It’s hard to believe, with daily news of fevered disagreement and of actual fighting, but human beings excel at cooperation. We mostly trust each other to do our part. Thomas Hobbes wrote, in Leviathan, that the state of nature was a war of all against all, exited only when our ancestors signed a social contract, waived their rights to take what they could, and authorized a King’s enforcement of their pledge. But “we have been obligate co-operators for hundreds of thousands of years; perhaps millions of years” (Kim Sterelny). Back then, no one could read a contract, or sign their name. No one could even think the thoughts the agreement contained. How then did they find a way to get along?
In social contract theory the stroke of a pen changes total war into civil society, ordered and secure. In reality this change took generations, step by gradual step, in “evolutionary time.” The Pleistocene Social Contract is Kim Sterelny’s account of those steps (he draws on the work of many others). I’m not qualified to validate or dispute his account, with archaeological evidence or expertise in evolutionary theory. But the account is interesting, and I can sketch its outline.
1. Collective action with immediate returns.
Many forms of cooperation are beneficial to all. Division of labor is familiar: I make the bread, you make the wine, and together we have a full meal. Sterelny says that the first form of cooperation was simpler: collective action with immediate returns.
The paradigm is group hunting. If we hunt together, and split the kill equally, we will each have more: we can take down bigger game, or more efficiently capture the small stuff.
But any form of cooperation, including this one, may be undermined by cheating. Free-riders enjoy the benefits while avoiding the costs. I drive the roads but don’t pay my taxes. Or someone may be greedy: participating, but taking too much of the profit—say, half the kill, when there are five hunters. Cheating in either form erodes cooperation's benefits. If it spreads too far there comes a “tipping-point where co-operators do best to cease cooperating and go it alone.”
Free-riding and greed are still with us today (you may have noticed). But we keep them well-enough under control for cooperation to remain beneficial. Techniques for controlling cheating have become more complicated, as cooperation has become more complex. In group hunting, free-riding is easily handled: since returns are immediate, they go only to those who participate. But greed is a problem, if bullying is possible. A bully is able and willing to beat up others, and take what they have. Bullies will take for themselves a too-large share of the proceeds of the hunt. It was a crucial moment in human history when
the great ape pattern was broken; the dominance hierarchy [organized around bullying] was suppressed.
What broke the pattern? The bad news is that “we do not know how or when bullies were stripped of their coercive power in the hominin lineage.” But there’s some good news: a speculative but plausible model of the process, which Sterelny sketches. Central to it is the evolution of the use of weapons. Weapons make bullying harder. Hand to hand, you might pack a harder punch, and be more willing to risk a fight. Knowing this, I’ll submit. But once projectile weapons are available, your advantage diminishes. Even with my flabbier arms, I could kill you with one shot from a safe distance. If everyone is armed, fewer will risk acting on their greed.
While the new weapons technology was a “difference-maker,” it was not by itself enough to end bullying: also needed was
enduring motivation [to retaliate against bullies] and an ability to act together....Over evolutionary time the motivation to suppress bullying shifted from flares of hot but brief reactive attitudes to become more sustained and persisting. Erectines or Heidelbergensians learned to treasure and sustain their resentments, a psychology not wholly unknown in [anatomically modern humans].
2. Reciprocation.
If I forage for food and share it around, counting on others to share with me later, we’ve moved past collective action with immediate returns, to reciprocation. Reciprocation is more beneficial than collective action: “foragers who share through indirect reciprocation gather resources more efficiently, and manage risk more effectively.” The flip-side is that cheating is harder to control.
Cheaters get more than they are due. Free-riders, because they get a “standard share” but put in less, or no effort; the greedy, because they take too much. (The line between these is blurry.) In group hunting, cheating is easy to spot; it’s less easy in reciprocation. “Commensurability” is a central problem. Suppose you return to camp with a bag of mussels; tomorrow I will hunt duck. If you take too much duck, you’re cheating. But “how much of a duck tomorrow is your contribution of mussels worth?”:
context and time complicate [this] still further. Suppose you were the only forager to succeed that day: everyone else would go hungry but for your mussels, and you did go somewhat hungry because there were so many wanting to share. Does that mean you can expect a whole duck tomorrow? Should you expect more, and how much more, if it is a couple of weeks before reciprocation comes your way?
Greedily taking more than your share no longer requires bullying, just the exploitation of this uncertainty.
There’s a related problem. Grog could end up with too much, on one occasion, through chance, or miscalculation. To know he is a cheater, one must know he generally tends to come out too far ahead. That’s easy in group hunting:
persistent greed will be clear to the affected parties. [But this is not so] in systems of indirect reciprocation....Band members do not have direct, observational knowledge of the contribution histories of others. So tracking cheating no longer depends on public information.
We solved these problems, somehow; Sterelny’s favored hypothesis has three parts: norms, gossip, and ritual. Norms (among other things) address commensurability problems. They say how a good is to be divided among the community. These norms are common knowledge, so those who, in an exchange, get more than the norms say they should, can be identified. Gossip then spreads this local information to the whole community. Cheaters become known as cheaters, and are punished accordingly.
But even in the best case, social tension will remain. Inevitably, suspicion will linger that someone or other is cheating. And even with no cheating, there will be inequality, which will generate resentment. Rituals function to reduce this tension. Through participation in them, individuals come to identify with the group. Rituals fosters feelings of loyalty. This identification and loyalty increase tolerance for uncertainty and inequality.
If norms, gossip, and ritual stabilized cooperation among our ancestors, then cooperation emerged only as we evolved to be a norm-following and gossipy species, capable of being inspired to loyalty by ritual. The story of how we did that is too long to tell here (nor do I understand it well enough to tell). Sterelny himself is brief on the topic, but one striking moment is worth mentioning.
I might know that according to the rules, Jones gets five mussels. And I might observe him take six. Cooperation is stable only if I am moved to punish him—even if I got my share, and so Jones’ gain did not come at my expense. How did we come to have this motivation? Sterelny, following Jonathan Birch, suggests that pride in performance is the mother of normative guidance. A motive to do what a norm says
evolved first in the context of skilled craftsmanship. The road to norms begins with pride (or with discontent, if something has gone wrong) in craftsmanship; caring about your work, perhaps partly for instrumental reasons, but importantly for intrinsic reasons as well. Skilled craftsmen hold themselves to a standard of achievement, and are dissatisfied with a product below that standard, whether or not anyone else knows or cares....Pleistocene craftsmen developed intrinsic motivations to get it right or put it right, where getting it right included shared standards.
Once standards are shared, it is short step to intervening when others violate them:
Craft norms can become other-directed, if and to the extent that a skilled artisan feels discomfort, disapproval and/or the desire to intervene at the spectacle of another agent’s inept workmanship.
You, perhaps, have had that feeling yourself. The last step is from being moved by norms of craft, to being moved by norms of social interaction. But the gap here is large. Sterelny suggests a stepping-stone: being moved by norms of shared activities, like hunting:
These have elements of both craft skill and social interaction. There is a way of getting a hunt right, which has both social elements—smooth coordination, seamless division of the spoils—and precise physical execution.
At first our conception of hunting well includes only what it takes to make the kill efficiently. But it is then extended, to include how the results are to be distributed. Thus a path is sketched from a well-made axe, to a well-ordered society....but only for small bands of hunter-gatherers. How Sterelny thinks cooperation was scaled up to larger, more complex societies, is a story for later.