#2 A History of Decentralization

This month's thematic has been a real challenge for us and raised many questions in our minds. Why? The history of decentralization is complex and non-linear. But most of all, it is difficult to be considered from an objective point of view, stripped of the predominance of the state.  

Talking about decentralization leads obviously to discuss about centralization; to find the ghosts of history, to cross-reference the victories and failures of social-political movements; to discover some contemporary alternatives to the generalized centralization of our lives. Unless we consider that a technology is neutral, in the end, we cannot talk about decentralization without talking about governance, suffrage, politics or apoliticism, autonomy, organization… and the dominant model of centralization: the nation-state. Still, if a very vast literature and documentation concerns rise of states, it must be stated that the one granted to the opposite, i. e. the absence of a state, is almost non-existent.

If we consider human history in its entirety and through the births of the first social groups, state dominance is a very recent phenomena. In his book "Homo Domesticus", published this year, James C. Scott writes, "The state is originally a protection racket implemented by a gang of thieves who overcame the others”. Could this be the explanation for this historical denial, through a “story” written by winners? Yet, as we shall see, there have been a number of attempts and this utopian dream, still very present today, often seems to be the last hope for a civilization running towards its own disaster. If there is one thing that blockchain allows us to experience again, in its very sensitive way, is this somewhat anarchist dream to be able to surpass state power, bypass its rules and create new ones.

A review of Nancy Huston's "The tale-tellers: a short study of humankind" reminds us that history is undeniably the “one” story  we choose to tell among a thousand others. As a consequence, this article is not a ready and obvious history of decentralization. We would like to consider it as an essay, and not a journalistic enumeration of historical and social facts. It is a choice, ours, to present "one" history of decentralization. We have no doubt that it would have been very different if others than us were involved in its writing.

Therefore, this article aims to propose a “certain truth”, but also to pay tribute to the history of decentralization, to describe social-political movements of men and women, whose names have been forgotten in the History painted by the winners/dominants. It seeks to pay tribute to all the areas that have been defended so far, and of course, to all the areas we will still have to defend.

This month, Black Monthly will not include any interview. With the same reserves we had in choosing the narrative elements of this “story” of decentralization, we preferred not to impose any "expert" view / analysis of this history.

Part One : Is Decentralization just a word?

When they founded Aragon, Luis and Jorge (see interview) decided to articulate it around a symbol and a political genealogy. This reference did not arise out of nowhere. Aragon is thus named in reference to the anarchist communities developed in Aragon during the Spanish War. This results in a personal and political topology of Aragon's decentralization project. Here a political legacy of decentralization is arming a very recent technological decentralization.

Spanish Civil War Aragon Front, 1936-1939, Volunteer washing his hands, © The Working Class Movement Library.png

But, in order to draw a genealogy of the term “decentralization”, we have to go back further in history. The term itself appears in reaction to the word “centralization” that arise around 1790, during the phase of the French Revolution also known as the Directory. “Decentralization” will therefore sound like a “brother-enemy” of “centralization”, much more prominent, and as that period brings to its climax. As a matter of fact, the French Revolution is often reduced to the advent of a centralizing Jacobinism replacing the absolute monarchy. But the struggle between the Jacobins and the Girondins is not limited to a simplistic opposition between centralization and federalism ; the federalism advocated by the Girondins had nothing to do with a real project of administrative and political decentralization as what could be seen later in Spain or Germany. Besides, it was later exploited for political and historical purposes in a binary opposition between Jacobinism vs. federalism, Paris vs. province. Either way, the Jacobins, triumphant over the Girondins, led France towards bureaucratic centralisation (which was previously only political).

We are not going to make a crypto-history out of the French Revolution! Let's just say that during strong political/historical events, in the spirit of the age, words and ideas flourish. The term “decentralization” gained usage in the 1820s. Tocqueville, in the middle of the 19th century, summed up the situation by seeing in the French Revolution a dynamic of decentralization that had given rise to an even greater centralization : “the French Revolution began with a push towards decentralization but became, in the end, an extension of centralization”. Interesting reflection that could be applied to the web: born for decentralization, instrument of centralization, challenge of redecentralization.

