David Lusseau always wanted to be a biologist. “Well, either biologist or clown,” he adds, “but I realized there was not much money in clowning.” When Marie the dolphin entered Lusseau’s life, she sealed the deal for him becoming a biologist. A bottlenose dolphin (Tursiops truncatus) who swam in the waters near the village of Cerbère on the border between France and Spain in the late 1980s, Marie set seventeen-year-old Lusseau on a path that would one day lead him to study social networks in her species. “When you look in the eyes of a dolphin, you realize there is a lot going on,” Lusseau says, reminiscing on his time with his cetacean friend. “It is something that is very hard to express or grasp or explain in a factual matter, but spending time with [Marie] got me interested in … trying to understand how dolphins work, [in what] I perceived as another intelligent species on the planet.”
As an undergraduate, Lusseau spent time as a research assistant working with a group studying bottlenose dolphins in Florida. When out in the water, he encountered dolphins swimming on their own or in pairs. On occasion he bumped into a trio, but dolphins always seemed to be doing their own thing, just in the company of one or two others. That view of dolphin sociality, or the lack of it, changed dramatically when Lusseau began his PhD research in the late 1990s at the University of Otago in New Zealand. His dissertation focused on conservation biology in bottlenose dolphins in a fjord called Doubtful Sound, but the social behavior of the dolphins there hit him like a ton of bricks. As soon as he got there, he encountered not lone dolphins, duos, or trios, but groups of thirty or more dolphins schooling and moving about in a coordinated manner. These were very different animals from the solo dolphins and very small dolphin groups he had studied in Florida.
Each day Lusseau rose at 4 a.m., grabbed some breakfast, swatted away an endless barrage of midges, and arrived at Doubtful Sound before the sun rose. He’d board a 14-foot boat, locate a group of dolphins, and do focal animal sampling, cycling through dolphins, each recognizable by natural markings on their dorsal fins, often from shark attacks. Doubtful Sound can be stunningly beautiful, but it is at a latitude called the “roaring forties” because of the strong winds from the west and six- to eight-foot waves at times, which make for rough going when watching dolphins from a boat.
As he spent time with the dolphins, Lusseau began thinking about how to understand their complex social dynamics, but he couldn’t quite figure out the best way to proceed. On one of his stints back at the University of Otago, Lusseau recalls reading a Proceedings of the National Academy of Sciences paper on social networks written by physicist Mark Newman and others. Soon after that, he emailed Newman, telling him, “I think you are doing really cool stuff and I can understand it, because you write so well. Would you like to have a look at what we’re doing?” Newman was interested. It wasn’t long before he and Lusseau were coauthoring papers on dolphin social networks. But before they penned any coauthored papers, Lusseau published a 2003 paper in the Proceedings of the Royal Society of London that is widely regarded as the first study explicitly on social networks in nonhumans.
Unlike animal social network papers in today’s journals, where readers are acquainted with how networks operate, to put readers in the right frame of mind in 2003, Lusseau opened his Royal Society paper using a strategy that Darwin had employed in On the Origin of Species. The idea was to introduce a phenomenon that readers already knew about (in Darwin’s case artificial selection, as in selection of different breeds of pigeons) and then make the case that what followed (natural selection), though it appeared radical, was really just another variety of what he had just discussed. In Lusseau’s paper, the opening sentences read: “Complex networks that contain many members such as human societies … the World Wide Web (WWW) … or electric power grids … permit all components (or vertices) in the network to be linked by a short chain of intermediate vertices.” And before readers knew it, they were learning about such social networks in dolphins.
