Dynamic Cooperation

[See also my blog post on this topic (including scientific references)]

Much of what humanity and humans have achieved has been through cooperation. This seems obvious and intuitive to us now, but the evolutinary process leading to complex, separate and sometimes competing entities cooperating at a human level of sophistication is nothing short of monumental.

Cooperation comes in many forms. For clarity, let us divide up cooperation into weak and strong cooperation. Weak cooperation is cooperation for some immediate and specific benefit, usually for all parties in the cooperative arrangement. Weak cooperative behaivour is present in nature even amongst parties of completely different species. The symbiotic relationship of small fish cleaning the mouth of a shark are the perfect example. In this arrangement, the shark gets an improvement to its cleanliness and health, while the fish receives a snack. The fish isn’t worth it as a meal for the shark, so it makes sense for both parties to behave in this way – the cooperation makes evolutionary sense.

Stronger cooperation is harder to explain. Strong cooperation is cooperation between organisms where the benefit is less immediate and does not neccessarily apply directly to both parties. Both parties do not neccessarily benefit in an obvious way – the cooperation can be altruistic. For example, in a simplistic sense a mother does not gain anything as an individual by risking her own safety to protect her young. Neither does a member of a small group have anything to gain by providing food to an injured member of the group. Yet strong cooperation, particularly in the form of altruism, exists in both humans and animals.

In scientific discussion of this phenomenon, there have been several explanations advanced as to why these traits exist not merely as a result of individual chance, but as apparently evolved, widespread, genetic traits. Chief amongst these explanations is kin-selection.

Kin-selection states that because organisms are essentially machines used by genes to promote their survival and reproduction, organisms will often behave in a way that achieves this by promoting the survival of genetically related organisms. That is, to the extent the another organism shares the same genes, it is as intrinsically important to the organism as the organism is to itself. So, behaivours that support the survival and reproduction of genetic relatives are likely to proliferate as part of the evolutionary process.

The existence of kin-selection implies altruism based on genetic similarity. Children share genes with parents, brothers and sisters with eachother, members of the same species even share a small amount. Provided circumstances do not place the organisms in immediate life-or-death competition, altruism and cooperation is extended to the other party in proportion to the shared genetics. Consistent altruism between close relatives can facilitate long term cooperative relationships and even social groups.

Another concept in this realm of scientific debate is group-selection. Group-selection states that groups of organisms that express cooperative and altruistic behaivour traits are more likely to survive, and thus proliferate than those who do not cooperate. For example, organisms that have a trait to share food with injured members of a group, appear to have an advantage over those who do not, because the cooperation strengthens the group as a whole.

Group-selection has been criticised on a number of fronts. Some critics have suggested that group-selection requires stable genetic groups that rarely exist in nature. Organisms do not usually survive or perish as a group. Instead groups expand, shrink, merge, split and mix. Evolution of traits happens over multiple generations, by which time the group composition will have changed drastically. Groups are not the unit of selection in the process of evolution.

Another, perhaps more lethal criticism of group-selection is that cooperative behaivours, while confering an significant advantage in the presence of others with cooperative behaviours, are in fact a major liability in the relationships with non-cooperative parties. For example, an organism provides food to an injured group member, at a cost to themselves, but the other when the roles are reversed the selfish party simply keeps the food and eats it. The more altruistic genes are quickly weeded out of the gene pool, because the selfish organisms outcompete the altruistic. The core of this criticism of group-selection is that it never has a chance to take hold, even where the group is stable, because it is simply wiped out at an individual level.

These criticisms seem to render group-selection as an obsolete theory, particularly in the light of kin-selection which partially explains altruism and cooperation in an easily understanble way. However, it is a major mistake to rule out group-selection, because both these criticisms have counter-arguments.

Imagine a population of organisms that each day form completely random and new social groups. This group stability only lasts for one day, then the groups are disbanded and new ones form. Each time the cooperative organisms begin to behave cooperatively, but only if the other members are cooperative too. That is, the cooperation is contingent – the organism only displays this behaivour under very specific conditions.

For example, each day the groups fight-off threats, search for food, and divide up findings amongst its members. If the group was cooperative, then by working together they were less likely to have succummed to external threats, more likely to have found a good supply of food, and more likely to have distributed it in a way that aided the survival of all members. The cooperation is temporary, but each day the cooperative organism has a greater than zero probability of negiotiating for cooperative group dynamics. The groups may be temporary, but cooperative members still walk away with the benefits of cooperation some of the time, whereas non-cooperative members walk away with no benefit at all.

Of-course, non-cooperative organisms undermine cooperative group behaivour because they exploit cooperation to receive individual benefits. There are two forms of this strategy. The weaker form is to simply avoid making cooperative or altruistic sacrifices, while attempting to still benefit from the cooperative behaivour of others. This is generally called “free-riding“. For example, a free-rider may sleep while other organisms expend energy hunting for food, but then collect an equal share of the food when it is cooperatively divided up.

A stronger form is betrayal. Betrayal is an active strategy of harming cooperative members of the group for selfish gain. For example, a bird minding a friendy relative’s nest might betray them by secretly destroying the relative’s eggs and replacing them with their own. Betrayal exploits the fact that cooperation essentially involves letting your guard down. Betrayers go further than just receiving the benefits of cooperation; they actively take measures to harm other friendly parties when they are vulnerable in order to reap a larger reward.

When we consider our example of randomly social groups, forming each day to hunt for food and defend against predators. The free-riders and betrayers gain all the benefits of cooperation, and more, without any of the sacrifices normally required. They outperform the cooperative individuals and quickly begin to replace them within the overall population. Considered this way, cooperation and altruism seem unlikely to survive. Evoluntionarly speaking, free-riding and betrayal suck the life out of simple cooperative behaviours.

