[This is the text of a paper presented at the Media in Transition Conference at MIT on October 8, 1999.]

Is the Internet a friend of democracy? The prevailing discourse says no, that the Internet is actually the end of democracy, and that democratic laws can no longer be enforced. This discourse is not only hostile to democracy, of course -- it is hostile to government as such, and it speaks of "government" in a way that makes no distinction between constitutional democracy and totalitarian fascism. This is the legacy of Friedrich Hayek, among others, extremist opponents of extremism for whom any amount of democracy, no matter how legitimate, inevitably leads to harder stuff.

Whatever their utility as political prescriptions, these philosophies have usefully directed attention to the complex and variegated institutional field through which the great bulk of any society is actually organized. Marx had no time for these institutions of civil society, which he regarded as epiphenomena of the essentially very simple structures through which a society and its citizens were defined (Keane 1988). But civil society is now exceedingly popular, in a striking way, throughout the world and across the political spectrum (Keane 1998), whether as a counterbalance to the overreaching of the state, as an integral constituent of democracy, or as the real and only substance of a free society.

Yet the libertarian commitment to civil society is unstable. Civil society, almost by definition, consists of intermediaries: organizations that orchestrate and subserve a wide variety of social relationships. But as Dominique Colas (1997) has observed, the concept of civil society did not enter European social thought as a liberal antidote to absolutism; its root meaning does not oppose it to the state. Rather, civil society was originally opposed to certain extreme forms of Protestantism that, in overthrowing the putative autocracy of the Church, also sought to destroy all intermediaries and all representations -- a mystical radicalism that sought to eliminate all obstacles to an unmediated communion with God. Political and technical ideas are routinely found to descend from secularized versions of medieval theology, and thus here the radicalism of modern libertarians echoes in some detail the origins of the concept of civil society -- not it supporters but its enemies: the smashers of idols, the extremist opponents of centralized authority, the militants seeking not to create their own intermediary institutions but to eliminate them altogether.

But it is not so. Nothing like that is happening. Intermediaries are changing, to be sure, multiplying and dividing, their functions rebundling into different configurations, but they are as necessary as ever. They are consolidating, indeed, increasing their geographic scope. States are not shrinking, and in fact they are compensating for the global reach of technology by creating a vast network of undemocratic and nontransparent global treaty organizations. Mediation and representation, with all of the good and evil that they imply, are the very essence of the age. Once we see this, we can see at last the real upshot of the technology, the real action that it has already set in motion. It is not the elimination of civil society, any more than of the state. It is, however, in both realms, the renegotiation of the working rules of every institution of society.

This conception of social institutions as sets of working rules that govern the roles and relationships of their participants belongs to John Commons (1934). Largely forgotten now, Commons was the mechanic philosopher of the New Deal. A printer, he eventually became a professor of economics and public administration at the University of Wisconsin, and in that position he trained many members of the generation that built the American welfare state. As the welfare state has come under ideological assault, Commons has been forgotten, mentioned only by a handful of legal theorists. All theories of institutions are largely compatible; they seem different on the surface because they all overgeneralize from the particular case with which the author is most familiar. Commons' theory started from his experience of the negotiation of work rules in printing shops through collective bargaining, and that was the paradigm that he brought to every institution he considered. He did not imagine that every set of rules arises through the same kind of formal mechanism by which union contracts are negotiated. He does not presuppose that organized associations of buyers and sellers will necessarily delegate representatives to negotiate over a long table the form contracts and other customary rules that govern a given industry at a given point in history. Nonetheless, Commons' project was to investigate the variety of mechanisms by which the stakeholder groups in a given institution do act collectively to carve out a space for their own customs and practices alongside and by compromise with those of everyone else.

As increasingly complex social relationships are mediated by networked information technology, we are becoming accustomed to the idea that the protocols of these mediated interactions -- the "code" in Larry Lessig's terms -- constitutes a set of working rules in very much the sense that Commons suggests. Computers, like institutions generally, both enable and constrain, and both computers and institutions are, in one important aspect anyway, discourses made material -- made, that is, into machinery that governs to some degree the lives of the people who use it. Even when they are not formally part of the government, and even when they have no legal force, institutions and computers both govern, and it is this much larger sense of governance that Commons views as the deep underlying unity of democratic government and democratic society. It is most unlikely, after all, that one can exist without the other, and if the Internet encourages a democratic society then it does so by promoting the diverse mechanisms of collective bargaining by which a democratic society orders its affairs.

