Inequality And The Anarchist Dream
The Anarchist Society’s charter city was beset with problems from the start. When Paul Romer’s Prospera began operations in 2020 on a Honduran island, most people had written it off as another eccentric millionaire fantasy play island. Then the coronavirus mutated and everything changed. Even as the 2nd and 3rd waves struck, the world held its breath waiting for the vaccine. It came. And didn’t work. The virus had changed its cloak. Now it wore a transgene from an old enemy, a gift from a lovely intelligent porcine companion we had chosen to eat rather than befriend. The mortality rate skyrocketed. Governments collapsed, order dissolved, and borders were finally seen for what they were, invisible not like a virus but like space. For decades the Anarchist Society had been printing pamphlets pedantically correcting people that anarchy didn’t mean chaos, and its use as a synonym for disorder was grounds for libel prosecution, with another pamphlet describing that this attack on free speech was not at all irony and in fact consistent with the anarchist belief of the sovereign individual island unto oneself. Now all over the world anarchy reigned and instead of correcting people, the Anarchist Society started owning their claim to the new world order, distributing pamphlets talking about anarchy and chaos as the natural tendency of man in his free state, and them as the defacto ruling arbiters of this freedom. They were mostly ignored. But Prospera prospered, and could not be ignored. Charter cities began to crop up everywhere. Rapidly. Many of them failed in their high ambition. Some failed in their pointlessly low ambition. Many more did just fine. Some were spectacular. All were too soon, according to the Anarchist Society, which wagged its finger at the fact that so many of society’s ills were faithfully reproduced in these charter cities, an exercise in reengineering societies to be the ideal versions of themselves, not mini-clones protected in bubbles.
They would get it right. Within the society, it had always been assumed that there were two basic irreconcilable factions, the Rawlsians and the Nozickists. But they reconciled fairly easily, agreeing on a multi-stage process on the veil of justice that made the exercise a little more complicated, but addressed all the concerns Nozick had raised. What threatened to be irreconcilable though was a small-voiced statistician who understood distributions and individual variance. In statistics, every data point had value and significance. Importantly, every data point had representation. Curves were defined as much by the points they covered as by the points they left out. Even the most normal distribution, the normal distribution, needed to be expressed both by its mean as well as its standard deviation. In the hard sciences, this representation was taken as it was. In the soft sciences, especially when the data represented individual humans, the normal distribution was approximated to a single line at the mean, dropping off to zero on both sides. There existed one average human being and nobody else. The Rawlsian and Nozickian camps, now arrayed all together in one big cloud staring down at the opposite camp, one lone statistician and a renegade psychologist who coughed awkwardly and stood up to go join him, having misunderstood the fairly obviously rhetorical question ‘and anybody else agree with this upstart?’. Individual differences mattered. Didn’t they?
Take inequality. The decision process was laid out this way. I would have to allocate $10M among 100 avatars, I could choose any distribution I wanted. Once it was time to populate the society, I would be randomly allotted to an avatar. As creatures seeking to minimize regret, surely we would prefer a fully equal society, $100k for everybody like a Monopoly game, which was renowned for its justice, fairness of outcomes, and ability to strengthen friendships and fix broken marriages. But it turned out some people were lottery players, if 1 person had $1M and the rest was equally divided among 99, then the 1/100 chance of being fabulously wealthy was preferable to a boring equal society. Most fell somewhere between these 2. Perhaps behavioral economics was wrong and we weren’t driven to making emotional decisions that were irrational in predictable ways. Maybe it’s just that the decisions are never presented to us in clear enough ways under material circumstances with consequences that activate our desire to make good decisions. Our decision making is either studied in the real world where choices are hazy and attention is scarce, or it’s studied in a controlled experiment where choices are clear, attention abundant but consequence and mental investment scarce. In this experiment, as a rational creature looking to maximize my outcomes, I should technically be agnostic between any number of scenarios with equal product of probability and payoff. There were a large number of income distributions with an expected payoff of $100k, so my choice between one that guaranteed 100k (all equal) to another was based on other preferences. Unfortunately, the statistician was correct, we had vastly different preferences.
