Debunking eight myths about multi-stakeholderism

From being the darling of civil society during WSIS (and soon after being repurposed by the technical community as their own guiding principle), multi-stakeholderism has entered a rough patch during the last 18 months. The phrase, and—worse—the underlying concept have quite swiftly begun to acquire a distinctly negative connotation in the public eye (at least, among those who have heard of multi-stakeholderism at all).  Amongst the civil society voices who are now treating multi-stakeholderism as a failed cause or worse, here are a few recent examples from commentators across a range of political views:

But these are not critiques of multi-stakeholder governance in general, or to the extent that some of them are, they are misplaced. Here I attempt to briefly debunk some of the fallacies that underlie blanket critiques of multi-stakeholder models of Internet governance:

  • That multi-stakeholderism aims to replace governments. Multi-stakeholder processes are for producing soft law principles, rather than hard law regulations. Admittedly there is more of a continuum between hard and soft rather than a clear dividing line, but it is clear enough that while multi-stakeholder processes may provide recommendations to the national and international institutions that create binding obligations, they won’t be replacing those institutions any time soon. Ultimately we can question whether nation states (and groupings of these) are still the right default agents for making the hard rules that govern our world, but the future in which they fade away in favour of multi-stakeholder governing bodies remains a long way off.
  • That multi-stakeholderism is incompatible with democracy. The “harder” that the outcomes of multi-stakeholder processes are (for example, some of ICANN’s policies have effects that would be difficult to distinguish from international legal rules), the more important it is to ensure that its structures and processes are legitimate. In this context, legitimacy—which, yes, means democratic legitimacy—doesn’t just come from representative democratic processes. There are other ways of reflecting the will of those who are governed than voting, which, anyway, is impossible to realise at the global level (as ICANN discovered with its 2000 At-Large election experiment). The method that ICANN has plumbed for is to create a structure aimed at balancing the inputs of all affected stakeholder groups, including those that cross national borders. Other methods, such as the deliberative poll that is (hopefully) to be held ahead of the 2015 IGF, utilise techniques of deliberative democracy with the aim of produce balanced and rational outcomes. Such methods, flawed as they may be, are no less democratic in principle than the equally or more flawed representative democracies that we live under—nor than international institutions such as the United Nations, which indeed suffer from very severe democratic deficits. Particular multi-stakeholder processes can and should be criticised for not being democratic enough, but it is false to claim that they are undemocratic by definition.
  • That multi-stakeholderism gives too much power to corporations. Do multi-stakeholder processes allow corporations to participate? In general, yes. But let’s compare apples with apples, here. Corporations already have too much power in global governance. Do multi-stakeholder processes give them any more power? No—or if they do, they’re doing it wrong. On the contrary, multi-stakeholderism is meant to be about balancing the input of all affected stakeholder groups, crucially including civil society, which is currently sidelined in the governance processes that corporations dominate. Thus although multi-stakeholderism does allow for the perspectives of corporations to be included in decision making, this remains relatively better for civil society, because our perspectives are also heard at the same level, which is an improvement over the status quo. By analogy, consider the reason the why free and open source software licences outlaw discrimination by field of endeavour (ie. prohibition of commercial use). Although FOSS licences are good for corporations, that is because they are good for everybody (and in fact better for those who are locked out of proprietary software). The same is true of multi-stakeholderism.
  • That multi-stakeholderism is a neoliberal construct. Again, if multi-stakeholderism is really a better form of governance for the borderless world we live in, then, like democracy itself, it ought to be better for libertarians, conservatives, liberals and progressives alike (to use the US terminology). Of course, it won’t ever satisfy radicals who are really calling for the complete overthrow of existing power structures. If you are not looking for a mechanism of global governance, but for a stick to beat corporations or governments with, then look elsewhere. That’s not governance, that’s advocacy (or, in the hands of the powerful, an instrument of class warfare). But whilst multi-stakeholderism isn’t going to completely upend existing power structures, neither will democracy. What you’re probably really looking for instead is a system of government that allows workers to seize control of the means of production.  I believe there’s a name for that.
  • That multi-stakeholderism requires corporations to have an equal role to governments. This has been debunked before, but it bears repeating that this argument is no longer tenable post-NETmundial, where it was established that “The respective roles and responsibilities of stakeholders should be interpreted in a flexible manner with reference to the issue under discussion”. As explained in my previous post  that means that depending on various factors it will be most legitimate and effective in some cases for governments to take the lead; in other cases, the technical community; in others still civil society; and in some cases industry.
  • It’s turtles all the way down. This fallacy tells that in order to sort out the roles and responsibilities of stakeholders in a given issue area or forum there needs to be a pre-existing (also presumably multi-stakeholder) process to determine these, and then another one to determine that one’s composition, and so on. Actually that problem mostly solves itself. Since multi-stakeholder processes are voluntary, if affected stakeholders are not able to negotiate for the roles and responsibilities that are appropriate to them, they can always opt out. Their refusal to participate will then reduce the legitimacy of the initiative, and consequently the authority of the outputs that it produces. This isn’t just theoretical; it can and does happen; the Licences for Europe initiative is one example.
  • That any process that calls itself multi-stakeholder really is.  Part of the reason for the backlash against multi-stakeholderism is that too many of the processes that use that label don’t deserve it. Not all multi-stakeholder processes were created equal. Many criticisms that have been made of the execution of particular supposedly multi-stakeholder processes are absolutely valid. Some such Internet policy processes are still legitimate in different terms—for example, they may be instances of state-led participatory democracy—but to call them multi-stakeholder, when they do not attempt to incorporate the viewpoints of all affected stakeholders into the development of those policies in a balanced way, is misleading. One recent example of this, in which civil society’s participation was both narrow and token, is the Global Conference on Cyber-Security  Processes that do better?  Regrettably these are harder to come by as yet. ICANN is flawed and largely captured and driven by special interests, and the IGF has squandered its opportunity to actually engage in governance rather than just discussing governance. However, both institutions continue to evolve, and we should not expect the multi-stakeholder model to be perfected within just a few years (how long did it take for the Westminster system to reach its present form?).  Even so, multi-stakeholder advocates are often criticised as not presenting a good example of the ideal that they propose, which brings us to the last fallacy: 
  • That there is any alternative. It really boils down to this: if not multi-stakeholderism, then what? Do we want to establish a new, global parliament to make global rules for the Internet? Or do we expect (and trust) existing intergovernmental bodies such as the ITU or the (thoroughly undemocratic) UN Security Council to do so? Do we have any faith that the rules that 196 nation states make for the Internet will interoperate in a sane way if there is no process to guide their development along compatible paths? Or that companies will develop terms of service in consultation with users from all over the world, taking into account all their local needs and circumstances? Or that technical standards are a magic wand to wave away all of the governance problems of the Internet? There is such folly in these scenarios that it makes proponents of the development of multi-stakeholder models of global Internet governance look like staid conservatives by comparison. There really is no alternative.

So if multi-stakeholderism is not the dead end that some claim it is, but rather an emergent model of governance whose name is often taken in vain, how are we to recognise (and to shape) better examples of it? What is the gold standard of multi-stakeholder processes to which we aspire? That is, indeed, something that needs to be better documented, and I hope to be able to work with others this year to produce an accessible “quality seal” of multi-stakeholder processes that we can use to help chart the path forward.