The necessity for such mechanisms is clear. By providing a general mechanism for moving digital information and a general platform for constructing digital information utilities, the Internet provides new opportunities; it opens a vast new design space both for technology in the narrow sense and for the institutionalized social relationships within which the Internet is embedded. The Internet also necessitates a renegotiation of institutional rules in a more urgent way by destabilizing the balance of forces to which any successful negotiation gives form; by lending itself to the amplification of some forces and not others, the Internet undermines many of the institutionalized accommodations through which stakeholder groups with distinct interests and powers have gotten along.

Fortunately, what the Internet necessitates it also facilitates. If the working rules of universities will be remade through a negotiation between professors and students, among others; if the medical system will be remade through a negotiation between physicians, patients, and insurers, among others; if the political system will be remade through a negotiation among citizens and their representatives, among others; then the main impact of the Internet has been to provide tools that allow each of these stakeholder groups to associate and, each in their own way, to press their interests. Once again the paradigm of collective bargaining can mislead if it is taken too literally. The point is not that every social group forms its own union, or even necessarily its own organization, and the point is not that the Internet necessarily facilitates any kind of formal bargaining process. Collective bargaining can be mediated by a great diversity of institutional forms, and it is the genius of the Internet to be indifferent to the details of such things.

The Internet makes visible a layer of social process that is more fundamental than organizations, and just as fundamental as institutions, namely the customs by which people who have something in common think together. Before collective bargaining comes collective cognition, and collective cognition in its various modes is greatly facilitated by the various community-building mechanisms of the Internet. Ideologies can form in the networked community of computer programmers; news can spread in the networked community of nurses; experiences can be shared in the networked community of cancer patients; patterns can be noticed by the networked community of pilots; agendas can be compared by the networked community of environmental activists; ideas can be exchanged in the networked community of entrepreneurs; stories can be told within the networked community of parents; and so on.

The question, of course, is whether it does so enough -- whether the Internet provides the conditions for every social group, no matter how spread out, to take its rightful place at the table, to play its own role in renegotiating all of the social institutions in which it takes part. And the answer, just as clearly, is no. No technology is ever a sufficient condition for anything. It facilitates, but it doesn't do the job for us. To truly build a democratic society, it will be necessary to build new social forms -- new ideas, new movements, and new organizations that are adequate the opportunities and challenges of a networked world.

The role of political organizations must change. No longer must an organization carry the full burden of organizing the collective cognition of the social group that it claims to represent. This is good when it frees resources for other purposes, and it is bad when it reduces the binding force that makes membership in an organization attractive in the first place. It is good when it reduces the arbitrary power of the intermediaries through whom the information had flowed, and it is bad when it makes consensus-building and leadership impossible. What, then, is the role of an organization in a networked world? An organization can put people into complex situations like legislatures and standards bodies, where there is still no substitute for being there. It can conduct the research that requires pulling together more information than any individual could manage. It can maintain the relationships that make actual negotiations possible. And it can build the legitimacy that is required to call for a solidary action. These are all classical functions of an organization, and they will not go away. But they will all happen in a much more dynamic environment, and they will only work if they draw upon and encourage the power of collective cognition, rather than trying to channel it. This is hard, because it is much easier to deal with a centralized representative than a sprawling associative community. But it is the democratic way, and it is the principal hope today for a democratic society.

The answer cannot ride on the sort of bargaining that can be bought. Instead, it must ride on the massed creativity of a diverse people in diverse situations, all bringing their own experience to bear on the situations of others. If the Internet is a friend of democracy then democracy will be won principally on the ground, and the central task for democratic theory right now is to understand this ground, and to be useful to the innumerable people of good will who are out there trying to build on it.

References

Dominique Colas, Civil Society and Fanaticism: Conjoined Histories, translated by Amy Jacobs, Stanford: Stanford University Press, 1997.

John R. Commons, Institutional Economics: Its Place in Political Economy, Madison: University of Wisconsin Press, 1934.

John Keane, Civil Society: Old Images, New Visions, Stanford: Stanford University Press, 1998.