The Incoherence of the Tragedy
On the Fundamental Incoherence of the concept of the Tragedy of the Commons
“There are moments when people love crime,” said Alyosha thoughtfully.
“Yes, yes! You have uttered my thought; they love crime, everyone loves crime, they love it always, not at some ‘moments.’ You know, it’s as though people have made an agreement to lie about it and have lied about it ever since. They all declare that they hate evil, but secretly they all love it.”
All the world’s a game and all the men and women and plants and animals and even the rocks and dirt are only players.
What is the essence of the Tragedy of the Commons?
The answer to this question is not at all obvious, though there is an obvious answer, which gives the whole theory the appearance of self-evidence: people are selfish, therefore sharing is hard.
As-is, human selfishness, or enlightened self-interest, or whatever you want to call it is generally uncritically accepted as a kind of emotional black box that generates certain cumulative political-economic effects like markets and Tragedies of the Commons.
But what kind of thing is the Tragedy of the Commons?
The more I examine it, the less it looks like a coherent concept or theory and the more it looks like Socrates’ typification of rhetoric: a (mere) experience. Experience is the start of science, but only the start. The Tragedy of the Commons lacks some hard-to-define criteria that would make it a satisfying or well-formed theory. It can only look back at situations where some selfishness played a role in some Tragedy in the Commons, and retroactively infer causality: “Because we suffer, we acknowledge that we have erred.”
An experience framed as a theory, like rhetoric, functions like rhetoric. As Aristotle observed, the most convincing rhetorical arguments are those with the appearance of self-evidence. The deep mystery of how the psychological sums to the social (or doesn’t), of how selfishness comes to play such a key political role, the existential context for these apparently mutually-exclusive situations, etc., are all papered-over by the appearance of self-evidence of the theory. Like Aristotle’s enthymeme, the very vagueness of the Tragedy of the Commons’ premises makes the whole all the more convincing.
Selfishness is certainly a condition or facet of whatever the Tragedy of the Commons indicates, but it’s only that. Prior to the selfishness, there’s a framing of a mutually exclusive situation. More for me means less for you, and vice versa. This negation, this NAND gate, is implicitly acknowledged by those who formalize the Tragedy of the Commons as a Prisoner’s Dilemma Problem. In the Prisoner’s Dilemma, 2 options leading to 4 possible outcomes are presented by 1 prosecutor to 2 isolated prisoners: 1) snitch on your friend or 2) keep the faith and refuse to snitch. For the different possible outcomes, different rewards are given:
If A and B each betray the other, each of them serves two years in prison
If A betrays B but B remains silent, A will be set free and B will serve three years in prison
…and vice versa
If A and B both remain silent, both of them will serve only one year in prison (on the lesser charge).
BLAH BLAH, ipso facto the prisoners will betray each other even though it would be better if they didn’t.
What’s striking about this formalization is the requirement of some contextualization or framing, which generates the antagonism. The law of the excluded middle is only maintained by the will of a hierarchically privileged Oracle, Master, Prosecutor, Demon, RAND Corporation, etc. Reminds me of Leibniz’s critique of Descartes’ Demon: he introduces the problem when convenient, then just as quickly forgets it. So too, modal logic, the guarantors of the excluded middle: would, should, could, must. Don’t these operators smell a bit gregarious? A special kind of being can be glimpsed in the shadows of these oughts, a being external to the individual, but with some distinctly anthropological features. Philosophers have sometimes called this being a thou
The thou is evidently human. Perhaps in the same way that the moon and the sun are evidently the same size in the sky. Such a strange accident, since we can certainly imagine other possibilities. There can be many species of human, as DNA sequencing continues to reveal. In Red Star, Bogdanov imagines Martians as Humans too. Many animals and objects possess many thou-qualities, especially in the literary register: Moby Dick, gold, neural nets. But there are evidently many things that humans, alone in our universe, can do. For one, only a human can communicate the terms of the Prisoner's Dilemma to another human.
Animals cannot be ensnared by the Prisoner’s Dilemma because their languages apparently lack certain criteria to make them capable of forming contracts or snitching. Contracts and snitching are peculiar speech acts, modes of communication or symbolic exchange, assuming a certain minimum social milieu of thous.
In an autobiographical passage, Dostoevsky relates watching a civil servant in a hurry, beating a horse driver, who was in turn beating his horse. Another Tragedy of the Commons! Now a horse is a simple creature, it knows nothing of language or game theory. A horse can revolt, I’m sure. And a herd of horses can revolt, if they have solidarity and learn from one another. I watched three little girl goats revolt after they were kept from their mothers and fed too little for too long. We named the youngest and the apparent instigator “demon goat.” Finally, these goats threw themselves against the wire fencing so violently that they learned how to kind of burrow under it. Finally, the women working on the farm (Anna, et al) revolted and just put the goats in with their mothers, against the wishes of the farm’s owner. Now the owner does not get quite as much milk, and the kids still sometimes wander under the dead electric fence and browse in the street, and absolutely refuse to obey humans, but the goats are happier, and so are we. And I half-consciously drink less milk. At all levels, the situation does bear a striking resemblance to a kind of game, but a game in a spiritual space: all animate beings in sympathy with one another. But while a goat can be a fenced in, and can formulate appropriate modes of gaining agency, and a tomato can give off quite a smell when picked, goats and horses and tomatoes cannot be ensnared by the terms of the Prisoner’s Dilemma.