Centralization/decentralization is not only a French matter, of course : it should be noted that federalism is an idea as ancient as politics and really took shape with the United States or the Swiss Confederation. Let us also imagine that there was decentralization against which a desire for centralization was formed; one example of this being the Grand Siècle which sought to centralize French language by eliminating regional and dialectal singularities and thus, in an attempt to consolidate the absolute power of the monarchy. This led to the creation of the Académie Française and a rigorous codification of grammar. In short, more than decentralization, we should probably talk about redecentralization.

This combination of centralization and decentralization, which can be traced back to the 18th century, will haunt the 19th century until today. Meanwhile, in 1937, the Aragonese anarchists mentioned above launched their project of collectivization and federation of agricultural lands and villages (such as that of Lecera, in Aragon). It took a myriad of thinkers and socio-political movements revolving around federalism, municipalism, libertarian communalism and democratic confederalism for the concept of decentralization to deepen, enrich and come to the complex meaning it has nowadays . Its relevance may never have been so strong: let us simply think of the Commune of Rojava inspired by the libertarian ecologist Murray Bookchin or of projects such as the Suite du  Monde, a blockchain-based municipalism.

This tension between centralization and decentralization persists through the advent of the blockchain and Decentralized Autonomous Organizations. We can draw here a quick parallel.. Just as blockchain with the  premises of a decentralized utopy in the early Web, Anarchists, pro-Federalism and libertarian organization of the 19th and 20th centuries dreamed to revive the decentralization expectations of the French Revolution. While anarchist and libertarian organization of the 19th and 20th centuries wished to bring back the spirit of decentralization inspired by the initial project of the French Revolution, technologies such as blockchain could reconnect the Web 2.0 to the decentralized world initially offered by internet. Undoubtedly, the common and main challenge of decentralization, within a political state as well as in a technological ecosystem, is to avoid  its corruption from bureaucracy /technocracy with a mode of governance that would allow its balancing: a glance at the current debates on the governance of the blockchain ecosystem shows us that this issue is far from being closed.

Part two : Decentralization, Centralization, Collusion, Coordination

In the major text The Meaning of Decentralization, Vitalik Buterin showed that he understood both the semantic vagueness and the critical issue of decentralization: “Decentralization is one of the words that is used in the cryptoeconomics space the most frequently, and is often even viewed as a blockchain’s entire raison d’être, but it is also one of the words that is perhaps defined the most poorly”.

He swaps the definitions of a decentralized and of a distributed network: the network labelled as decentralized is usually the one labelled as distributed and vice versa. Because what is distributed is what does not have a central control body. Let us take a state that would show its willingness to decentralize - any resemblance to existing states is fortuitous, of course - and appoints a regional president in each region, give each region a regional budget: this system would actually be distributed, because it would be subordinated upstream to a central unit.

As we have seen earlier, centralization/decentralization is not only a matter of political power; since the French Revolution, for example, its bureaucratic and administrative aspect has been involved. Vitalik Buterin's article contains an interesting tripartition. It defines three complementary axes:

  • Architectural (de)centralization
  • Political (de)centralization
  • Logical (de)centralization

To say that blockchain is decentralized (as it is also said to be transparent, among others qualities) is not really accurate. According to Buterin: “Blockchains are politically decentralized (no one controls them) and architecturally decentralized (no infrastructural central point of failure) but they are logically centralized (there is one commonly agreed state and the system behaves like a single computer).”

On this particular point  Buterin raises an interesting paradox: how do we match the principles of decentralization and the requirement of logical centralization? In short, how can immutability and decentralization be conciliated? To overcome these apparent contradictions, Buterin will develop "three reasons for decentralization": fault tolerance, attack resistance and collusion resistance.  For the first two arguments, Buterin explains that decentralized systems are less likely to fail and more expensive to attack because they lack central points. It recommends, among other things, the use of proof of stake and the multiplicity of companies and organizations. The latter measure implies the benefit of healthy competition and leads to the last argument: collusion resistance. This point allows him to redefine in new terms the principle of collusion/coordination.

Collusion is the "coordination that we don't like". Indeed, Buterin distinguishes between good and “bad coordination”: he gives as an example this “miners trying to screw everyone else over by repeatedly coordinating 51% attacks”. We could limit collusion and coordination in order to prevent monopoly situations (situations in which one single entity owns the market/power), but wouldn’t that create a breeding ground for centralization ? Is this what meant Lane Retting  (the Ethereum Core Developer) in his recent tweet about Ethereum supposedly governance failure?