Lusseau constructed dolphin networks based on thousands of observations, and one metric he looked at was network diameter, which measures the average shortest path between nodes. To introduce network diameter to readers, Lusseau first discussed psychologist Stanley Milgram’s “small world” research from the late 1960s. “The global human population seems to have a diameter of six,” wrote Milgram, “meaning that any two humans can be linked using five intermediate acquaintances.” The party version of Milgram’s small world is the parlor game “six degrees of Kevin Bacon.” The rules are simple: players choose a movie actor and then connect that actor to another that they played alongside in a film, repeating the process over and over, trying to link their original actor to movie star Kevin Bacon—who once quipped “he had worked with everybody in Hollywood or someone who’s worked with them”—in no more than six connections. It turns out the dolphin small world in Doubtful Sound is smaller than the human one (including Kevin Bacon’s), both in the size of the network and network diameter, the latter of which is approximately three, meaning any two dolphins in Doubtful Sound can be linked using two intermediate acquaintances.
Lusseau wondered what would happen if the dolphin network was culled by, for example, shark predation. To do this, using the network data he had collected, he built a computer algorithm that simulated predation, reducing the network size 20 percent by randomly removing 20 percent of the dolphins. The small world of the dolphins, it turned out, was unaffected by such a reduction. But if instead of randomly selecting individuals to remove from the network, Lusseau simulated removal of the 20 percent of dolphins who had the greatest number of ties to others, network diameter increased, which had the effect of slowing information transfer within the network.
As he came to know his dolphins better, Lusseau discovered that some individuals in Doubtful Sound give signals that affect group movement associated with finding new resources, including food. Side flopping, in which a dolphin leaps from the water and lands on its side, is seen only in males when they initiate a move to a new location, while upside-downing, in which an individual rolls onto its ventral side and slaps the water to signal an end to a group move, is seen almost exclusively in females. But only a few males do all the side flopping, and only a few females do all the upside-downing. Lusseau wanted to know if a network analysis would shed light on exactly which males and which females. It did. Males initiating and females terminating travel had higher betweenness— they were key hubs in this traveling/foraging network—than their non-signaling counterparts.
In a few populations of bottleneck dolphins on the other side of the planet, in Brazil, signaling and networking is not sometimes about feeding opportunities—they are always about that. And the dolphins have, rather remarkably, added humans to their feeding networks.
For more than three decades, ethologist Paulo Simões-Lopes has been studying dolphin populations in the lagoon systems along the coastline near Laguna, Brazil, about 800 kilometers south of São Paulo. The dolphins in nine populations along that stretch do something that no other dolphins—and almost no other animals, period— do. They not only network with each other, but cooperate with humans to secure more food for both themselves and their primate partner.
Each autumn, a huge mullet migration takes place in southern Brazil. Both the dolphins and the fishermen see the fish as prize prey. Up to fifty fishers, wading waist deep in very cold water, wait for the chance to cast large circular nylon nets called tarrafa over schools of mullet. The problem for the fishers is that the water is murky, and it is next to impossible to see the fish. The problem for the sixty or so dolphins at Laguna is that compared to their other prey, mullet are large and hard to catch. But dolphins aren’t especially troubled by murky water, as they detect mullet using echolocation, a built-in sonar system that would be the envy of most engineers.
Dolphins produce sound waves in their nasal sacs and focus those waves through fatty tissue and fluid in their foreheads. Once the sound waves are shot out into the water, they travel until they bump into an object, at which point they bounce back to the dolphins, who use their lower jaw as a receiver. From the lower jaw, the waves travel to the inner ear and then to the brain. Objects of different sizes and densities reflect back sound waves of different frequencies, and the dolphins use that information to “see” what is in the water around them. When their sonar detects mullet, dolphins signal fishers that the fish are present by curving their backs and then slapping their heads or their tails on the water surface. The fishers then cast their tarrafa and pull in loads of mullet. The confused mullet who escape the tarrafa often swim right into the mouths of waiting dolphins. It’s the perfect win-win situation.
Laguna newspapers from the late 1890s featured articles about this dolphin-human mutualism, and so Simões-Lopes knows that, at the very least, it has been going on for more than 130 years. And though many dolphins don’t signal fishers, every fisher knows which dolphins do. “It is famous [in southern Brazil],” Simões-Lopes says. “I grew up watching those dolphins … I would sit on a rock in the canal and watch for hours. I knew it was unusual … I knew there were dolphins in a big harbor farther south where dolphins and fishermen don’t interact.”
