Skip to main content

What Can Dolphins Tell Us About The Evolution of Friendship?

This article was published in Scientific American’s former blog network and reflects the views of the author, not necessarily those of Scientific American


A version of this post was originally published on November 18, 2010. Click the archives image to see the original post.

Scientists thought they had a pretty good handle on the social interactions of bottlenose dophins (Tursiops). They've used the term fission-fusion dynamics to describe dolphin (and non-human primate) society and so far it has served researchers well. Fission-fusion societies among dolphins are characterized by two levels of social hierarchy: groups of two or three related males ("first-order alliances") which work together to guard one or more females from other males, and larger teams comprised of several first-order alliances ("second-order alliances") which cooperate to "steal" females from other groups. Since the individuals in first- and second-order alliances are related and therefore share genes, this sort of cooperation can be explained by kin selection.

For six years, Richard Connor and his team studied the social interactions among the bottlenose dolphins of a six hundred square kilometer portion of Shark Bay, Australia. In a paper recently published online in Biology Letters, they describe a new third level of social hierarchy among bottlenose dolphins: a set of alliances among second-order alliances, that they describe as second-order "super-alliances" (or, simpler, "third-order alliances"). The groups which combine to form these third-order super-alliances were often comprised of individuals who were not related, and therefore kin selection can not adequately explain these relationships.


On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.


To test if there was any predictable way in which the different dolphins cooperated with each other, the researchers conducted an analysis on the thirty-four male dolphins on which they had data regarding third-order alliance interactions. It is particularly important to note that all third-order alliance interactions occurred only during conflicts over females and not, for example, during foraging or hunting expeditions. Below are the results of the cluster analysis. I've indicated in the graphic how each group sorts into each of four alliances. The groups were labeled WC (green), KS (brown), PD (blue) and RHP (purple). Each three-letter abbreviation on the far right represents an individual dolphin. RHP is a first-order alliance of three individuals. PD, KS, and WC are all second-order alliances. As the diagram indicates, WC and KS combine, for example, to form a third-order alliance.

Consider the following interactions, which were observed in 2002 and 2006. RHP is a first-order alliance of three individuals, PD is a second-order alliance of seven individuals, and KS is a second-order alliance of fourteen individuals.

2002: Three KS males were approached by the seven PD members and the three RHP males. One PD trio had a female and the other took the female from the KS males.

2006: Four KS males attacked three PD males with a female and immediately the four other PD males and seven of the eight KS males in the area joined the group for totals of seven PD males and eleven KS males. One of the KS trios that joined had a female throughout the skirmish. Aggression (vocalizations and movement) escalated 20 min later when two RHP males (who had a female) entered the group. Four members of the RR alliance and a few immature unallied males were in the immediate vicinity and may have participated. Six minutes after the RHP males joined, the KS males split off but continued to follow the RHP and PD males, who remained together. The following day we encountered a resting group of three PD males and eight KS males. The three PD males still had the same female and the group included the four KS members that initiated the conflict the day before.

But another observation in 2006 showed PD and KS teaming up against WC, a second-order alliance:

2006: The fight between 12 KS, three PD and eight WC males was joined in progress. After the fight, the WC males left with a female, and the KS and PD males remained together travelling.

In two interactions between PD and KS, RHP always aligned with PD, allowing PD to prevail both times. But then when WC launched an attack on KS, PD showed up to help KS defeat WC. A theory of dolphin sociality that was based on reciprocity would not predict that KS and PD would work together, given their history of antagonism. Dolphin friendships, then, appear to be more about alliance formation than about reciprocal altruism. What this comes down to is that males tend to team up with other males in a consistent, predictable way, depending on who else is around and not solely based on prior interactions. And this is critical, as these dolphin brawls have important implications for reproductive success. If this sounds complicated, it's because it is. Not only must each dolphin rank the relative importance of his friendships, but he must also be able to predict how the other dolphins rank him.

Connor and colleagues note that:

Only humans and Shark Bay bottlenose dolphins are known to have multiple-level male alliances within a social network. It is unlikely a coincidence that humans and dolphins also have in common the largest brains, relative to body size, among mammals. Our evidence for a third level of alliance formation in the dolphins should refocus attention on the potential cognitive burdens for individuals embedded in such a system, where decisions at one level may have impacts at other levels.

