Distributive justice: the evolution of an instinct

The story of humanity is one of working together, of pooling our efforts to achieve more than possible alone. For millennia we teamed up to hunt, build homes, defend communities, raise children, harvest crops and tend cattle; today, some of us are lucky enough to live in countries blessed with a welfare states and a national health service.

For all this time, we’ve faced the basic problem of how to divvy up the goods of society — from meat and maize to power, wealth, social status and respect. Who should get what? Should allocations be based on individual merit? Or need? Or should things be distributed along strict egalitarian lines, with everyone receiving the same? What is fair and just when it comes to redistributing the collective output of communities and nations? These questions of distributive justice are of course well-trodden in political and moral philosophy. But they also arise in day-to-day life, and there’s reason to think that we might come into the world predisposed to lean towards certain distributive principles.

They take a little while to kick in, however, and they evolve over time. Children younger than 3 tend to be almost entirely selfish, a direct conduit between world and ego. Give them a choice to share a hoard of sweets or keep them to themselves, and self-interest triumphs. By 5 or so, kids tend to be a bit more sharing and egalitarian, and even more so by age 8 — and this pattern is seen across cultures [1]. At the same time as this generosity blooms it becomes more focused: kids become more likely to share, but this largesse is preferentially directed towards other kids in their social group [2]. That is, these prosocial, egalitarian sentiments become more parochial as they become more pronounced, which is particularly interesting in light of evolutionary theories suggesting that parochial, in-group altruism evolved hand-in-hand with out-group antagonism, or what we normally call war [3]. (And this isn’t just a coincidence: the threat of inter-group conflict creates a selection pressure for communities to pull together to fight the common enemy, so in-group amity and out-group enmity spur each other on.)

The factors that feed in to thinking about distributive justice also become more nuanced over time. Children younger than about 5, when they do share the rewards of a collaborative effort, tend to do so along egalitarian lines, such that each contributing member gets the same. By the time they’re entering adolescence, they’re much more likely to take a more meritocratic approach, such that those who worked hardest in the collaborative endeavour reap the greatest share of rewards [4].

In a new paper published online at Nature [5], Katharina Hamann, Felix Warneken, Julia R. Greenberg & Michael Tomasello explore this developmental story even further, and give it an evolutionary twist by comparing the behaviour of very young kids with adult chimpanzees. The team came up with some clever experimental set-ups in which children or chimps could collaborate to retrieve rewards — toys for the kids, grapes for chimps — but they were designed so that even if the kids or chimps put in the same effort the rewards were divided unequally. This created a lucky and an unlucky child or chimp, who could then give away some of their extra bounty to restore equality. (The exact details of the set-ups used are quite complicated — but if you want the details, I’ve explained what occurred in an Appendix at the end of this post.)

So what did they find? Well, kids 2 or 3 years of age were more likely to share after they worked together to obtain the rewards, compared with simply being given the rewards followed by an opportunity to share. Chimps, on the other hand, showed no such inclination to share after collaboration. This suggests that at an early age children’s sense of fairness is sensitive to the significance of collaboration, a sensibility that even adult chimps lack. As such, it’s plausible that this is an evolved skill, one that emerged in the human lineage sometime in the seven million years since out species diverged.

Why should humans have this subtly of thought when it comes to distributive justice, when chimps don’t? Hamann et al. argue that this cognitive difference could turn on the different foraging strategies that the two species have historically adopted. Unlike humans, who gather and produce much of their food by working together, the great apes (which includes chimps) are not generally collaborative foragers; although chimps will team up to chase and kill monkeys, they then hustle to get their share of the kill, or share only with established collation partners.

Because collaboration has been so important for human survival, it paid to be the kind of person others would like to work with. Trying to selfishly keep the output of collaborative efforts for oneself is hardly the way to go to establish the requisite reputation; much better to be known as the kind of person who will take on a share of the effort, and be willing to share the benefits. Chimps, not relying on collaboration to anything like the same extent, had less need to adopt such collaboration-enhancing sentiments.

This study was done with very young children, and possibly taps into the earliest sense of distributive justice that humans develop (and maybe the earliest to have evolved after we parted ways with our chimp cousins). Yes, it’s quite basic: it is simply concerned whether or not we collaborated in assessing how spoils should be doled out. But it sets the stage for a more complicated notion of distributive justice to be built up, one that factors in egalitarian considerations, and later the significance of effort and luck in sharing resources. And it’s striking, to me at least, that such young minds should already be on the road to becoming natural political philosophers.