The statistician and psychologist, hereby designated as S&P, were overjoyed. It was often joked that they got off on correlation porn, but when an unsuspecting intern walked in on each of them alone in the data center, it quickly became far less funny and far more disturbing. They really liked numbers. Really really. Like really really. She quit and joined the Mormon society charter city. She writes once a week. She’s doing really well. Last we heard, she became a statistician. And then the letters stopped. We assumed she gave up language as useless waste of symbolic space. S&P’s correlation task was theoretically simple to frame. The Anarchist Society was dedicated to protecting individual preferences, so it would not ask me to participate in this random allocation into a society whose construction I did not agree with. Yet, it was theoretically not impossible to construct the society that everybody agreed with. All we needed to do was find the correlates of differences in preferences and then adjust for them by adding new variables into the mix, a superimposition of multiple distributions if you will, until we found the superimposition that perfectly mapped every data point onto a consistent curve, a mean curve, a higher-dimensional normal distribution. Simple.
First they got rid of all the pesky noise, by controlling for physiological correlates. One person was really hungry before the initial experiment and therefore chose a very unequal distribution. He tended towards equality on the second go, well fed. Another went the opposite direction. One person had fought with his wife. Another had been spoken to sternly by his father. Another had just been helped to cross the road. All these people gave different answers on the second go. All the data was thrown away. Subjects were kept in a controlled environment, their vital statistics tracked, their nutrition and hydration optimally maintained. The experiment was repeated and the resulting distribution was taken as the base set. It took a month. Then the Anti-Elitist faction of the Society objected to the fact that mild classical music had been playing in the controlled environment and that our conceptions of what constituted the optimal conditions for Homo Philosophicus were inextricably linked with our socially constructed oppressive historical narratives. S&P started weeping. P just mumbled to himself and walked in circles. The faction backed down compassionately, and grudgingly accepted that perhaps we could keep this point in mind while the experiment progressed and would S&P please add a *mark below the graph explaining this possibly important fact once they stop crying.
Some rich people chose very steep distributions, severely overestimating their ability to live with less money. Some poor people chose highly unequal outcomes, following the general idea that lotteries are predominantly patronized by the poor, although why that would be the case in a new reality with a fresh start was questioned by P. The matter was brought before the panel. Should the sum total of our life experiences one way or the other be a factor in our decision for building a new reality? The panel was displeased. This was like giving the Manhattan Project the task to split the atom and make a bomb, only to have them come back on day 2 with a very grave ‘but what exactly is the meaning of life, perhaps we should meditate on such matters for a few years’. The whole point of upholding the value of individualism was that it was an end in itself, and not a means to achieving something else, a waypoint in some mythic journey towards platonic truth. The panel wanted diversity of experience and preference and opinion, and the fact that some of them maybe ill-begotten or unfortunate in their background or plain wrong was less relevant in a world where your opinions were your own. Unfortunately, in this stage of the exercise, your opinions were not your own, there was need for consensus. And while this higher-dimension blahdiblah that they drew a complicated flowchart for sounded well and good, there remained the issue that to get there we needed to poke at the heart of individual differences. Seeking to understand was different from seeking to correct, so the panel allowed S&P to control for long-term lived experience.
