Building the Bottom Up From the Top Down

A. Michael Froomkin

University of Miami School of Law

Workshop on 'Governance, Regulations and Powers on the Internet (GRPI)'

Paris, France

May 27-28, 2005

Excerpted from Workshop Draft

I welcome your comments.


I.         Introduction

            In November, 2004, the British think tank Demos, an influential advisor to Tony Blair and the New Labor Party, issued a iconoclastic paper entitled The Pro-Am Revolution. Footnote Demos argues that advances in technology and culture increasingly are being driven by a new breed of serious hobbyist, "amateurs who work to professional standards" Footnote --thus the infelicitous "Pro-Am" of the title. The work of these Pro-Ams, Demos argues, increases 'cultural capital': "when Pro-Ams are networked together they can have a huge impact on politics and culture, economics and development. Pro-Ams can achieve things that until recently only large, professional organisations could achieve." Footnote It follows, argues Demos, that if this new class is valuable, it should be encouraged, even subsidized, by the British government.

            While I disagree with several of Demos's specific suggestions -- notably the idea of direct subsidies -- the underlying idea motivating the Pro-Am paper offers a healthy challenge: if we do generally agree that the public sphere, or civil society, or our life together, would be improved by having more self-organized groups and self-governing institutions, then how exactly does one get from here to there? More specifically, what role is there for government to play in encouraging the rise of new groups and new self-governing institutions? Does it even make any sense to think of building the bottom up from the top down, or is the role of academics and especially policy-makers limited to that of participant-observers and cheerleaders waiting for the public to spontaneously organize itself? This paper is an attempt to respond to that challenge.

            It would be nice to write to write a paper that contained a roadmap for the creation and nurturing of self-governance among spontaneously formed groups, a Habermasian community in a box. Unfortunately, this paper is, at best, a stepping stone in that direction. Part of the problem is that we know too much about what does not work, but too little about what does. But there is a more fundamental problem: before one can talk about the details of self-governance one must confront the dilemmas of self-organization. Unless groups are forming there is nothing to (self-) govern. This paper therefore concentrates on the first steps needed to get from where we are to where one might seek to be: the encouragement and nurturing of self-organized groups.

            Self-organization of groups, however, does not always mean self-governance in any important sense, and the growth in the variety and number of groups does not necessarily translate into anything with effects outside those groups. Civil society is a very broad tent; a book club may be rewarding to its members but it translates at best imperfectly to any greater form of social action. Nevertheless, the experience of participation in a type of self-organizing and self-regulating organization may incubate habits and practices that could find wider application even if the initial groups themselves remain insular and inner-directed, that "a multitude of subspheres of interlocking, cross-pollinating, discourses would provide an environment in which an informed citizenry could revitalize the public sphere as a whole." Footnote In the service of that objective, many important broader questions of self-governance, such as how the habits of self-organization are translated into more outer-directed social movements and practices, are of necessity left to another day. Footnote

            Good discourse is at the center of liberal democratic theory. Many modern democratic theorists, such as Jürgen Habermas, rely on the right sort of discourse as a fundamental justification for the legal or political system. Footnote Many other democratic theories treat discourse -- whether labeled political speech or the marketplace of ideas -- as an essential element of a healthy, functioning polity. Footnote If discourse is so central to democratic government, then finding ways to enable and improve the quantity and quality of the sorts of groups that nurture good discourse should be one of our very highest priorities. Footnote And, fortunately, the Internet and the cell phone seem to provide a platform for the deployment of new tools and practices that hold out the hope of overcoming some of the obstacles that tend to hold back group formation, endurance, and healthy self-governance.


II.        Groups Are Good -- Why Don't We Have More of Them?



III.       Enhancing Groups With Social Software



IV.      Solutions From the Top: How Government Can Help

            In a perfect world, self-organization of groups would be a spontaneous bottom-up phenomenon. There is indeed something ironic, even illogical, about wanting large national (or, in the case of the EU, supra-national) institutions to act to stimulate bottom-up group formation. It may be even more unreasonable to expect top-down control institutions such as governments to nurture self-governing institutions that may in time grow to become rival power centers. Economics, and economists, teach us that we live in a world that is far from perfect. Citizens in liberal democracies traditionally look to their governments to serve ameliorative functions. Other institutions, such as churches, charities and other NGOs also play important roles in fixing market failure, and failures caused by the markets. Thus, despite the irony, it seems plausible to ask what governments and other large established institutions can do to induce more groups to form and to help them engage in workable self-governance. The Internet makes this challenge more interesting, and perhaps more feasible, because it presents a plethora of new opportunities for self-organization of groups and deliberation. Governments should seize the opportunity.

