These concluding remarks aim at following up on the presentations that took place at the Max Planck Institute Luxembourg by clarifying three closely related philosophical and anthropological issues that emerged during the seminars.
As it appeared from several presentations, blockchain technology has forced us to reconsider the received wisdom regarding the way legality was previously understood in legal circles. Sketched briefly, the story has three chapters. Until the 2000s, a relatively virtuous cycle had been performed and maintained between the written production of legal texts and their reception in society through various kinds of institutions, from parliaments to courts and law schools. This virtuous cycle was based on a shared level of literacy that had tremendously improved during the last two centuries in many societies, making it possible for more and more people to read the law and participate in the public debate.
The emergence of digital law disrupted this state of affairs: the ‘Code is Law’ era that started around 2000 internalized legality in pieces of software and nearly rendered obsolete the institutions intended to maintain the virtuous cycle at the core of the rule of law by replacing them with digital platforms that were supposed to do the same job more quickly and cheaply. Digital technology took precedence over traditional legal institutions, and at the same time, almost everyone—even highly qualified individuals, when confronted by the digitalization of legal systems—suddenly became illiterate either via lack of programming skills or lack of legal knowledge; reading the law was not possible as it used to be because it was encoded in a way that escaped our shared social literacy. In a more recent and rather unexpected step, some promoters of blockchain technology went even further in their wish to bypass institutions that were thought to be necessary for maintaining the rule of law by removing the traditional institutions’ control on the actual making of the law. Not only had the law become partially unreadable through digital technology, but sovereign parliaments were now deprived of its very drafting since it was delegated to self-executing operations such as smart contracts that were functioning over networks without direct human control, thereby avoiding the authority of common law. The once-shared literacy was now scattered in various social groups, and technology had become the only common mediation between them—the ability to decipher and to obtain access was now the main stake in the struggle for social fairness. This was indeed a “revolution within the digital revolution”1 that has hit hard on the notion of common legality.
But what kind of revolution is blockchain technology, exactly? Hannah Arendt in her reflections from the now published conference ‘The Freedom to be Free’ (1966) points out two different conceptions of revolution in modern times: the American conception which was mainly focused on removing an oppressive colonial power from overseas, but which was completely blind to the economic dimension of freedom since it could unquestioningly rely on slavery, and the French conception, which discovered at its own expense that political freedom is a mere word as long as a relative equality in economic and social conditions is not established as a precondition.2 This divide can help us characterize the revolutionary nature of blockchain technology as it is defended by its promoters today.
Blockchain technology is less concerned with the positive notion of freedom than with the negative notion of injustice, although both belong to the concept of legality. But is it political or social injustice with which blockchain technology is mainly concerned? Having diagnosed a persistent failure in the rule of law, blockchain advocates locate injustice not in the inequality of economic and social conditions but in human arbitrariness and prejudices. In this respect, they are more sensitive to the repressive aspect of power than to the economic and social equality necessary for freedom to become effective. Why is it so? Essentially because the solution to injustice which is promoted by blockchain advocates is purely technological, and technology by and large ignores the social dimension of injustice. Hence its political dimension comes to the fore: no control exercised by the politically powerful can limit the exercise of blockchain technology, which is claimed to help overcome human bias and arbitrariness in a way that no political regime has ever dreamt of. But the relationship between technology and the social world is far more complex than what it seems at first because technology has the hypnotic power to be entirely forgotten at an individual level—just as we do not think of electric potential when we switch on a light—and also at a social level, as was the case with slavery for the American Founding Fathers. Blockchain technology stonewalls what does not belong to its own realm of exercise although it was devised as the precondition for solving a social problem which has to do with injustice.