Rich people were given a primer in living poor. Poor people were given a primer in living rich. Some responses became more moderate. Some became more extreme. This was fine. Utility curves were becoming more nuanced. Risk-averse people chose equal outcomes. Risk-seeking people chose unequal outcomes. Base income levels had nothing to do with the distribution of these risk attitudes, although this was not the case before they started controlling the environment. Stress made people more risk-averse, and the poor had more stressful lives unless their immediate pre-decision environment was controlled. They went to the panel. Should we control the risk-attitudes of people by sanitizing their lives? The rich have stressors too, but only not right now, it isn’t constant like for the poor, it comes and goes in bursts, and for some the stressors are huge, like a business failure, a mean comment on social media, or assassination by a disgruntled actor from the south. The panel was displeased. But one of them agreed. He had recently been called ‘stupid’ on twitter and described it to everyone as literally like the worst day of his life and like everyone’s worst nightmare, and how it was so poignant and humbling that things could so suddenly and dramatically fall apart. He made a speech about how, today he would make a very different decision about creating a twitter account, and it might possibly be the better decision. So it wasn’t clear that stress made us take bad decisions, as opposed to focusing our minds on high-impact decisions whose import we dismiss because we’re bad forecasters or ostriches or just lazy. The panel had a suggestion, they kept the tissues handy before telling S&P. How about conducting a range of experiments for each person across different levels of stress, thus getting a distribution of distributions? S&P beamed at the suggestion, proving either that they were losing their grips on sanity or that the panel couldn’t quite figure out how their brains worked.
S&P didn’t like throwing out data. Everything else was fine. The Anti-Elitist faction realized their error. They just needed to have said hey what happens if we play hip-hop in the controlled environment and S&P would’ve beamed at them. 2 is more than 1. Just never tell them to throw out data and they’d be happy sitting in that abandon-all-hope-ye-who-interns-here cave. They churned out additional distributions of distributions. The distribution for endowment of prestige was overlaid on income and found to have a neatly negative correlation. Some people were agnostic between money and prestige. Many were willing to take a cut on one for a bump in the other. The extreme lottery-style distributions suddenly went away on this overlaid distribution. There was more consensus now. They were trending in the right direction of higher dimensional distributions. They went to the panel. Should we create a bouquet of benefits, like political power, social capital, prestige, admiration as role models, along with initial wealth endowments such that we achieve the perfect distribution of distribution of distributions with complete consensus? The panel was displeased. This was sounding a lot like regular society. Worse, this was sounding a lot like Marxism, with the structure of society engineered down to the color of cobblestones. The fact that it was guided not by Politburo but by the rigorous statistical treatment of individual preferences was a consolation, and the panel wasn’t the romantic kind to value nature as supreme law thus prioritizing emergent phenomena over designed simulations. It didn’t stick well, but this was just a theoretical exercise so they agreed on a set of criteria and endowments. They also agreed to a more flexible treatment of wealth endowments, with S&P showing how some people would be willing to take a haircut on the initial endowment in favor of a larger regular cashflow, while others were willing to go the other way around. This was already a dimension that, when added onto the inequality curve, tended towards a consensus response.
As S&P retreated to their cave, the panel stayed back to reflect on how this must fundamentally change the idea of the veil of justice. What they were doing was less about finding the consensus distribution to which everybody would agree to be randomly placed on. They were finding a far more complex solution. They had removed the randomness from the equation. Here there would be no allotment. Individuals would choose their exact set of endowments in this new world. Not only that, they are perfectly happy with everybody else’s exact choice of particular endowments. This had gone way beyond the traditional anarchist dream of the veil of justice and the individual sovereign. Imagine a world where I choose not only what I have but what everyone else has. This isn’t hard to do, this is a dictatorship. Now imagine a world where everybody chooses what they have but not what anyone else has. Technically this is an anarchy. Now imagine a world where everybody chooses what they have and also what everybody else has, and somehow these two things are consistent and offer up a non-trivial solution that actually exists. This is the new anarchist dream. They sank all their resources into the S&P cave. The data being churned out of that single room soon required the computing and storage space of a small country. S&P headed the effort. They excitedly reported they had achieved 90% consensus, and still had a whole host of untested variables. They were almost there.
The funds ran out. The Anarchist Society declared bankruptcy and was bought up by the Neocapitalist-Marxist Collective that was headed by a former mormon and Anarchist intern who promptly fired S&P and declared as gospel the mean value of the initial experiment in inequality behind a veil of justice. The panel was displeased. They were fired too. The NMC Charter City formally began operations in 2030. It closed in 2034 after a period of intense turmoil that was later without any irony described by their CEO as ‘very anarchic times’. S&P were never heard from again. Neither was all of that data they generated, the whole lot of which they had taken with them.