1. Rochat, P. et al. Fairness in distributive justice by 3- and 5-year-olds across seven cultures. J. Cross Cult. Psychol. 40, 416–442 (2009).
2. Fehr, E., Bernhard, H. & Rockenbach, B. Egalitarianism in young children. Nature 454, 1079–1083 (2008).
3. Choi, J.-K. & Bowles, S. The coevolution of parochial altruism and war. Science 318, 636–640 (2007).
4. Almås, I., Cappelen, A. W., Sørensen, E., Ø. & Tungodden, B. Fairness and the development of inequality acceptance. Science 328, 1176–1178 (2010).
5. Hamann, K., Warneken, F.,  Greenberg, J. R. & Tomasello, M. Collaboration encourages equal sharing in children but not in chimpanzees. Nature doi:10.1038/nature10278 (2011).

Appendix: How the experiments were done
There were two broad sets of experiments done is this study: one with kids aged two or three, and another with adult chimps. I’ll run through the kid experiments first, then the chimps.

Collaboration and sharing in kids
In the first experiment with the kids, they were presented with the following apparatus:

In this task, one kid sits on the right, the other on the left. Both pull the ropes simultaneously — that is, they collaborate — which brings the sliding tray forward. As it moves forward, the black stops at each end of the tray move out of position. This allows one of the balls on the right-hand side to roll to the left hand side, as the whole apparatus slopes right to left. When the tray reaches the front, with the balls aligned with the access holes, the lucky child on the left can reach three balls, and the child on the right just one. Afterwards, they lucky child could give a ball to the unlucky child to restore equality.

The control for this experiment was a ‘no work’ condition. The set up was as follows:

Everything is the same as before, except there are no ropes to pull, and the tray is already at the front of the apparatus, with three balls on the left and one on the right. As such, the rewards are ‘windfalls’ and not the result of collaborative effort.

In further rounds of experiments, the apparatus was modified somewhat. For the collaborative condition, the new set up was this:

In this case, pulling the ropes brings a sliding block forward that pushes the four balls over a ledge, where they roll down two inclines to the left and right. However, the balls are positioned so that three fall on the left slope, and one on the right, resulting in the same unequal distribution as before.

This set up allowed for a variation in which both children put in the same effort and executed the same mechanical action, but in a way that their effort was merely carried out in parallel, and not collaboratively. Here’s the relevant illustration:

Now instead of one rope pulling one block, there are two independent ropes each of which moves one block.

Finally, there was a control no-work condition to compare to (c) and (d); in this case, the balls are presented as an unequal windfall:

Collaboration and sharing in chimpanzees

The apparatus used in the experiments with the chimps was considerably more complicated (hats off to the team for devising this!). Here’s an image of the set up, with some explanatory notes added (I’ll explain this further below, though you may also want to open this image up in a  new tab to see more detail):

OK, so what’s going here? The chimps were positioned on the left and right of this cage, and were separated by a mesh that bisected the apparatus. By pulling the ropes ‘out’ of the plane of the page, they could move the upper shelf towards the front of the apparatus (red arrows), thereby bringing the grapes into alignment with the access holes (dotted green circles). The upper shelf rests on two support shelves, but the one on the right slopes up, which causes the right-hand side of the upper shelf to tilt (see red dotted oval). This in turn causes a grape to roll away from the unlucky chimp on the right-hand side and over a blocked hole, eventually falling through a hole on the left-hand side of the pivoting upper shelf, after which it drops into a sloping tray and through another hole onto a see-saw near the lucky chimp. At this point the chimp on the right can push the see-saw down so as to cause the lost grape to roll back, and restoring equality. Meanwhile, the lucky chimp could also push down on the see-saw, ensuring it gets all three grapes (the one on the see-saw plus the two on the upper shelf). (It seems to me that in this illustration the blocked hole, drop chute, and catching tray on the right-hand side are redundant, and confusing – so ignore them!)

The researchers used two variations of this basic experimental set up (each of which was paired with a no work, windfall condition, in which the grapes were already distributed unequally). In one, only the lucky chimp could move the see-saw; in the other, if the lucky chimp tried to get the grape to roll off the see-saw for its own benefit, the grape would be lost.

In the first condition, in which the lucky or unlucky chimp could manipulate the see-saw, the most common outcome was that the “unlucky partner almost immediately tipped the see-saw and took the fallen reward for itself (63% of trials); in no cases did the lucky chimpanzee actively tip the reward to the unlucky partner (even though in pre-training they often tipped the food away from themselves into the other room, if they themselves could then go through an open door and get it). In the remaining trials, the lucky chimpanzee took the reward for itself.” The fate of the fallen reward did not differ between the collaboration and control conditions.

In the second variant, in which only the lucky chimp could move the see-saw, they exercised this option and tipped the see-saw towards themselves in 98% of trials. Again, there was no difference between experimental and control conditions. In the final variant, in which the lucky partner could not get the fallen food itself, sharing was increased: the lucky partner tipped the reward to the lucky partner more often than in either of the previous two conditions. And as in the previous two experiments, the results did not differ between the collaboration and control conditions.


About Dan Jones

Dan Jones is a freelance science writer
This entry was posted in Uncategorized. Bookmark the permalink.

Have your say!

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s