Governing Online Spaces: Virtual Representation
David R. Johnson, David G. Post, & Marc Rotenberg
”The introduction of this new principle of representative democracy has rendered useless almost everything written before on the structure of government . . .”
Thomas Jefferson, August 1816
Facebook recently terminated its commitment to hold a vote on all policy changes that received comments from thirty percent or more of users. In defense of the move, it explained that the system encouraged quantity rather than quality of comments; some defenders of the move also have pointed out that reaching the requisite minimum number, on a platform with almost a billion users, was impossible to achieve in any event.
We believe that this presents an opportunity to rethink the ways that meaningful participation by users in the development of policies that will govern large (and arguably essential) online social spaces can be achieved. In the online world, website policies, incorporated into their Terms of Service (TOS), “regulate” the activities of large numbers of people during increasingly substantial portions of their lives. In effect, TOS represent a new kind of law – an amalgam of principles borrowed from property law (and a service provider’s right to impose conditions on access to its servers), contract law (although TOS terms are not the result of negotiations or meaningful acceptance by users, and, indeed, most service providers reserve the right to change the terms that users supposedly accept at any time), tort law (although TOS-law doesn’t generally provide for compensation for any injuries), and criminal law (although TOS-law doesn’t provide for due process or impose external sanctions). Terms of Service govern not merely the relationship between individual users and the online service provider, but the relationships among users. They matter, and they will matter more and more as more and more of our time is spent in online spaces. The question is: who will make this new kind of law?
We believe in the principle of self-governance and self-determination: that all users have a right to participate in the processes through which the rules by which they will be bound are made. This principle is today widely accepted throughout the civilized world when applied to formal law-making processes, and we believe it applies with equal force to the new forms of TOS-based rule-making now emerging on the Net. Facebook’s privacy policies, its data-retention policies, its intellectual property policies, its rules regarding permissible content – all will have a far more substantial impact on the lives of many users than most of the more formal Law to which those users are subject. It is reasonable for users to demand that those policy-making procedures comport with this fundamental principle, and it is in the interests of service provides to find some reasonable mechanism to implement it.
Some will argue that users need not and should not be given any voice in the development of website policies because they can simply express their preferences through their ability to choose among competing sites. We certainly agree that the market can and does serve as a significant and largely effective check on the ability of website operators to impose arbitrary or unreasonable conditions on user activity. But markets are never perfect; in some contexts – Facebook being a prime, though not the only, example – switching is extremely costly for users, many of whom have invested substantial amounts of time and effort in organizing their own experience at the site and are understandably reluctant to have to repeat that investment elsewhere. Users are not exactly “trapped” on Facebook’s platform, but their ability to switch is severely constrained [see Google+], and to that extent the market is both an ineffective check on Facebook and an imperfect reflection of user preferences and values. Moreover, the positive network effects associated with social sites distorts the market for such sites even further, by making user exit to sites with smaller numbers of participants an even less viable alternative (and an even less effective check on the policies of the incumbent sites).
We propose that Facebook implement a system of virtual representation, whereby every Facebook user would be given the ability to grant a proxy to anyone who has volunteered to act on his/her behalf in policy discussions with Facebook management. These proxy grants could be made, revoked, or changed at any time, at the convenience of the user. Those seeking proxies would presumably announce their general views, proposals, platforms, and positions. Anyone receiving some minimum number of proxies would be entitled to participate in discussions with management — and their views would presumably carry more or less weight depending upon the number of users they could claim to represent.
This scheme serves the interests both of Facebook and of Facebook users, giving the former a far more reliable guide to user preferences and user values than it has under the current scheme, and giving the latter the opportunity to have their voices heard – indirectly, to be sure, but with appropriate weight, and potentially with great effect – in the development of Facebook policies that affect their experience at the site and their interactions with other users.
Virtual Representation Would be Good for Users
It is, we recognize, entirely unrealistic to expect large numbers of users, with busy lives and many other competing priorities, to delve deeply into arcane questions regarding the Terms of Service of increasingly complex social networking sites. It is equally absurd to imagine a sensible and truly deliberative discourse emerging out of the hundred million or so people who might actually care about any particular Facebook policy alternative placed before them.
But these are problems that we have confronted and solved before, through the 18th century invention of representation and representative democracy. As Alexander Hamilton and James Madison (writing as “Publius”) noted in The Federalist over 200 years ago, “representation of the people in the legislature by deputies of their own election” was one of the principles that was “now well understood” (i.e., in the late 1700s) but “either not known at all, or imperfectly known, to the ancients.” A method of designating individual representatives for large numbers of citizens was the only way, Publius (and others) realized, that democratic governance could scale over large territories and large populations like the newly-independent United States. It was, in Publius’ words again, a “powerful means by which the excellences of republican government may be retained and its imperfections lessened or avoided.”
