A Constitution without Constitutionalism

“Digital constitutionalism” has attracted a good deal of scholarly attention in recent years, much of it enthusiastic, some more sceptical. Just what constitutionalism means, and how this meaning can be transposed into a realm of private ― albeit increasingly regulated ― interactions rather than traditional public law, is part of the debate between the enthusiasts and the sceptics. All agree, however, that it is a normatively charged idea, a shorthand reference to certain values which include ― whether or not they are limited to ― respect for certain human rights.

In this post, I argue, drawing on a recent article, that while we can indeed think of internet regulation in constitutional terms, we must first understand what I shall call the constitution of cyberspace. A descriptive effort must precede any normative projects directed at imposing values allegedly inherent in the notion of constitutionalism onto cyberspace. And further, understanding the constitution of cyberspace should at least make us wary of digital constitutionalism’s normative ambitions.

Digital constitutionalism and its discontents

To briefly summarise an already substantial and growing body of scholarship, the proponents of digital constitutionalism, such as Nicolas Suzor, Claudia Padovani and Mauro Santoniello, Giovanni de Gregorio, hope that injecting constitutional principles into the regulation of the internet and especially the design of online platforms will make them more hospitable to human flourishing. These scholars are concerned that, left to the devices of self-regulation through the media of electronic “code” and contract law, cyberspace will a realm where private actors exercise power unredeemed by any commitments to human rights and dignity, and hence illegitimate. To tame online private power, they call for regulation subjecting it to constitutional principles, notably respect for human rights but, at least for Suzor, also the Rule of Law.

In a recent article, Róisín Á Costello has expressed considerable scepticism about the “constitutionalism” label. Her argument is not that the aims of the digital constitutionalism scholars are misguided, but rather that they do not amount to constitutionalism. This is because they pay no attention to “structural restraints” on power exercised in cyberspace and do not seek to make power accountable to those who are its objects, i.e. primarily the users of online platforms. In Costello’s telling, this makes it too easy for the governance of cyberspace to be described as constitutional and so to benefit from an undeserved perception of legitimacy.

Both the “digital constitutionalism” scholarship and Costello’s critique are valuable because they point us, respectively, to questions around the rights protected and the governance structures present in cyberspace. Both are, of course, typical constitutional concerns, and it is understandable that public lawyers should take a keen interest in the regulation of cyberspace. But interest is one thing; takeover would be another. We should not be too quick, indeed I think we have no particular right, to insist that cyberspace must take on our normative commitments.

Constitutions: another perspective

There is another way of thinking about constitutions that is helpful if we want to understand the regulation of cyberspace without warping this understanding through normative theorising. Its most extreme version is captured in J.A.G. Griffith’s notorious remark that in, the United Kingdom, “the constitution is no more  and no less than what happens. Everything that happens is constitutional. And if nothing happened that would be constitutional also”. However outré this may seem, especially to those of us who have been brought up in legal systems with legally entrenched constitutions, Griffith got at something important.

There are all kinds of constitutions, not all of them concerned with human rights, or structural limitations on public power, or accountability, or the public good. Statements such as the one found in the French Declaration of Rights of Man and Citizen, to the effect that “[a]ny society in which no provision is made for guaranteeing rights or for the separation of powers, has no Constitution” are vain both in sense of being futile and in that of being arrogant.

A private club or a corporation, for example, can be meaningfully said to have a constitution: that is to say, a framework of rules, although necessarily legally enforceable or permanent rules, about how its other rules change and who is involved in this process, as well as the limits, if any, to which these rules are subject and the values they serve. But such a constitution does not serve the relatively narrow ends recognised as constitutional by supporters and critics of “digital constitutionalism” alike.

The oddball constitution

In my view it makes sense to see cyberspace as having a constitution in the sense of a framework of rules about rules, even though it does not conform to a normative vision of constitutionalism. Admittedly, it is a peculiar constitution, in at least two ways that are worth highlighting. First, it is unsettled and subject to vigorous contestation by actors, state and non-state alike, trying to maximise their power and autonomy at the expense of competitors. Second, it is fragmented in a way that illustrates Heather Gerken’s concept of “federalism all the way down”. Neither of these traits is unique, but it is true that the constitution of cyberspace is an outlier on both these dimensions.