So the thou is human, a rational animal. The I’s relationship to the thou is obscured, but it somehow mirrors the paradoxes of consciousness: consciousness is one, consciousness is many. I is one, I is many. I think Nietzsche refers to consciousness as a kind of higher body. In a similar sense, the thou is a kind of higher body. When a tree falls in a forest, if anyone sees it, then it happened. When I walk in the forest with Anna, I know what salal is, because she does.
The thou is also the cause and effect of the NAND gate: the Law of Contradiction assumes a milieu of thous: the essential unit of a multiplicity: thous grouping themselves, sharing and withholding information, separating the smooth from the striated, separating prisoner A from prisoner B, forming contracts, snitching. And no, we cannot generalize these processes to all objects, because the thou is manifestly unique. The existential role of the thou can be formulated in Kantian or Hegelian metaphysics, with their peculiar inflections. For a Kantian, the thou emerges from the categories, particularly at their “third level”: community, the disjunctive syllogism, necessity, etc. For a Hegelian, the dialectic starts with master and slave, the master and slave relationship presupposing a milieu, a multiplicity of thous. For 26-year-old C.S. Peirce, the Thou was Thirdness, which would make abduction something a lot like “the reports of others.” And then there’s Marx.
In physics the object is opaque, hiding behind the veil of perception. In psychology the object is transparent, because the object is the veil. But the thou is something else: neither I nor grass, neither quite transparent nor quite opaque, or perhaps more transparent and more opaque, but according to different criteria.
The existential peculiarities of these thous are the suppressed premise of both the Prisoner’s Dilemma and the Tragedy of the Commons. The Prisoner’s Dilemma predicts that the prisoners will fail to cooperate because neither has any reason to trust the other. Experimentally, however, people often do cooperate, because of course they would. Which still means that if you choose to snitch and the other prisoner chooses to not snitch, you can get the best deal! All you must do is lack a conscience: betray your fellow human, leave them to rot, murder the mean old landlady, take more than your “fair share,” because why not?
The existentialists will say that those who do not snitch do so for the wrong reasons, and if they’re optimistic existentialists, they will imply that the knight of faith or the overman or whoever will pass through some stage of evil, of snitching because he is liberated from bourgeois norms, before coming full circle to essentially quietist, socially-average behavior, still not sacrificing Isaac, but now with some aura of enlightenment. I reject this formulation as too psychological.
I suppose I’m team Dostoyevsky here. Yes, there’s the moral hazard and the existential redemption arc, but there’s also: polyphony! Not only the “higher men” are thous. As the prisoner, locked alone in a cell, is told by the prosecutor that they may take a few years off their sentence if only they snitch on some other prisoner, what is really the determining factor here? These prisoners are simple people, no fancy game theories. Well, in fact, they do have some fancy game theories. “You gotta know when to hold ‘em, know when to fold ‘em, know when to walk away, know when to run.” A skill Dostoyevsky never acquired. Now you think these prisoners are going to trust the prosecutor more than another honest guy just stealing bread for his family, trying to make a buck, buying a few counterfeit 20s? It may simply be wiser to do whatever the prosecutor doesn’t want you to do. But even if it is unwise to antagonize your lord and master, or to pity prisoner B, then prisoner A may still do so, because of the power struggle. The stubbornness of the underground man, the tendency to revolt, a constipated will-to-power; it is not always Big Brother imposing 2+2=5 on the dissident, it is sometimes the dissident asserting that 2+2=5 when cornered by an all-too-human game. If nobody snitches, maybe the prosecutor doesn't even have a case! Most prisoners will react against the prosecutor, as well as in sympathy with the other prisoner, as well as in tune with the polyphonies of cultural mores, moral expectations, bad conscience, the Freudian death-drive, Nietzsche’s much-maligned “gregarious instincts,” the pleasure principle, enlightened self-interest, selfishness, etc.
If there is a commons, like two shepherds with two herds of sheep trying to share a pasture, then you have the potential for Tragedy. But Tragedy cannot occur without some hierarchy of thous. The Tragedy’s implicit need for hierarchy makes the term “commons” misleading. Don’t these shepherds have bosses? In Governing the Commons, Elinor Ostrom shows how people do indeed cooperate well to manage resource exploitation. She claims that common-pool resource management only devolves into a Prisoner’s Dilemma when the resource is purely open access. But does even this situation contain the required criteria to make it a Prisoner’s Dilemma? The need for some hierarchical, human-made contextualization is such a given that we forget to account for it in the first place. An open access resource already mapped by humans, for exploitation. Grass grows in other pastures, but in this pasture, in this time, a special kind of contradiction is manufactured, between shepherd A and shepherd B. The structure of the Prisoner’s Dilemma looks a lot more like situations that States and Firms and Class Society exist to generate: heads I win, tails you lose.