            Demos's report recommended very targeted subsidies:


The government should launch a ... fellowship programme, investing small sums in community [leaders]. This might be modelled on localised versions of the National Endowment for Science Technology and the Arts, which provide fellowships for innovators, and be funded by the Big Lottery Fund." Footnote

Demos proposed this not only to encourage and reward group formation, but because it believed that encouraging Pro-Ams is good for the economy. Footnote

            Direct subsidies may make sense as a stimulant to the production of industrial capital by infant industries, but it is a bad policy as regards the production of social capital. Although self-organized groups are not necessarily political, some will be; others may in time take on a political tinge as the members find themselves already organized should they become politicized. Still other groups may never be overtly political, but may nonetheless have political side-effects by their very existence as training grounds in communal relations. Given these roles, it is unwise to encourage governments to pick winners and create losers. Doing so invites governments into considering the content of projects. In the US at least, it is now widely agreed that this is not an appropriate role for government--a sea change from thirty years ago when the question was hotly debated. Footnote

            As Clay Shirky says, "Social Software encodes political bargains." Footnote Software helps make difficult things easy, but too often at the cost of defining roles. Thus, for the same sorts of reasons that it would be undesirable for governments to subsidize some community leaders, it would be preferable if governments did not try to pick a narrow group winners among the types of software that structure conversations or helped groups reach decisions or govern themselves. Given the diversity of organizations and objectives, software that defines a "chair" or a "board" or even "moderators" would be a Procrustean solution for too many groups.

            The government's role should be facilitative yet entirely content-neutral. Even ostensibly non-political rules such as one that limited subsidies to non-political activities should be avoided. Human time and energy is limited. thus, even if one could craft a program that had no class-based discrimination, any rule subsidizing gardening but not community organizing would inevitably cause a shift of time and energy away from politics towards the subsidized activities. If, as Habermas persuasively argues, public engagement is already too weak then it makes no sense to discriminate against it.

            Thus, the state's ideal role is primarily in creating a climate in which groups can form, and resources that they can use to organize themselves, govern themselves, and achieve their aims. Given the speed at which communities such as Slashdot (with more than half a million members) and the so-called blogisphere are forming, much may be achievable without much in the way of direct state intervention. There are, nonetheless, some areas where government action would be helpful and appropriate.

            The state has a role in protecting the participants in online activities from lawsuits designed to chill their participation -- especially since it is law, a governmental action, which creates the danger in the first place. SLAPP suits -- Strategic Lawsuits Against Public Participation Footnote -- are a danger to any group that engages a public issue in writing in a form that, like many virtual communities, can be ready by anyone. So-called "cyber-SLAPP" suits target those who host allegedly libelous anonymous comments made by others. Footnote Several US states have anti-SLAPP laws Footnote which seek to protect public discussions against those who would shut them down because they are being criticized by shutting down SLAPP suits at an early stage in the proceeding. Jurisdictions that lack such protections, or which do not extend them to cyber-SLAPP cases should be encouraged to offer legal safe harbors, at least for public criticism of politicians and corporations and public figures if not necessarily for attacks on private citizens.

            Second, the state has an active role to play at least in policing markets for communications technologies. Anti-trust vigilance is essential to ensure that no software or hardware maker can exert control over citizens' means of communications. If group-formation and self-governance is going to be enhanced by access to information and to communicative tools, then attention needs to be paid to the nature of the markets for those products. Unfortunately, there are powerful tendencies in both software markets and even hardware towards winner-take-all outcomes that can create monopolies over tools needed for communication. Footnote The government can, both by regulatory and purchasing choices ensure that it does not encourage these tendencies. Similarly, the state should encourage the production of information in the open source and creative commons Footnote models rather than continue to strengthen intellectual property rights. Footnote Encouraging people to put their works wholly or partly into the public domain enriches discourse. Too much intellectual property protection can stifle discourse. (Arguably, too little intellectual property protection is also harmful, but we are nowhere near that point at present.)

