Red giants: statistical fat tails and revolutions as inverse risk-management


I consider my small intervention, in the form of this blog, in the continuity of the Bolsheviks and the October Revolution.   Not only because I want all forms of  oppression, exploitation, and misery to be destroyed, but I also desire an ambitious, positive program as well,   such as how Bogdanov dreamed of socialism in Mars, or how delegates representing millions of workers around the world, met in the First Congress of the Communist International to  draft a program for a world, socialist republic.  I desire a planetary, democratic community, where individuals of all nations, genders, and ethnicities, get together to direct and plan the world-system for the purpose of  emancipating humanity from toil and misery. In contrast,  the capitalist economy  is a blind and purposeless demiurge  that turns children in the periphery into soldiers and  slaves,  citizens  in the first world  into emotionally damaged and  medicated  automatas, and cruel idiots into presidents, CEOS and autarchs.

I expect  for this month that the internet will be bursting with  didactic lessons on the October Revolution – how certain, socio-economic conditions  of Russia and tactical decisions of the Bolsheviks determined  the revolutionary overthrow of the Constituent Assembly, and what these lessons mean for socialists today.  Different groups, depending on academic pedigree or sectarian loyalties, will excavate different lessons  – with some claiming that we need a party modelled  by the bolshevik approach, while others will suggest that the era of ambitious political programs  is over, given that the structure of capitalism today has essentially fractured the identity of the old mass worker which the  radical socialist program was built upon.  Finally, some like certain maoists, will argue that revolution is a “science” that is updated by revolutionary events, with China updating Russia, and Russia updating Marx.

The large spectrum of different conclusions  implies that the problem of didacticism is hard. However, I think it is because all these agents are asking the wrong questions.  Revolutions are abrupt changes: extreme, highly variable, non-linear, almost unpredictable, but usually under-predicted – in short, they are black swans. I’ve talked about the black swan before – basically it is a pop-finance/statistics term that describes highly impactful but unpredictable events, like the invention of the internet, the publication of Ulysses by James Joyce, or  the October Revolution. Other black swans are earthquakes, nuclear meltdowns, and terrorist attacks. This “black swan”  dynamic makes revolutions incredibly epistemically opaque to us, and also bounds the type of questions we can ask about them.

Trying to forecast whether revolution is possible on not in a given timescale is like trying to forecast the next earthquake and its magnitude.  Many radicals treat revolutions as a function of certain inputs – for example, how the pressure of a gas is proportional to its temperature.  So if you can excavate from history the variables that are coupled to revolution, you can analyze those parameters and predict whether revolution is in the horizon and how it will look like.  However, the statistics of revolutions are more comparable to earthquakes – earthquakes, like revolutions,  have technically concrete, quasi-deterministic causes, but they are usually epistemically opaque to us. You can in hindsight analyze the dynamics of earthquakes, by solving a set of physical equations in a computer, but you cannot possibly know when will the next earthquake be, or what will be its magnitude. Similarly,  historians can analyze in hindsight what were the causes of a specific revolution, but they are incredibly ill equipped to delimit the possibilities/impossibilities of the next revolutionary event.   This does not mean that future prognosis based on longue durée history is epistemically prohibited, but that history can only be used to forecast slow-changing, long averages, not shocks and sudden jumps. Revolution, if anything, is the quintessential  example of a historical shock, therefore no amount of PhDs and “brilliance” can  prognosticate the possible horizons of the next revolution.  Whole traditions of socialism  exist that have “caked” in them a strategy of revolution or social change, but many of them are nothing more than dogma.

The October revolution actually gives us an empirical example on the unpredictability of revolution.  The soviets and the bolsheviks didn’t entertain the possibility of overthrowing the Constituent Assembly until  a few months before the assault of the winter palace, after a long and gruelling polemic about whether  Russia was ripe for socialism or not.  The revolutionary timescale where everything was done and decided in their case was less than a year long.  In fact, the Marxist orthodoxy based on  longue durée historical analysis at that time predicted that Russia was not ready for socialism, given that it was underdeveloped and hadn’t gone fully through capitalism yet. Yet, countering all intellectual and theoretical expectations, and therefore embodying a statistical black swan,  Russia experienced the first socialist revolution in history.

In contrast to the “historical-academic” approach, a better method  for the study of revolutions from the perspective of a socialist is inverse risk-management. Just as how engineers and planners might design buildings to withstand earthquakes or massive floods – events of low probability but with high impact, socialists  should be “prepared” for the revolutionary window.   Revolutions  are low probability, high impact events, and the only possible epistemic approach to them is to be prepared. Just as how the capitalist state has plans in case of social decomposition, such as how the United States  subsidizes agricultural production in order to have food sovereignty in case of a world-historic break-down, socialists should build their own infrastructure and presence in society, in a patient, thoughtful manner, in order to be prepared to ride the earthquake, so to speak.