Today Simões-Lopes has a team of ten working with him, but he began on his own in 1988. Soon thereafter, he entered a PhD program and built his dissertation around his research on the dolphin-human foraging mutualism. Each day he brought a folding chair with him and set it up on a rock, watching the dolphins through his binoculars, taking photos—he had compiled a mug book with photos of all the dolphins in the lagoon—and filling notebook after notebook with data on dolphins signaling fishers.
Simões-Lopes began to know the fishers, and they began to know him. He also was starting to get a good feel for which dolphins at Laguna signaled the fishers and which did not. Not surprisingly, the fishers also kept tabs, telling Simões-Lopes about the “good dolphins” (who signaled fishers) and the “bad dolphins” (who did not). The fishers know not only which dolphins signal, but which dolphin will give which signal: “Each dolphin gives the signal in a different way,” one fisher said, “and we need to know [the different signals] in order to catch the fish.” Another fisher was more of a romantic, telling Simões-Lopes and his colleagues, “This is beautiful. It doesn’t happen everywhere.”
The more that Simões-Lopes thought about those “good” dolphins and “bad” dolphins, the more he wanted to understand them better. Years later Mauricio Cantor joined Simões-Lopes’s team; Cantor had worked with Hal Whitehead, a leader in early social network analysis. Simões-Lopes and Cantor decided that a network analysis might help them delve deeper into the between-species cooperation they observed on a daily basis. In 2008, they contacted David Lusseau, who had done the network studies on bottlenose dolphins in New Zealand, and asked if he would be interested in serving as a sort of conceptual consultant specializing in social networks. Lusseau was more than happy to join their team.
This article appeared in Skeptic magazine 29.3
Buy print edition
Buy digital edition
Subscribe to print edition
Subscribe to digital edition
Download our app
Simões-Lopes and his team assumed dolphins learn how to signal humans from other signalers they associate with, so for their social network analysis, they were especially interested in whether signaling dolphins preferred spending time with other signaling dolphins, both when they were chasing mullet into nets and, just as importantly, when they were not. To test whether there were cliques of signalers and cliques of dolphins who didn’t signal, Simões-Lopes’s team looked at clustering coefficients of sixteen cooperators and nineteen dolphins who did not signal and cooperate with fishers.
What they discovered were three cliques within the larger network of the thirty-five dolphins. Clique 1 had fifteen dolphins: each and every one of them cooperated with the local fishers. Dolphins in this clique associated with one another not just during the autumn mullet fishing season but the rest of the year as well. Clique 2 had a dozen dolphins, none of whom cooperated with fishers, and dolphins in this clique were not as well connected to one another as the individuals were in Clique 1. Clique 3 was made up of eight dolphins: seven never cooperated with fishers, but one—dolphin 20—did. And of all thirty-five dolphins in the network, it was dolphin 20 who spent the most time interacting across cliques, acting as what Simões-Lopes and his colleagues call a “social broker” between the signalers and non-signalers.
This behavior is all wonderfully complex, and we humans—and I don’t just mean the artisanal fishers of Laguna—should be grateful to play a role in understanding it. 
Excerpted and adapted by the author from The Well-Connected Animal: Social Networks and the Wondrous Complexity of Animal Societies by Lee Alan Dugatkin, published by The University of Chicago Press. © 2024 by Lee Alan Dugatkin. All rights reserved.
About the Author
Lee Alan Dugatkin is an evolutionary biologist and a historian of science in the Department of Biology at the University of Louisville. He is the author of sixteen books and more than 200 articles in such journals as Nature, The Proceedings of the National Academy of Sciences, and The Proceedings of the Royal Society of London. Dr. Dugatkin is contributing author to Scientific American, The American Scientist, The New Scientist, and The Washington Post. His latest book is The Well-Connected Animal: Social Networks and the Wondrous Complexity of Animal Societies.
This article was published on November 1, 2024.