            Third, the state has an essential role to play as provider of reliable data, and as honest archivist. The Library of Congress, a digital National Archive, and every bureau of statistics are critical support institutions for an engaged citizenry. To be relevant and useful, political discourse needs facts about what governments are doing, what they have done, and as much reliable information as possible about the rest of the world. The government can serve as honest purveyor of raw data, and as reliable archive of information -- an especially important role given the ease of digital forgery. Furthermore, the government's ability to set reporting standards takes on added importance in this connection because a common data format enables a world of easy comparisons that would otherwise be much more difficult. Food labeling regulations that require packed foods to report ingredients, fat, calories, and other dietary information in a standard format provides a model of this type of standardization. Footnote By setting standards for the presentation of data, and in some cases collecting and reporting reliable data, the state can set the stage for informed debate on what the data tell us, rather than on what the data might be.

            One problem, however, is that the temptation for governments to become propagandists seems almost irresistible. In the US, the government has been caught producing fake news videos. Footnote In Europe, the European Commission has not sunk quite that low, but its current campaign to 'educate' citizens as to the benefits of the European constitution is certainly one-sided. Footnote Government has a role in providing information for the public, especially reliable statistical data. But the task of fairly setting out all the different possible sides is one that government will find hard to do; even if it is done well, it will be criticized by those who think their side is entitled to pride of place.

            Fourth, the state may have a role in helping people fight spam and other forms of discursive sabotage. Online communications in general continue to be burdened with high levels of spam; meanwhile, unless they are designed with community moderation in mind, virtual communities frequently find that they are vulnerable to other forms of discursive sabotage. Spam has now reached epidemic proportions, and is causing a significant fraction of users to rely less on email. According to a survey by the Pew Internet and American Life Project, 22% of email users say that spam has reduced their overall use of email, and 67% of email users say spam has made being online unpleasant or annoying. (These numbers are slightly down from 2004, when 29% said they used less e-mail and 77% said spam was making online life unpleasant. Footnote One does not know, however, if this means the problem is cumulating or ameliorating.) If the exponential growth of e-mail spam continues, email may no longer be a useful tool. Meanwhile, spammers have aimed their messages at blogs, wikis, slash servers and other community software. It would be disastrous to give governments the power to directly censor content, but governments may have a role in providing users with tools that they can use to defend themselves against spam.

            Fifth, rather than subsidize individual participants in groups, governments should seek to subsidize facilitative technologies. Perhaps the best way would be for governments to begin to use social software to involve citizens in decision making. Government demand for open-source decision-enabling tools could create a paying market that would stimulate open source software designers to improve their products.

            Sixth, governments should recognize that Meetup-style services are essential to group formation. If the industry moves towards a charging model, governments should investigate the elasticity of demand for meeting-creation services. If demand is elastic (is sensitive to price), then governments should offer the service for free, or encourage the provision of free private meetup services. Given the evidence noted above that there is a "virtuous circle between ICT use and more traditional neighbourly relations," Footnote governments might also wish to encourage the development of highly localized versions of Friendster that could be marketed as social icebreakers rather than as romance sites. Footnote

            Seventh, just as the state regulated labor relations by defining the role of unions in the centralized workplace, so too may the state have a role in creating new unions, or substitutes for unions, in the new decentralized workplace. A digital workers rights policy should include a component that encourages digital or even physical meetings at which workers could socialize, network, share skills -- and foster their solidarity. Realistically, developments of this sort are more likely in Europe, where there is a tradition of co-determination and worker participation, than in the US where most businesses are quite happy to be rid of their unions.

            Last, but not least, while it is would be impossible to craft a one-size-fits-all solution to the internal governance needs of what are currently an extraordinarily heterogenous variety of virtual fora, it is not too soon to try to craft solutions to specific and observable problems. Rather than try to craft model constitutions or bylaws for groups that will be as variable as the human imagination, running from lightweight mailing lists to complex virtual worlds, it would be better to offer services that designers of virtual spaces and participants in those spaces could avail themselves of when they need them.

            Perhaps it might be possible to design circuit breakers for familiar discursive pathologies. For example, one problem that seems to crop up time and again in fora as disparate as mailing lists and virtual worlds is the absence of a dispute settlement mechanism. Virtual communities often have very lightweight governance structures. Even those which are somewhat more thought-out are of necessity new, or old forms adapted to new circumstances. Inevitably unprecedented disputes occur, ambiguities are found, or the collective finds that it needs a way to prevent part of the group from out-shouting the rest. Mailing lists are notoriously vulnerable 'flame wars', or find that their community is being distracted by procedural wrangles, most often over what ground rules should govern posting rights to the list. Footnote Advising the community to migrate to a new list if it is unhappy with some of the participants in the old is not much of a solution if the troublemakers follow along, especially if they use pseudonyms.