So far as the lack of settlement and stability is concerned, it is important to keep in mind that no constitution is perfectly ordered, let alone static. Even those that are (partially) codified in supreme laws meant to resist easy amendment are subject to change by the development of practice and convention. Of course, those constitutional courts that adhere to some version of a “living instrument doctrine” thereby undermine the stability of the constitutions they apply.

The constitution of cyberspace lacks institutional mechanisms such as amendment procedures that channel and defuse some of the instability of many real-world constitutions. But even state constitutions evolve outside of these formal channels. The difference is one of degree ― although it is a very large one ― rather than of kind. To the extent that Costello suggests that the instability of online governance arrangements prevents them from being regarded as constitutional, I believe she is mistaken.

But there is no gainsaying the reality that all key elements of the constitution of cyberspace are vigorously contested. This applies to what we might describe as structural questions, more or less of the sort that Costello rightly draws attention to. Thus: will the constitution of cyberspace be mainly the product of national or supra-national legislation, or will it emerge from private contractual arrangements and practices? And who will decide this? For example, what role, if any, will the users of online platforms play in their governance? Conversely, just how far can governments extend their regulatory ambitions ― can they, for instance, penalise platforms for behaviour that does (directly) impact those under their jurisdiction, as the EU recently threatened to do with Twitter?

Instability affects rights questions too. Most fundamentally, of course: what rights, if any, will the constitution of cyberspace protect? The focus of the “digital constitutionalism” literature is on freedom of expression and perhaps equality ― though the two are arguably in some tension. Suzor adds a concern with due process. But what about property rights, especially intellectual property of course, and freedom of contract? For better or for worse, these have been relegated to the status of an afterthought if not a nuisance by most public lawyers, but it is not obvious, to say the least, that they should be so in the realm of private interaction online.

Turning now to the fragmentation of and distribution of power in cyberspace, it too has some models, albeit imperfect ones, offline. Federal constitutions are of course familiar phenomena, and the idea of pushing the distribution of powers “down” ― and “up”, too ― is readily recognisable from discussions about the rule of cities or other local governments on the one hand, and entities such as the EU on the other.  The all-the-way-up-and-down federalism of cyberspace is, once again, different in degree, but a very considerable degree.

Cyberspace is not just one normative order but a collection of local communities, each of them with its own rules and values, enabled by (computer) “code” to organise themselves on the principles most congenial ― and/or profitable ― to those who control the code. This blog, for example, controls both its posts and the comments published on them, in the latter case explicitly referencing its owners’ “property rights” as a justification for a moderation policy that wouldn’t pass muster on any conceivable approach to freedom of expression. Other communities make very different choices.

And, public agonies over, say, Elon Musk’s idiosyncratic approach to running Twitter notwithstanding, the ability to exit and enter these communities freely is all that is needed to keep most people mostly happy most of the time. The costs of entry and exit tend to be radically lower online than “in real life” ― which, to be sure, does not mean that they are nil; they might even be unusually high in a few cases, calling for specific regulatory responses (though for my part I am sceptical about this). Whatever else we think about it, our consideration of the constitution of cyberspace is radically incomplete if it does not account for this, its most distinctive characteristic.

Getting normative, after all

The kinds of issues outlined above ― how rules get made and by whom, what rights, if any, are respected in a given community ― are typically constitutional issues. When they arise in the context of internet governance or regulation, it makes good sense to think of them in terms of a constitution of cyberspace. This constitution is unusual, to be sure, when compared with more familiar real-life examples. But it is not unrecognisable.

However, it is a mistake to proceed from this observation to the normative view that cyberspace not only has a constitution but also needs constitutionalism, at least if constitutionalism is understood as a more or less pre-determined package of liberal democratic values. To impose these values on cyberspace would destroy its most distinctive and arguably most valuable trait, the radical up-and-down-federalism that marks its structure. On this point, recent concerns about “constitutional scholars as constitutional actors” and “scholactivism” are directly relevant to the scholars of “digital constitutionalism”.