Buterin defines the logical centralization of  blockchain in these terms: "there is one commonly agreed state and the system behaves like a single computer". However, is and will there be a commonly agreed state?

On this point, the difference between the informatic level and the political level should be highlighted. A decentralized system like Ethereum is not governed as a political state. Nevertheless, in order to reach a commonly agreed state, must we agree upstream on the mode of governance (and therefore on the political side of decentralization)? When Vlad Zamfir, Ethereum lead researcher, has challenged Nick Szabo, he opposed what he called Szabo's Law, a minimalist attitude that would not want any change in the protocol except for maintenance purposes. The Szabo line would not include any policy in the protocol; while the Zamfir line, while respecting the games of power, would accept its politicization and would like “the inception of a new crypto legal system”.

We come back once again to the mode of governance. Buterin, to access a commonly agreed state, is particularly interested in different forms of voting. In collaboration with Glen Weyl and Radical Markets, he proposes a form of consensus through quadratic voting. Would this voting resolve a deadlock or endorse the governance of the blockchain on the model of census suffrage? It should be remembered that a census suffrage is a method of voting in which only citizens whose total direct taxes exceed a threshold are voters.  The question of how to vote is just as important for democracy today as it is for the blockchain. The importance of delegative democracy (or liquid democracy), a mode of governance whereby an electorate has the option of vesting voting power in delegates rather than voting directly themselves, should be raised here - and will be the subject of a full article on the Black Blog.

Part Three : Some concrete cases

We, Aragon Black's team, recently passed the AGP - Aragon Governance Proposal. This vote and all the proposals generated enthusiasm but also raised some concerns which is a sign of a healthy community.  Evan Van Ness managed to highlight the problem of onchain governance in a recent blog article. The votes would at the last moment switch because of the vote of one or more whales - in short: the votes would be subscribed, at the end, to a kind of plutocracy capable of taking all the decisions. A few minutes before the closing of the voting session,  a  whale would change the outcome with his own single vote.

This outcome immediately raises the problem of decentralization which is at the heart of the blockchain.A lot of people offered their own solution to this issue, sometimes radically opposed in their vision of the problem. Let us identify two trends, one on governance and the other on scalability.

Vitalik Buterin, as we say, is closely interested in QV, Quadratic Voting. Its inventor, Glen Weyl, wanted to overcome the tyranny of the majority and binary choices in votes (based, in fact, on a yes or a no) while eliminating uniformity between weak and strong preferences. The vote is called quadratic, because the cost of the vote represents the square of the number of votes. Thus, voting 4 times on a project would cost 16 tokens, 5 times would cost 25, etc. There is therefore the possibility, either to vote for several projects, or to express an interest in a project by voting several times for it. The more votes there are, the higher the cost increases.  This is not only a step forward for governance within the blockchain but also for democracy. There are already applications of this model. A year ago, the State of Colorado tested it in partnership with Democracy Earth, a network that combines blockchain and civic tech to create a borderless community. Voters had 15 tokens each (without financial values). However, how can this voting mechanism be tweaked for particular use cases? How to choose the number of tokens to distribute? Glen Weyl proposed that the money be redistributed among the participants once the vote was over. This does not prevent the rate of participation in the election from being indexed first and foremost to a cost. Eximchain, a public blockchain networks with privacy for enterprise supply chain applications, uses QV for its governance and specifies: "Voters can vote as many times as they want, but they are assigned a set number of voting tokens over a certain period of time and the cost of each vote/token increases in a nonlinear way". This system of governance is undoubtedly largely perfectible: perhaps we cannot perfectly solve the problem of governance, simply test it, improve it empirically, as it is in fact an aporia.

We have only mentioned the Democracy Earth project as a use case where blockchain and civic tech intertwined. Another case involving a form of decentralization through the problem of governance is Bitnation. Moreover, Bitnation presents itself as governance 2.0. Pangea, both a decentralized market and a secure mesh network, takes its name from the supercontinent that previously brought together all the current lands. The project seems bold, almost oversized: Bitnation is the first Decentralised Borderless Voluntary Nation (DBVN). The goal? Allow the creation of nations, without borders other than technological ones, where everyone could adopt their own constitution, mode of governance, code of law. Each decision or conflict would be resolved through P2P agreements. Bitnation hosted the world's first blockchain marriage, birth certificate, refugee emergency ID, World Citizenship, DBVN Constitution. 500 nations have been registered, for about 15,000 citizens. The blockchain and the Ethereum protocol, it seems, are equipping a new kind of decentralization, a network of cross-border micro-nations. Behind this nebula lies the model of the nation-state, without borders nor the birthright citizenship. Now several questions arise; first, the possible risk of a collusion between physical, territorial and virtual citizenship; then,  the idea of replicating the model of the nation-state in a nebulous and dematerialized global space.  In fact, the issue is also related to the scale of a technology and its relationship  to its environment, namely: should we create a form of decentralized organization as vast as Pangea? Is a fully virtual decentralized organization possible?