            Disputes of these sorts are ill-suited for traditional terrestrial courts. The noncommercial issues at stake -- how a virtual community should be governed -- do not involve the type of issues that courts are likely to think they have jurisdiction over. Venue questions will be far more complex than a small-dollar-value matter justifies. Nor, for that matter, would many courts know much about how to resolve them. There is no traditional property at stake (and often no intellectual property as such), and to the extent that it would find "property" in the right to a list's name, it belongs to whoever runs the machine or owns the account that hosts the software. If the question is who owns the clubhouse for a club that meets in a virtual world, a court is likely to apply rules that will feel and in fact be arbitrary to many of the participants. If there are written, oral, or even implied contracts they will likely be silent on the issues at stake; in any case many disputes of this type may require the application of equity, or at least of a law sensitive to the evolving customs and folkways of the Internet generally, of the type of virtual community and software at issue, not to mention sensitivity to the legal and cultural expectations of participants who will in some cases be geographically concentrated but at other times will be internationally diverse.

            Disputes about the nature and practice of self-government of online communities call out for a new type of neutral and external referee -- a light-weight online arbitration, the cyber equivalent of a court, or perhaps rather of a small-claims court, but one empowered to primarily do equity rather than law. Footnote Quality arbitration costs, and the expense of bringing problems of this nature to a traditional arbitrator would be prohibitive for most virtual communities which, after all, may not have legal personality, a clearly defined membership, or a bank account. Just as states staff and subsidize their courts in the interest of providing a trusted and neutral means for citizens to resolve their disputes, so too might governments establish an Office of the Referee to arbitrate online disputes about the governance of virtual spaces.

            Like an arbitrator, an online Referee could offer binding opinions when the parties agreed in advance to be bound. Unlike a court, the Referee would have no power to compel participation -- a concept of dubious applicability in disputes where the parties may be international, or may have masked their identities behind digital persona. And unlike both arbitrators and courts, the Referee might also be allowed to educate participants and the public by giving advisory opinions when invoked by fewer than all the parties to a dispute. Like a court, the Referee could charge a small filing fee to discourage frivolous requests, but the state would both select the Referee and subsidize her operations. Footnote


V.        Coda

            Demos's The Pro-Am Revolution contains an implicit challenge: can governmental power be enlisted to aid decentralized self-organizing and self-governing groups? The difficulty, of course, is that existing governments are the institutions that in the long run have the most to lose from the growth of rivals who might claim to have greater legitimacy. In the short run, however, governments have much to gain, as do we all. Involvement and participation in all kinds of groups, including those with no political objective, can benefit both individuals and communities.

            Even if governments are willing to act, the economic obstacles to group formation remain formidable. There is also a danger that any governmental policy that sought to 'pick winners' would be smothering. A better policy would be for the government to create a legal climate in which groups can flourish both on and off line, and to provide generally facilitative, supportive, non-financial resources.

            Above I identify eight specific governmental policies that could usefully be adopted in any relatively wealthy liberal democracy to promote the formation of groups and assist them once they are formed:


          Enact legal reform (if not already in place) to prevent cyber-SLAPP lawsuits.

          Apply competition law aggressively to markets for communications technologies in order to ensure that no software or hardware maker can exert control over citizens' means of communication.

          Provide reliable data, and act as honest archivist.

          Assist those who desire aid (but only them) to fight spam and other forms of discursive sabotage.

          Subsidize facilitative technologies, such as group decision-making software by purchasing it for government use.

          Ensure that Meetup-like services are available at low (or no) cost (if demand for these key services proves to be elastic as to price).

          Enact a digital workers rights policy including a component that encourages digital or even physical meetings.

          Provide a corps of subsidized online neutrals to settle non-commercial disputes among members of virtual communities.

            Of course, there is only so much that communications technology can do alone, even with government assistance. Encouraging group formation will not by itself end poverty, raise standards of education, equalize gender roles, nor ensure a supply of reliable and affordable childcare. Good software can help make good relationships, yet cannot wholly transcend the problems that define the society in which it is deployed. But it's a start.