

   On a grander scale with greater publicity, George Soros' International Science Foundation helped ignite a virtual explosion of networking in Russia that began by putting in place a direct 64 Kbps satellite link from Moscow to the United States in July 1993. This program is further expanding interconnectivity in Russia with the construction of a fiber "Moscow Backbone" (Shkarupin and Mafter, 1995). The initial ISF project budget was only $10 million in private funds, a small sum indeed in terms of funds dedicated to U.S. national security interests.

   Existing federal resources offer substantial interconnection possibilities. For example, U.S. Information Agency offices or Embassies around the world could serve as a net access point through which indigenous organizations in foreign countries could interconnect with the global information infrastructure.

3. The United States can unilaterally affect the dictator's decision and weight the scale in favor of interconnectivity and openness by augmenting the perceived and actual economic value of being interconnected. In this arena, domestic policy choices will have distant consequences. Policy options between those that could either stimulate or stifle commerce on the networks, debated within the national context, will reverberate in non-commercial ways in places far from the United States. Any autocratic conclusion and politically motivated decision to restrict information flows in China, for instance, become more economically painful to implement when the United States, its major trading partner, selects policies that support the use of electronic networks for trade and commerce.

   Generally, internal U.S. policy deliberations on issues nominally related to domestic networks are thus artificially and unfortunately constrained if not conceived most broadly in the international context. In fact, the word "national" as in "National Information Infrastructure" is at best a misnomer and at worse can mislead policy in spheres where political boundaries have no practical significance.

   Commerce friendly policy regarding networks is one means by which the United States can tip the economic freedom versus political control balance. Another is the demonstration effect by the world's strongest economy, which has already proven to be a powerful motivator. The greater the quantity and higher the quality of public information available on-line in America, the stronger the incentives will be everywhere to join the information revolution, both as recipients and as providers. With such projects as "Thomas," providing current legislative information on the Internet, to others such as making census data available online, the U.S. government has begun to use the networks to provide volumes of useful information aimed towards domestic audiences, thus setting a positive international example. A far-sighted policy would consider foreign audiences as well. Translations of useful documents and information available electronically could assist aspiring democrats overseas. International human rights information could be made available from the same sources as commercial data such that one could not be blocked without also, painfully, losing the other.

   There is also moral encouragement. When highly centralized countries, such as China and Saudi Arabia, take tentative steps in the direction of broader network connections, the United States should be liberal with laudatory motivation, if not material support for networking.

4. Similarly, the United States can unilaterally lower or eliminate barriers that prevent foreign access to useful information. Exemplifying the frustration that some foreign networkers encounter, "Iran's computer specialists worry as much about hard-liners in the U.S. Congress as they do about their own Islamic clergy. For some years, Iranian academics could not browse the library at, say, the University of Michigan, because the NSFNet was barred to them" (Bogert, 1995: 36).

   Export controls on encryption technologies provide another distressing example. In many countries of the world, those who may have a need to guard against state scrutiny are not likely to be the "bad guys" as one might argue is the case inside the United States. Instead, those seeking to protect themselves may well be the democratic opposition or human rights activists. Prohibitions on encryption technology can inhibit communication and increase peril for such would-be networkers in their home countries.

   Again, the demonstration effect is a potent force even when it provides an outstanding negative example. The Russian anti-encryption law, for instance, appears to have been inspired by aspects of the U.S. proposal. If both countries were to enforce such laws aggressively, it is perhaps an open question as to whether or not American democracy would be better off, yet it is almost certain that prospects for Russian democracy would unambiguously suffer. Thus, ultimately, both countries would be less secure.

5. The United States can more carefully align communication aid with the incentives of recipients. Traditionally, aid is provided through government-to-government programs, although foreign counterparts may not have the right motivations to move in the direction of openness that are mutually beneficial. The implication is to reconsider the appropriateness of certain foreign assistance recipients in target countries relative to their natural inclinations. The probable outcome will be a switch in focus from governments to citizen groups whose natural inclinations are more compatible with U.S. national interests. The trend that analysts have begun to observe is that "the audience for American foreign policy is changing; it is less other national governments and more their publics--who are watching and listening and to whom the power is shifting" (Builder and Bankes, 1990: 22).