Our proposal updates the principle of representation for the global network, and we believe that it constitutes an important step towards development of a mechanism for scaling up democratic principles and democratic institutions on a truly global basis. There is no longer any need for Facebook to make the difficult determination, UN-style, of how many representatives any particular interest group or geographic region “deserves”; let the users decide. Giving users the freedom to allocate, or re-allocate, their proxies at any time obviates the need to schedule synchronous “election days” when everyone has to find his/her way to a polling place. Unlike specially timed elections among competing candidates, a rolling proxy designation mechanism would allow non-activist users to be involved in (indirectly) shaping TOS policy at their own convenience. This would allow more moderate voices to be more fully reflected in the discussion. Extremists, who currently participate in Facebook policy forums on equal footing with all other users, will be (appropriately) marginalized if their views are unable to gain the support of large numbers of users. Ambitious would-be representatives would have an incentive to crystallize the issues, and to promote themselves as standing for particular points of view. Advocacy groups might find it in their interest to promote individual candidates for such a representative role, because showing up with a binder full of proxies would make it more likely that the company would listen to their suggestions. The need for those seeking proxy designations to reach out to the broadest possible audience would help to assure that all interests were represented and the extreme positions could be put into perspective. This mechanism would also allow representation of user viewpoints on a global basis; users could make their views known from any location, and designation of representatives could proceed from discussions in any language. Users around the globe could take some comfort in the realization that their diverse values would, to some appropriate extent, be taken into account.
Virtual Representation Would be Good for Facebook
Facebook, like most benevolent dictators, might have doubts about the wisdom of giving users the power to govern (even in part) their own affairs; it might fear that any mechanism for user involvement would give disproportionate power to extremists and activists pursuing their own parochial agendas; and it might fear that it would become somehow obligated to take actions that disserve its shareholders or even the bulk of users themselves. Facebook might argue that it already has adequate incentives to make its product as valuable as possible, and that its users are always free to migrate elsewhere.
But Facebook is both a product and a polity. While it has an ordinary producer-consumer relationship, mediated and constrained by its TOS, with each of its users, its power and its success derive from its ability to empower users to form relationships among themselves, relationships that are also mediated and constrained by those TOS rules. Making good policies for complex social networks requires thoughtful tradeoffs among many competing values; the only way to get a sense for how most of the affected users would make those tradeoffs is to ask them, or to ask some small group of active users who can be counted upon to reflect the perspectives of the larger group.
An online polity that is experiencing some success in governing its own affairs would be less likely to desert en masse for another platform. Really dumb mistakes, like sudden actions that defeat user expectations of privacy, would be less likely to occur and more quickly remediable when they do occur. Users, possessing a degree of participation in policy-formation procedures, couldn’t blame any rule they object to entirely on The Man.
Facebook’s compliance with the clearly expressed will of the online polity would also surely help to keep real-space regulators at bay .Facebook now confronts a need to comply with a complex array of local regulations, reflecting a diverse set of values articulated by local governments around the globe. Those won’t entirely go away. But the pressures to pass local regulations would diminish if Facebook itself could credibly claim to have a mechanism that produces TOS policies that reflect the views of its entire global polity. Many governments are happy to defer to self-regulation if their is some evidence that the regulated are satisfied with the results and are mostly making rules that govern themselves rather than pushing negative externalities and costs onto others. The TOS rules that guide user interactions primarily affect the way that users interact with one another. If users can be shown to be satisfied with the rules, and have a chance to influence them, it will be easier to persuade local governments to step back from aggressive efforts to shape and constrain Facebook products and services.
We are not (yet) suggesting that Facebook change its TOS to provide that it must adopt whatever policy a majority of representatives favor. Such a provision would, we recognize, pose a difficult conflict with the fiduciary obligation of the corporate directors to act in the interest of the company’s shareholders. But providing a mechanism to distill and discern the strong and concentrated views of large numbers of users, even by indirect means, is perfectly consistent with corporate self-interest, and does not pose any such conflict.
Adopting an innovative mechanism for virtual representation would place Facebook at the forefront of public spirited innovation in Internet governance. By affirming that its users constitute a true polity entitled to some measure of self-government, Facebook would gain the “high ground” in the governance debates, avoiding the need to constantly defend itself against the negative reactions engendered by its seemingly arbitrary changes in policy. Other important online sites might follow suit, perhaps even using mechanisms for proxy-identification and proxy-allocation provided by Facebook itself. Over time, Facebook might be viewed as a founding parent of global online democracy — quite an improvement over its current image as an unaccountable, all-too-powerful, not-so-loved, 18th century monarch.
Representative democracy is not perfect. But we believe it is the best available alternative. It would be relatively cheap to implement – indeed, it would likely be a relatively simple task for Facebook itself to make an app available to all users for the purpose of facilitating the designation of representatives and the allocation of proxies. The personal investment of time and effort by a small number of people who have incentives to become able to claim (accurately, for a change) that they “represent” the views of large numbers of users, would lead to more constructive discussions and better decisions on what the TOS ought to say and how it should be implemented.
Lots could go wrong, there are many unanswered questions, and implementation will undoubtedly prove to be more complicated than it might appear at the outset. But that has always been true when facing the challenge of creating new governance structures for new situations. The time for thinking more creatively about how to set the rules that regulate important online communities has arrived.