(Re)Public
Political commentators may spend screens pondering on the true meaning of the word republic, maybe eliciting subtle variations and differences with the related term democracy. I am going to speak of it here from a computer scientist point of view. But first things first: republic comes from “res” (thing) and “publicus” (of the people). You also have to realize that “realize” means cause to happen and shares the same etymology. This remark is not without its importance in an era where information technologies have impacted how we perceive reality and virtuality and have produced a new kind of culture.
Elections, Consensus and Distributed Systems
The central problem of distributed systems is to ensure that the system works while it can be subjected to many transmission errors and or node failures. The failures can be intentional, due to hackers, or not, it can range from software bugs to hardware issues passing by power supply issues. The point of a distributed system is to be able to continue no matter what. Since any node can crash, or can be isolated from the rest of the network, there has to be a way for the participants of the system to settle down on what the state of the world is. Bitcoin is just a peculiar solution to this broad problematic. In the case of Bitcoin the service is only a distributed ledger. Participants have to agree on who owns what?, which is equivalent to who sends how much bitcoin to who?
I have advocated a lot in this substack that using distributed systems lenses to analyze how power is expressed in our society is worthy, see for instance:
2024 is the year when the most people are called to vote in world history. This number has increased at the margin since the French president Macron call to snap elections following the disastrous results (from his perspective) to the European polls. Lets observe how consensus building is thought in distributed systems to compare with what happens in the political realm.
The first difference that appears is the general framework. The idea in the Distributed Systems is that things are not going to work as they claim to. The whole point of research and engineering of Distributed Systems is precisely this one: how can we make sure that the system is still going to fulfill its duties in presence of errors/corruptions? It used to be the complete dual of what was the political framework: you were supposed to choose an administration and legislative body from a supposedly honest pool of propositions. The idea was that your political opponents also wanted the good for the country -hence far from the distributed system stance that the default is that nodes will be corrupted-, but disagreed on how to achieve it. Increasingly it is less and less true. Political oppositions are viewed as evil. The assumption that political opposition has good motives is vanishing at the same time as the trust in institutions is falling. The fact that both Republicans and Democrats claim that if they lose the elections, then this is the end of times. It is not entirely new, but it has reached so much importance than every other subject is dwarfed in regard. The same phenomenon appears in french politics but with a triangle: far right, far left and extreme center are each pointing at the other two saying that the world is going to collapse if they are elected rather than them. It is ironically the single political message that emerges from all side: me or the chaos.
The second difference is that, unlike in distributed systems, citizen are trusting the result of an election less and less. This phenomenon is mostly an American one. Accusations of frauds are not new but since 2016 they are running on deep state steroids. The Russiagate, which is the name given to alleged Russian interference with the 2016 elections, spanned over four years, involved almost all three letter agencies for not much to show. This was followed by the infamous January 6 Capitol attack, which started on the idea that 2020s election were stolen. For the 2024 elections everything is in place for a major denial of results by both sides. There are decent chances than one incumbent will be jailed at election time, while the mental presence of his main opponent can be questioned too: if Biden wins the election can we say that “he” won since there will be no measurable “he” at the time? Actually it looks like the election campaigns are more and more focused on about making elections less trustworthy in case of a loss rather than actual policy decision. So elections are becoming maximum dissensus points rather than consensus points.
A third difference appears in the interactions between local and global aspects. In a distributed system local decisions are made in order to insure global behavior of the system. Each node follows a script that when replicated will end up insuring global properties. It was also the nature of political power for centuries simply because there was not a lot of choice: information was hard to spread and it took time. Nowadays things have been completely turned upside down: first the agenda is set at a global stage then it percolates locally. This is now technically possible because we can synchronize policies at a global stage in a glimpse of an eye. Examples are everywhere to be found: from the whole COVID shit show to the Ukraine war (the international community — meaning the USA, Europe, Japan, Korea and Australia mostly— thinks that …). Your tax dollars are spent on a globalist agenda: it is not a conspiracy theory, just the observation of what is happening. One direct observation is that local towards global is inherently more stable than global towards local. Indeed remember how fast the agenda can change: from mandatory vaccine passports to everyone is forgetting to reclaim their extra shot… At the lowest level, the one of the citizen, you have no way to anticipate such brutal changes in global politics. Maybe because it is not the aim of the system in the first place?
Herd Immunity
When you look at a crowd going out of a subway train and crossing the flux of passengers that are going to enter the train, you have a very good example of a distributed system. Every passenger is like a node and the property that the system ensures is the peaceful replacement of passengers. The protocol used to make this system work is culture, understood in a broad sense: crowds in France do not behave the same as in Argentina or in the USA.
Now the interesting question is what happens when there is a system fail? In the case of truly distributed systems, one has to rely on local reactions to handle errors. At the end of the day, a program has no bearing on what its remote. The best case scenario is when you can spot that something has gotten wrong. The same thing can be observed in the metro crowd. Everyone is equipped with its own survival instinct. The process is completely decentralized and asynchronous: people act on their own and, at best, rely on “women and children first” rule of thumb.
But what does it mean in the political realm? There is this paradox: elections are those massively parallel synchronization points. Every political institution halts and waits for the results to be known. The theory is that the system is restarted fresh with the new crew. The idea is that by correcting every so often the course of event we can control them. But just like in a subway station the options are very limited. Even if everyone wants to flee from a fire, the number of escape routes is fixed. There is only so much choice. All the fuss around the Deep State is the political equivalent. You may vote every x years but unelected bureaucrats are staying. They are like the walls of the subway station. you can only have an action inside a very limited range. This is how you end up with 15 min cities and with not what you wanted to in your plate for dinner.