"Decentralized institutions can be supported by parallel, decentralized technologies," Erselan Serdem told CoinDesk. He is the leader of Rojava's technology development program. Rojava, or  “Democratic Federation of Northern Syria”, has been fighting Daech and the Syrian regime since 2012. Based on Murray Bookchin and the Kurdish political theorist Abdullah Ocalan, who promotes a form of governance called "democratic confederalism", Rojava tries to organize itself into autonomous and confederal cities and defend direct democracy, feminism and ecology. This political experience is currently focusing on cryptography and crypto-currencies. The aim is primarily strategic: not to use the Syrian pound and thus avoid economic sanctions from neighbouring countries. Rojava wishes to set up technological academies to raise public awareness on these technologies. Amir Taaki, one of the first Bitcoin developers and which  fought during the Rojava revolution, is part of that change. Beyond the monetary use of Bitcoin, Rojava would like to use blockchain technology for its mode of governance, i.e. decision-making at the community and confederal level. The expectations are that  this infrastructure could meet Rojava's democratic values A concrete example would be to control in a secure and transparent way the management and distribution of resources, monetary or human. Here, logical, architectural  and political decentralization could coincide.

In a project such as Rojava’s one, blockchain technology comes to escort and equip an organization that, in a way, resembles it through its decentralized nature. Its use on a local scale is therefore confronted with a secondary and physical ecosystem, so to speak territorial: in fact, the blockchain could exist in Rojava not as an entity separate from the world but rather as a concrete and conjoined interface with a social and political organization. This dual character, local and territorial, can also be found in the Suite du Monde project. The goal? To acquire and free agricultural land and real estate in order to build a ground for any activity that would allow more autonomy: i.e. communalist organization, issues related to habitat, energy, agricultural production, education, etc. What separates them from usual alternative community projects is among anything the use of private property, then the use of new technologies such as blockchain. “We are getting organized," says their website. In partnership with Osmose, a crypto-cooperative seeking to build the tools necessary for widespread cooperation in purchasing, investment, products and housing, the Suite du Monde wants to use DAOs as their mode of governance.

Towards Barbarians Insfrastructures

The State is a central issue on the matter of decentralization. Actually, to ensure its stability, the state would always rely on a more or less graduated form of centralization. Therefore, it seems obvious that blockchain(s), in its decentralizing aim, is in turn interested in the question of the State. The State is also interested in blockchain, sometimes wary, sometimes genuinely attracted.

In the Blockchain ecosystem, as anywhere else, there is most of the times a double alternative: either destroy the State (or at least do without it), or expect it to improve. In any case, the State is an ever-present factor. It is the privileged place of social conflict. However, in his anthropological essay, The Art of Not Being Governed: An Anarchist History of Upland Southeast Asia (2009), James C. Scott shows that the State is a recent invention which actually isn’t self-evident. Better still, he explains that states have always been crossed by flight movements of populations migrating to freer areas. There would only be autonomy where territories could be appropriate, so that autonomy would be achieved in the flanks of the state, where it is not present, where it is not watching.

“Civilizational discourses never entertain the possibility of people voluntarily going over to the barbarians”, James C. Scott writes. The idea would be to revive the attractive figure of the barbarian, who stands on the periphery of a centralized space (or the exciting figure of the pirate). The risk would be to end up in a spatial and territorial vision of social conflict (on the model, for example, of guerrilla). But, conversely, thinking of decentralization in a purely virtual way would lead to the derealization of this conflict and subscribe it to an abstract nebula.  This is the paradox on which we would like to end: the creation of barbaric infrastructures, the technological possibility of decentralization, in the blind spot of the state.

  • Aragon Black

Read more posts by this author.