Is there evidence that human friendships have parallels to dolphin friendships? One prevailing theory in social psychology suggests that people choose their friends on the basis of homophily, which is the notion that people associate with others who are similar to them. This is not a new idea: Aristotle wrote that people "love those who are like themselves," and Plato noted that "similarity begets friendship." C.S. Lewis mused, "Friendship is born at that moment when one person says to another: 'What! You, too? Thought I was the only one.'" The other main theory in social psychology predicts that people choose their friends on the basis of propinquity, or geographical closeness.

The social alliances described by Connor and colleagues are really complicated, requiring big dolphin brains to navigate the subtle social nuances of multi-level dolphin culture. In contrast, the leading social psychological theories of friendship are strikingly simple. Is dolphin society really that much more complex than ours, or are the prevailing theories of human friendship insufficient?

The majority of human friendships occur between individuals who are not related, nor between sexual partners, ruling out kin selection as the underlying mechanism. Reciprocal altruism, as you might expect, is the oft-employed explanation for human friendship formation. In this model, individuals in a friendship each receive benefits in a tit-for-tat, you-scratch-my-back-and-I'll-scratch-yours system of exchange. But social psychologists have found that people do not keep score in close friendships, nor do they resist helping their friends when repayment is unlikely. An evolutionary theory for friendship needs to account for more than just reciprocity.

A new model of human friendship was recently proposed in PLoS ONE, by Peter DeScioli and Robert Kurzban. Participants were asked to list the initials of their ten closest friends (excluding family members and sexual partners), and then to rank them from first to tenth on the basis of closeness. Then, they were asked to divide and distribute one hundred points among each of their ten friends. In the "public" condition, individuals were asked to do so on the assumption that their friends would find out their allocations; in the "private" condition, the allocations would be kept secret. In the diagram to the left are the allocations given in the private condition. On the right is the percent change between the private and public conditions.

In the private condition, individuals gave the most points to their best friend, the next highest amount of points to the second-best friend, the third highest among of points to the third-best friend, and so on. In the public condition, the allocations were made uniformly. The differential allocation patterns between conditions suggests that humans maintain awareness of the implications of the short- and long-term repercussions of the way they interact with others. And the variable that best predicted how the participants ranked their friends? How they assumed their friends would rank them. That is, Carl would place Arnold high up on his list if he thought that he ranked high on Arnold's list.

These experiments suggest that human friendship is driven by something more complicated that simple similarity or familiarity. DeScioli and Kurzban suggest that, for humans,

Friendship is generated, in part, by cognitive systems that function to assemble a support group for potential conflicts...Human conflicts are usually decided by the number of supporters mobilized on each side (rather than strength or agility). ...Therefore, individuals can increase their power by creating and maintaining a network of allies, well in advance, before the onset of an argument or quarrel.

Indeed, this is reminiscent of how relationships are built and maintained between nations. For example, as DeScioli and Kurzban point out, the US and Mexico are geographic neighbors and share economic interests, but Mexico would not rank among America's "best friends." Also, in 2006, the US and China traded three times more than the US did with the UK, though the UK would be ranked higher among America's friends. Alliance-formation explains international relations better than reciprocity does.

If this sounds a lot like the dolphin society described by Connor's team, it should. The friendship literature may need a major revision, thanks to the bottlenose dolphins of Shark Bay.

Update: Evidence just released today (11/18/2010) indicates that the same alliance-building model of friendship has been seen in male Assamese macaques as well!

Connor RC, Watson-Capps JJ, Sherwin WB, & Krützen M (2010). A new level of complexity in the male alliance networks of Indian Ocean bottlenose dolphins (Tursiops sp.). Biology Letters. PMID: 21047850

DeScioli P, & Kurzban R (2009). The alliance hypothesis for human friendship. PloS one, 4 (6) PMID: 19492066

Dolphins image via Flickr/jurvetson.

This post was inspired by Robert Kurzban's brief mention of this study on the Evolutionary Psychology blog.

Jason G. Goldman is a science journalist based in Los Angeles. He has written about animal behavior, wildlife biology, conservation, and ecology for Scientific American, Los Angeles magazine, the Washington Post, the Guardian, the BBC, Conservation magazine, and elsewhere. He contributes to Scientific American's "60-Second Science" podcast, and is co-editor of Science Blogging: The Essential Guide (Yale University Press). He enjoys sharing his wildlife knowledge on television and on the radio, and often speaks to the public about wildlife and science communication.

More by Jason G. Goldman