Post 4 in the Guest Blog Series Preserving Public Values in The Automated State by Francisco de Abreu Duarte

We all know the famous duck test scenario. Although it often leads to absurd results, as Monty Python superbly demonstrated, it is a reasonable and very human reasoning to make: if something acts, sounds, and looks like something we know, chances are that they are the same thing. Even if they are not, they should likely be treated the same. Now, what has this to do with Law & Technology and, more precisely, with digital public values? 

Law & Tech scholarship is recently undergoing what I would call a ‘Public Law Duck Test’. This is most evident in an increasing movement I am part of called Digital Constitutionalism. Although there is so much more than meets the eye, people tend to read our work as reflecting a new application of old-style constitutionalism (meaning a new version of Liberal Constitutionalism), and we have been severely criticized for this. In these critics’ eyes, we were doing precisely what I claim in this post: we see public law everywhere. However, Digital Constitutionalism is far from the only attempt to look at Law & Tech from a Public Law perspective. Some scholars now propose direct comparisons with Platforms’ Administrative Law, while others describe their power as ‘kingly prerogatives’, calling them ‘emergent transnational sovereigns’ or States. Others prefer the term special-purpose sovereigns, custodians, governors, privately owned bureaucracies or data oligarchs. Habermas himself has called these companies ‘semi-private, semi-public communication spaces’, and a quick look at the DSA seems to confirm the same spirit, with classic administrative due process being the core of the regulatory effort.

So the question is rather simple: why do we compare these new realities to public law constructions?

Hierarchy and Power

One of the dimensions that seems to welcome this Public Law reasoning is the idea of hierarchy and untamed power. In all of the comparisons above, the problem that worries scholars the most is the unequal standing between platforms and users. Prima facie, the former seem to possess the characteristics that should worry public law scholars: the power to censor us, to exclude us, to select what is relevant for us, to downgrade us, and, in some cases, to addict us. It is not only that there is a David – Golias relationship, typical of Administration-Subject, but also that Golias seems to have the discretion to do as he pleases. It looks very much like 18th-century problems. 

However, this argument raises some doubts, typical of the duck test fallacy.

First, where do we draw the ‘Golias’ line? Is it any large entity that deserves to metamorphize into a public law entity? Does it have to do with its size, leading us to choose some arbitrary number, such as 45 million monthly active users? If that is the case, then should that suffice to say the entity has transformed its private nature (which, ironically perhaps, liberal constitutionalism so desperately tried to protect) now to be bound to similar obligations as an administrative agency or a government? Is that what we mean by enforcing ‘digital public values’?

In addition, one company can be big and abusive and still not considered worthy of this ‘publicness’. If Coca-Cola decides that Portuguese are unworthy of its soda, and decides to stop selling in that country, can I personally ask for a justification? Do I have a right to challenge that decision in a specialized body for ‘beverage disputes’? Likewise, one can be small and irrelevant and be very abusive. If I create a chess club with two colleagues and decide that only great dancers should be admitted, do I have a duty to explain my reasoning and have an extensive appeal mechanism? In both cases, the answer would be likely no.

So, size cannot be the sole answer.

Market Power/Choice

Perhaps the answer lies less in size and hierarchy but more in choice. One could wonder: had there been different ‘States’ in the same given community, and would liberal constitutionalism ever be necessary? To some extent, despotic public power is problematic precisely because there is no exit. There is no other government, a different set of courts, or multiple administrations. Perhaps the DSA and the DMA are sisters because our public law duck test inherently depends on the inexistence of market choice. The follow-up question is similar to the previous one: how much market power is then needed to trigger publicness? Do they need to be a monopoly? Do they need to be digital gatekeepers in the sense of the DMA? Does market concentrated power always necessarily lead to despots, especially in highly innovative markets such as technology? 

This argument also raises some doubts. There are hundreds of dominant companies in the world, some abusive, for which we would never think of creating a due process scheme when they make decisions. Competition law remedies, perhaps until the DMA, were considered specific market corrections to protect consumers via protecting fair competition: it was not the consumer that would challenge a particular decision to go through a procedure to tame the company’s power, and in the EU, she could not even do it if she tried. It would be a competition authority/the EC that would do that for you, ultimately considering that a competitive market would be better for us than an oligopoly or monopoly prone to price-fixing or cartels.

Choice matters, but it was always conceived as an individual market choice (e.g., I switch companies), not an individual due process (e.g., I challenge that company’s act).

Does it have to do with Public Interest?

Those two characteristics seem to contribute but are not sufficient to justify publicness. Something about values permeates the discussion (this project’s DigiPublic Values perhaps?). While for centuries, public lawyers have endlessly debated the limits of public and private interest, whose strict boundaries are an 18th-century creation, we consider some goods as inherently more public than others. If a platform deals with online speech, it seems more quickly drawn into the public law test than if it only sells clothes, even if both have hierarchies and market power. There is something about freedom of expression that is almost inherently constitutional: it is freedom of expression that despots quickly target when they abuse their power. It is then easier for us to make the same claims we once did against the State, redirecting them against private platforms.

However, this is just as slippery a slope as the others. What is considered public interest can vary and is undoubtedly contentious in the constellation of constitutional traditions in Europe. For example, should sports be considered a public interest and hence, should FIFA become a public administration? Should big telecom or energy companies, or infrastructural AI companies such as Microsoft or Nvidia, be treated as public administrations too, as they become critical for our lives? And can a company start its life as a non-public interest and then become so relevant as to become ‘public’ (in the European sense) as public utility theories advocate? 

I have some ideas of my own for this, but this is not the purpose of this blog post. The idea was to simply point out that we need to devote more effort to understanding what triggers the ‘metamorphosis’, from private to public, before we can confidently build entire systems of governance that mimic public law but are ill-adapted to the more complex realities of Law & Tech.

BIO

Francisco de Abreu Duarte

Francisco de Abreu Duarte is co-founder and CEO of The Legal Place, a legal edtech. He holds a Ph.D. from the European University Institute focusing on Law & Technology and two LL.Ms from New York University (NYU) and the European University Institute. Francisco has worked as a legal advisor in both the Portuguese Government and the European Parliament. He is also the founder of The Digital Constitutionalist Blog and the EUI Metaverse RegLab.