What does this preparation consist of?  That is the million dollar question.  However one cannot extract  the solution to this problem by looking simply at longue durée history, because such a study concerns with long, slow moving time-scales, while the concrete actions that need to be taken are short-termed  and granular and extremely contingent to the conditions that exist now. Therefore, an “apple to apple” comparison between the old socialist parties, and how socialist parties in the 21th century should look like is impossible.   Indeed, some socialist “pessimists” explain their position by simply stating that the conditions now are very different than the ones of 1917, and therefore a  revolutionary, socialist party is an impossibility.  However, the only thing we can know from such an argument is that a socialist party would look different than the one of 1917, which is a very banal assertion. Perhaps, a better approach is to look at the current existing social averages to start building a party in the mean time, while having flexible enough tactics that can be modified depending on future social contingencies. Such an approach would require engaging with current tried and tested modes of financing, organization, and media that are being developed by current, bourgeois organizations, while at the same time, taking into account longue durée historical analysis to develop a radical, principled socialist “maximum program”. Therefore, such a program would be  inspired from the historical experiences of socialists, but “filtered” by granular tactics informed by modern scientific disciplines, current aesthetics and present values.

6 thoughts on “Red giants: statistical fat tails and revolutions as inverse risk-management

  1. Tunisian man self immolation, Egypt uprising, Bahrain, events of Arab Spring (going on memory here). Would you comment on those events? Cracks in the ice?

    Liked by 1 person

  2. since i’m sure you’re sick of me already, i’ll try to keep my comment here short and give some final reflections on your interesting article and our earlier conversation on twitter.

    this is a good piece which lays out the argument in a more sophisticated way than the fragmentary nature of 280 characters allows. and certainly there is much to commend in the argument: the ideology of “revolutionary science” has generally speaking led to nothing more than cargo cults and tragic farces — see, for instance, the history of the “new communist movement” in the united states. in that sense, the social revolution of the 21st century cannot take its poetry from the past but only from the future.

    all of this so far the “bleak left” would no doubt agree on. they are not interested in didactics in any positive sense of providing a “model” — as in the maoist or trotskyist case — nor in forecasting the revolution, but in analyzing the “conditions really in existence” (the german ideology) to provide a hypothesis (as marina vishmidt has put it) about how revolution might come about today (as the immediate production of communism). the purpose of their historical analysis is made clear in the quote of mario tronti which adorns the top of their “history of the workers’ movement” in the fourth issue: “We have no models. The history of past experiences serves only to free us of those experiences.”

    nevertheless, historical analysis remains central. you may term this a “negative didactics” if you wish. you concede that “historians can analyze in hindsight what were the causes of a specific revolution.” but this is the entire point. we can look at the conditions of possibility for the workers’ movement and its characteristic forms of struggle through historical analysis. we can then look to see if these conditions exist in some way today. if they do not, then the question becomes how can one simply repeat the forms of the workers’ movement without these conditions.

    to say that this resolves itself into the banality that “a socialist party would look different than the one of 1917” is somewhat misleading. the very notion of “the party” is inherited from the workers’ movement. indeed, as a number of historians and political scientists have pointed out, the workers’ movement pioneered the use of formally structured mass political parties, spurring on other movements (christian-democrats, etc) to mimic the format. to simply say that it “would look different” conceals the question of why it should be assumed, a priori, that it is appropriate at all. certainly, the revolution will look different, although the banality of this statement should be weighed against its evident unpopularity among the ocean of trotskyist sects and maoist fronts. but to say the party will look different is assuming precisely that which is in question.

    to put it in the language of the article: we can use a “historical-academic” approach to research the conditions of possibility of old forms of organization and struggle. we can then reason inductively about the forms of the workers’ movement, in all their diversity and connectedness, and generalize about them and what made them possible. the black swan here, however, is not the revolution (on which we are all agreed is possible). the black swan is the existence of a party (or more accurately: programmatism as an entire comprehensive strategy) which defies what were thought to be its conditions of possibility, proving the induction made about its necessary conditions to be invalid. but what you have done is merely to assert that such a black swan might exist, not prove that it does, or that it is probable. the possibility of an incorrect induction – the black swan problem – is intrinsic to induction itself. but that doesn’t mean all results of inductive reasoning are invalid. you first have to find the swan.

    for my part, i’m open to arguments that it exists, and i hope you continue your project of reflection on revolution for our time on your interesting blog.

    Liked by 2 people

    1. Hi. The part that the workers’ movement invented the mass party is interesting. Do you know of any references about this?

      This opinion is interesting, but I would be more willing to accept it if the “party-form” would have faded in general, but it is the defacto “form” for doing politics for the bourgeoisie as well. So you would have to explain why is it possible that the party became historically “prohibited” for the working class but not for factions of the bourgeoisie. Because as it stands now, the “party” isn’t an exception/black-swan, but just “normal” politics for the vast majority of politicized humans iMO.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s