   Therefore, best interconnectivity programs will probably not be of the sort that helps to advance connections through government ministries or PTTs, at least not exclusively. The central players' instincts are biased toward monopolizing power and control. A preferable strategy would target numerous private groups, schools, libraries and other non-governmental organizations. The variety of programs through which the United States supports small business development may provide useful templates for programs to seed the proliferation of electronic networks. Important policy choices will consider tradeoffs between supporting indigenous foreign NGOs or trans-national NGOs linking with local partners.

6. The United States should elevate the attention to and priority of information and communication programs in planning and implementation throughout foreign policy, from economic assistance in which the Maitland Commission found telecommunications to be a legitimate and critical element, to national security. Proponents have cogently argued that "information should now be considered and developed as a distinct fourth dimension of national power--an element in its own right, but still one that, like the political, economic and military dimensions, functions synergistically to improve the value and effects of the others" (Arquilla and Ronfeldt, 1995: 31). Whereas promoting democracy globally has been identified as a primary national security interest that the United States pursues with the tools of the three traditional dimensions and whereas this security interest is powerfully correlated with the fourth, an information dimension, the last dimension ought rightly to be included among the others as a recognized instrument for the exercise of national power.

On the other hand, objections may be raised against the implementation of international communication policies that arguably reflect "cultural imperialism." These are serious concerns for which there are several salient responses. First, the United States regularly attempts to influence the nature of foreign regimes with policies that range from economic assistance to troop deployments. Many of these interventions are far more invasive than simply encouraging interconnectivity. Military operations are an obvious trespass but less ostensibly imposing interventions can still be viewed as quite threatening by the ruling elite. One such example would be the National Endowment for Democracy programs that assist opposition groups to create credible political alternatives in countries without them. Even previous generation communication technologies such as state radio and television broadcasts assailing foreign borders ought to be considered more intrusive. President Clinton's Radio Free Asia and VOA television proposal to "tell freedom's story to the people of China" is more offensive than electronic computer networks yet it is less likely to be successful. Broadcasts provide no opportunity other than passive reception subject only to the interests of the propagandist. Networks, much to the contrary, empower foreign participants to actively pursue their own communication objectives both as information producers and consumers. For the same reason, inherent "cultural imperialism," Clinton's plan is doomed to be less auspicious. Conversely, the power of the personalized media in energizing political change in Eastern Europe was augmented precisely because "It was indigenous. It did not bear the stigma of originating with the American government" (Rosenstiel, 1990: A15).

Second, with due respect to Marshall Mcluhan who said "the medium is the message," this essay has specifically addressed the former but not the latter. The research and policy conclusions focus on the deliberate spread of a technology not an ideology. Democracy is a dependent output of theoretical and statistical models, not the centerpiece of a proselytization campaign. Both the theory and the statistics rest solely on the measured ability of people to communicate efficiently and effectively. The essence of the policy recommendations is merely to assist in establishing the technological and social infrastructure for robust international communications. In doing so, one can expect democratic outcomes even when diverse people everywhere are using this infrastructure for their own chosen ends.

Third, "revolutions do not occur until people learn that there is an alternative to their way of life" (Wriston, 1993: 121). It is essential to consider who might raise the charge of cultural imperialism, the dictator or Es public? Whom would the United States most want to benefit? Whose interests are closest to those of the U.S.? When the Tienanmen Square demonstrators chose Lady Liberty as their symbol for universal human rights it was obvious that cultural imperialism was not the source of their grievance or oppression.

Fourth and finally, the implementation of policies advocated here impose nothing on anyone, not even democratic governance. More subtly, the policy goal would be to help create conditions that, were a people to choose a path toward democracy, the available means of communication would assist rather than hinder their progress.

Clearly, a hopeful metaphysical premise under-girds these arguments leading to the conclusion that inexorably the very tools required to build a strong economy will also be used to topple political autocracy. While there are competing beliefs, this premise holds that freedom and democracy are, indeed, universal values. Empirical evidence seems to support this optimism thereby reinforcing the policy recommendations. Although not everyone subscribes to this vision, Secretary Shultz was prescient in his observation regarding the "Dictator's Dilemma" and he may well be right also when he proclaimed, "History is on freedom's side" (1985: 721).