Games Without Fairness

Why do people play games that they know are unfair? Casino games are the most obvious example, but I have another in mind: EVE Online.

EVE is a notoriously demanding and engrossing game. And it is manifestly unfair. New players start with a pittance in cash and skill points. Existing players are almost all far ahead of them in both, because skill points accumulate as a function of real-world time. Still worse, a lot of how one earns cash in the game relates directly to those skill points, so the new player can likely forget all about catching up.

And it’s even more unfair than that. EVE has a significant sandbox component; players form quasi-political units called corporations, and corporations form alliances; these alliances can conquer territory, which they go on to control until some other alliance deposes them.

Territory is the real cash cow of the game, and those who control the most valuable territories are insanely wealthy. There is almost no practical prospect of any new player catching up to them. Indeed, there is almost no practical prospect of any old player catching up to them. ¹

And wait, there’s more: The real-world corporation that runs EVE Online has been known to play favorites among the user base, giving favored corporations money-making options not available to anyone else.

Fans of political theory have probably noticed that I’ve just walked them through the complete and total violations of several plausible norms of fairness: It is morally arbitrary, and thus unfair, when social and economic power reliably tracks age without reference to responsibility. It is barbaric, and also not fair, when those who conquer the most are allowed to become the wealthiest and most esteemed. And it is not fair (or it would not be fair) if God himself gave me the blueprints to a faster-than-light starship, allowing only me to profit from it.

But that’s just life in EVE Online.

Now, not only do people still play EVE, but usually they know the relevant facts about the game before they sign up: EVE has exactly the reputation it deserves. Facts about the game are available on Wikipedia. They’re covered in gaming journalism outlets. They’re discussed exhaustively in the game’s own forums. People even make memes about how brutal and arbitrary EVE is, and those memes are everywhere.

Analogies are rarely perfect, especially in the social sciences, but it’s hard to avoid thinking at least briefly about one of the most celebrated thought experiments of the last century, the Rawlsean veil of ignorance. I suspect that EVE says something important about this thought experiment.

Rawls asked us to imagine persons divested of all their particular attachments, meeting with others who were similarly divested. These persons would be allowed to know all the relevant general facts about how human life works. They would also know the general structure of various possible societies that they might be asked to join. Their task: agree on one or more of them.

Choosers would know nothing about their own future place within the society that they chose. For that, they’d effectively be playing a lottery, one whose precise odds were also unknown to them: If they agreed to create a society in which one social position was downright appalling, they’d best be careful, because it could turn out to be their social position. And if they created an exalted social position, they could never count on it being their own.

Lacking specific demographic details about any social position’s frequency, people behind the veil of ignorance would have no way to minmax the problem. They’d therefore have to treat the least well off with decency, because the least well-off could soon be them. And they would have no incentive to create exalted positions if these favored spots came at the expense of others.

For Rawls, a fair society was one in which individual non-economic rights were as extensive as possible, the general equality of rights permitting, and in which differences in economic holdings and social position were permitted only if such differences worked to the absolute benefit of the least well-off, relative to how those least well-off might otherwise be. Thus we might allow an extremely rich capitalist class, for example, but only if it’s necessary for creating better absolute conditions for the least well-off.

Here’s where EVE presents a challenge: It suggests that hundreds of thousands of people in the original position have been eager to take a bad place in a manifestly unfair society. Indeed, they’ve been paying a subscription for the privilege.

This certainly doesn’t sink Rawls, of course; Rawls is talking about real life, and EVE is only a game. ² But it does suggest that justice as fairness may not be a standard that we readily apply in all contexts, and that we should inquire further about when we do and don’t want to apply it. It may prove — though I haven’t proven it yet — that a real-world society doesn’t need to be fair to be worth inhabiting. The unfairness of a given society may be real and persistent, and it may even cry out to be fixed, and yet the reasons for wanting to live in that particular society may be overwhelming anyway.

On reflection, other things do seem able to make social participation worthwhile, or not, the society’s relative fairness notwithstanding.

Imagine a perfect instantiation of Rawlsean justice, albeit one in which the available life paths are very few, not through stigma or restriction, but through some entirely blameless mechanism. Perhaps in the distant, AI-run future, all deliberative decisions are best made by computers, and yet humans are paid, and paid well, but only because we house within ourselves such wonderful enzymatic mechanisms: In this society, “work” consists of providing urine and stool samples. The AIs appreciate and affirm our work, exactly as section 67 of A Theory of Justice would demand. We are fine little chemical factories.

In this society, all the basic non-economic liberties are strictly respected; the difference principle for justice in holdings remains in force, and perhaps it is more scrupulously observed than today. Work isn’t even a drudgery. It’s a nullity. As a result, and without any malice aforethought, no one can feel a sense of self-authorship anymore; no one ever sees their will carried out in the world, not even in a trivial way. The social basis for self-respect is abundantly present through the AIs’ gratitude, but it does not appear to satisfy.

Would you take that life? Or would you prefer a grossly unfair but extremely exciting life? What if you had no hope of catching up to the Space Lords, and yet you still faced meaningful choices, and you felt yourself to be of some consequence, and you had projects you could sincerely delight in, even if they were small and humble, just as most of our projects always are? Rawls may be right that self-respect is the most important primary good (section 60), but he may be wrong that it is in society’s power to grant it, whether by fairness or otherwise.

What if we met behind the veil of ignorance and judged societies not exclusively on the basis of their fairness, but also in some way on the basis of their capacity to foster self-authorship, considered independently of social esteem for the author? We should certainly reject some types of unfairness, of course, but what if in some other cases the pursuit of fairness came to undercut self-authorship, or perhaps to efface it entirely? Games like these exist, but they don’t sell well, at least not to adults.

In which circumstances do we choose fairness? And in which do we choose self-authorship, even at the cost of some fairness? Is some bounded version of fairness an appropriate side-constraint? But what would that look like?

1. Yes, it’s possible in effect to buy in-game currency with real-world dollars. Insofar as the real world is also unfair, this does nothing to obviate the charge that EVE Online is unfair.
2. One obvious point of difference: In EVE, no one ever permanently dies. It’s literally impossible, short of deleting your account. But the question here is “Do we join clearly unfair societies?” — and not “Do we join societies that are so unfair that some may kill others with impunity?” Those are different questions, I think.