The case of blockchain technology keeps the question of human arbitrariness separated from that of economic equality by relating to the physical and social external world only through “oracles,” i.e. discrete filters capable of transforming external reality into computable data, thereby ignoring the tremendous energetic cost necessary for blockchains to function3 as well as the inequality in internet access and economic conditions; in other words, everything that does not belong to the digital literacy specific to blockchain networks in which users are supposed to be peers. The question of injustice requires nonetheless a stronger collective awareness of the fact that blockchain networks are part and parcel of an ecological and social environment which cannot be limited to the exchange of computable data only: in order to become free, citizens have to be liberated not only from fear but also from want, as Arendt puts it. It therefore should be kept in mind that there is a connection between the political and the social dimensions of injustice which, although they are almost unconsciously kept apart in blockchain technology as in technology in general, have to be investigated at the same time in order for the problem of injustice to be properly addressed. We must resist the technological temptation to treat the two sides of the problem separately because fighting injustice politically is not yet fighting it socially. In other words, the fight against injustice is not yet the fight for justice; strange as it may seem, injustice and justice are not two sides of the same coin.
The solution to the problem of injustice put forward by blockchains relies on a trust in the autonomy of technology which is based on a two-step reasoning that connects technology to the notion of social exchange. Firstly, the social trust in blockchains is based on the technological trust in the reliability of computer processes. Because of this technological trust the problem of maintenance is globally overlooked, and information technology processes are not supposed to require permanent human intervention—which they do. Secondly and more importantly, computer reliability depends on self-executing programs which make it possible to extend social trust to any type of social exchange without bringing in human biases and power struggles. It is therefore the delegation of social trust to autonomous self-executing computable processes which lies at the core of a technology like blockchain that aims at promoting social justice. And it is therefore this autonomy in programs which enables the claim that the symbolic mediations of traditional legal institutions can be disposed of. Is this two-step reasoning correct? The first question to be asked is: Why should computable processes be considered reliable in the first place? There are at least two levels to consider here—that of the connection between computable processes and blockchains and that of the connection between blockchains and their social consequences.
Blockchain processes operate within a finite world. The recordings of exchanges on a blockchain are by design traceable, for it is computably possible to track them down (they are logically decidable) as they are part of a finite list of items (all of the diamonds of a certain category, all of the vintage cars of a certain type, all of the products arriving at a certain harbor), even if a new item can always be added to the list, however large the number of items in the list is. From a logical point of view, a blockchain thus operates in the same kind of environment as fragments of first order predicate logic or Presburger arithmetic, both of which were proved to be complete and decidable.4 But due to the finiteness of the list involved in the case of blockchain processes, they only operate on a small part of the computable space which also comprises infinite undecidable processes. The division of the computable space into decidable and undecidable processes was made possible by a negative theorem5 which proved that the act of listing items was at the same time a computable process that could generate the list mechanically, but that this computable process could not be put into the very list it was mechanically generating. There was therefore an intrinsic limitation of the listing process by computable means since the process of listing itself, although computable, could not be listed. By this reflective twist, the demonstration established that there was more in the computable space than a finite listing process. This should remind us of the self-executing programs that are at the core of blockchain processes because the self-executing programs are of the same logical nature as the reflective twist just mentioned. Although computable through and through, the idea of using self-executing programs does not pertain to the decidable domain; it is itself an oracle. It should therefore be underscored that the notion of a self-executing program cannot establish its own validity without being established by a human decision external to the framework of self-executing programs. This is, for example, evidenced in the blockchain legal world by the recourse to human expertise in decentralized juries as soon as the conflicts under scrutiny become of a technical nature.
It would therefore be illusory to claim that legal blockchains are able to bypass traditional institutions because of blockchains’ ability to stick to only traceable processes. They do not. If it were the case, this restriction would render the delegation of legal decisions to self-executing programs impossible because the usage of self-executing programs implies more than traceable processes—the usage of self-executing programs and the necessity to go beyond traceable processes go hand in hand. Politics comes back where it was not expected, the use of technology itself. The two steps which connect technology to social exchange should therefore be interpreted in a different way than what was thought at first; technology cannot expand to the legal domain without a collective decision that does not depend on technology only but on a prior collective agreement to use technology or not. This agreement lies at the very foundation of the rule of law: we are the only ones to decide whether and to what extent we want to delegate legal decisions to traceable processes.