However, contingent cooperation serves to offset this effect. Where other members of the group appear to free-ride or betray the group, cooperation is temporarily turned off. This is the first part of a more sophisticated cooperative formula that we can call “dynamic cooperation“. Dynamic cooperation is the employment of strategies to defend cooperative behaivours and their benefits. These measures ensure cooperation remains viable by protecting cooperative individuals and negating the benefits to non-cooperative free-riders and traitors.

There are several important components to dynamic cooperation:

Contingent cooperation – Rather than naively extending cooperation to all parties, contingent cooperation involves applying cooperation selectively based on observed cooperation displayed in other parties. (One similar concept describing this is the so-called “greenbeard” gene, though such a behaivour is almost certainly not simply a matter of a single gene). Extending cooperation at a slowly increasing but roughly equal level to the cooperative behaivour observed in other parties is a sign of this phenomena.

Harm minimisation – Because there is a evolutionary force encouraging free-riding and betrayal wherever there is cooperation, dynamic cooperation will usually involve harm minimisation strategies – behaivours that systematically attempt to reduce the instances in which free-riding or betrayal might occur. For example, an organism might emphasise careful observation of fellow group members in order to predict free-riding. Or activities that are vulnerable to betrayal might be carried out in view of the entire group, so that betrayal cannot occur unnoticed.

Active deterrance – In addition to more passive harm minimisation, organisms can actively target free-riding or betrayal and provide deterrance. This may involve punishing free-riding and betrayal by causing harm to the non-cooperative organism, or by revoking group membership and the extension of cooperative benefits. In doing so the uncooperative organism receives an evolutionary disadvantage to their behaivour and free-riding and betrayal become an increasingly unviable evolutionary path.

Cooperative pyramids – Intelligent multicellular organisms to prioritise other organisms for cooperative alliances and altruism based on kin-selection. A cooperative pyramid is an approximate visualisation of this genetic strategy. Organisms are arranged with the self and close relatives at the centre, and progressive more distant relatives fanning out to each side. The prioritisation of organisms is highest in the centre and gradually decreases, forming a pyramid shape.

By prioritising cooperative behaivours based on this pyramid, an organism is safer because cooperation is more likely to be reciprocated. This is addition to the benefit of simply helping a relative to make sure their genes survive. The cooperative pyramid is distinct from pure kin-selection, in that it is ‘magnified’ or ‘extended’ in comparison.

Cooperative pyramids only evolve after ‘mature’ social conditions arise. For example, a species cannot evolve altruistic behaivour if organisms rarely come into contact with another.

It is important to note that though very distant relatives might have small weightings, as a large group they collectively become quite significant.

For interest, there is also two less important components:

Oxygen principle – Organisms are less altruistic when they are near death. Once an organism fufills its basic requirements, the risk of extending cooperation is reduced (it probably won’t die from free-riding). A thriving organism is also able to provide better help to others than a struggling one, and so altruism is weighted accordingly.

Zero-sum principle – Likewise, if a situation is unambiguously competitive, and cooperation does not improve collective fitness, then there is no evolutionary advantage at all to extending cooperation. Therefore, altruism and cooperation drop off in situations that are essentially zero-sum (or perceived as zero-sum).

Cooperative pyramids, subject to contingent cooperation, harm minimisation and active deterrance, are dynamic cooperation.

Dynamic cooperation renders altruism defensible enough to be stable at the level of individual organism. In turn, this allows altruism to be further evolved by group-selection. The result is increasingly broad cooperative pyramids – and the existence of highly cooperative, altruistic organisms.

Dynamic cooperation is not cost-free, but providing the organism is intelligent enough to evolve its required components, and provided the benefits of cooperation are outweigh the cost of its defence, it is an effective strategy. Dynamic cooperation restores the viability of cooperation and altruism as evolutionary strategies, allowing them to proliferate in genetic populations.

Dyanmic cooperation also fits well with much of human altruism and cooperation. It resides at the heart of human morality. It explains much of what we see in human moral behaivour and in human ethics. Moral humans look out for one another, cooperate, protect eachother from harm. They offer cooperation based on certain contingencies, such as group membership and familiarity. And, in defence of cooperation and altruism, they take a variety of measures to prevent betrayal and free-riding, such as imprisonment and social sanctions for those hurt others by behaving immorally.

If we visualise human morality as a cooperative pyramid, then the cooperation is often very broad indeed. Humans usually protect themselves and their immediate family first, but they will also cooperate in very large social groups, helping strangers and considering the welfare of other humans in a broad or sometimes universal way. In many cases humans will behave in a way that seems to value non-human life – protecting animals or ecosystems.

It is only possible for humans to have such wide cooperative pyramids because they also have very sophisticated cooperative defences – they are good at dynamic cooperation. They also live in very large social groups that through membership confer groupwide benefits. In other words, humans are subject to the forces of group-selection. It is true that in evolutionary terms this very broad cooperation and altruism is newly formed, somewhat irratic and as yet unrefined. Despite all this, human levels of dyanmic cooperation is one of the most significant processes in the development of life on Planet Earth.

When you sit and contemplate moral philosophy, you are actively taking part in this development. Your morality and your altruism is part of an evolving cooperative quality that is manifesting itself in human society. Where the social conditions are right, that morality is expressed, strengthened, and refined into something truely remarkable in the natural history of this planet.

Next – From Morality to Ethics