Thousand-Mile Song

Sperm whales make two kinds of clicks: regularly spaced, even ones, whose function is generally believed to be echolocation; and quieter, uneven rhythms, which were originally hypothesized to be signature sounds that identified individual whales. These are called “codas” (even though they don’t always come at the end of something). On the first expedition aboard the Tulip, Whitehead and Gordon did not pay too much attention to coda clicks, since they happen only when the whales are socializing, and that is only about 20 percent of the time. That study ended in the mid-eighties because of the Sri Lankan civil war. Gordon went into conservation, and Whitehead married a whale-obsessed woman named Lindy Weilgart, who became his companion on a series of journeys across the Pacific Ocean, between the Galapagos Islands and the coast of Chile. The couple knew from old records that sperm whalers had done well there in the past.

As their research progressed, Weilgart and Whitehead were able to create a lexicon of sperm whale beats, analogous to the catalog of cries and shrieks assembled by John Ford for the Vancouver Island orcas. In the South Pacific they identified thirty-three different coda types, all little rhythmic riffs, from three regular beats, to two beats, a rest, and a third beat, to four beats, a rest, and a fifth, to seven irregular beats. The catalog of patterns makes quite a comprehensive chart, as shown in figure 30 on the preceding page.

From a musician’s standpoint this looks like a list of possible rhythmic patterns, say the range of beats that an Indian tabla player might use, or a Latin percussionist playing timbales. The scientists, however, didn’t know what to make of this information, since it seemed so anomalous. These clearly weren’t echolocation clicks: they were neither extremely loud nor directionally focused in a way that might help locate prey. Nor could they be signature sounds, which would identify individual whales, since there are far too many whales, and far too small a number of different codas to reliably identify a single whale. “They’re very strange,” says Whitehead. “They have a clear and distinct pattern but they’re very simple, at least as we hear them. It makes them easy to study, to quantify and to organize.” So what were they for?

Weilgart and Whitehead sailed around the South Pacific, trying to discover how sperm whale behavior, and especially codas, varied over vast distances. “This was a lovely excuse to spend a year on a boat with our family,” said Hal. For much of the trip, they saw no whales at all, even in areas where whalers always found them. They didn’t expect that. The two researchers also imagined that the codas in the eastern part of the Pacific would be different from those in the west, due to genetic differences. But when they did hear whales, Whitehead and Weilgart heard something unprecedented.

There were groups of whales scattered all over the Pacific that seemed to be identifiable by virtue of the kinds of clicks they favored. Tissue sampling showed the whales in any one group shared mitochondrial DNA, suggesting they were related on their mother’s side. But this is not enough genetic similarity to have inherited behavior. Instead, it suggests that sperm whale clicking behavior is indicative of a matrilineal sperm whale society—the whales seem to learn their group characteristic clicks from their mothers and their mothers’ relatives. Each group consists of one, two, or three “units” of perhaps eleven animals each. The whole group includes up to thirty-five animals in total.

While on the water, Hal tends to collect the data and not analyze anything prematurely. “There are people who are very good at understanding what animals are doing in the wild, but I’m not one of them. I go out and collect the data, then take it all home. I can only pick up this stuff through analysis much later.” This explains why it can take up to a decade for these whale science papers to get into print.

When he got to the stage of charting and tabulating which units used exactly which codas, Whitehead found a surprising pattern of organization at a higher level: certain units seemed to share a definite repertoire of clicks. Whales separated by thousands of miles knew the same codas.

“When you listen to these sperm whale groups,” notes Whitehead, “you find they are genetically unrelated, they are all female, but they have the same codas. This posed a dilemma that we did not fully understand. The groups are only together for a short time, and they’re made up of two or more independent units of whales. So how could they all have the same sounds?” They must possess the ability to learn rhythms, the beats that you need to be part of the band.

“Of course, how could they not? The codas are made almost exclusively by the females. The females have this whole social system that they have to mediate. With other toothed whales like orcas and belugas you have a wider range of sounds, but the same division into sonic groups. But only sperm whales have this matrilineal society where the males are chucked out.”

That’s how Whitehead and Weilgart identified twenty units of sperm whales throughout the South Pacific, which they named “A” through “T.” Different units would associate together to form temporary groups, and each unit could most easily be distinguished by the coda clicks they were making.

“As I tabulated the data I was just blown away!” It made more sense than Whitehead expected. “I found these two sets of units with very distinctive coda repertoires, which is something we had never noticed before.” Groups of whales that shared the same mother made similar sounds. The kind of click favored by a group could identify its matrilineal family. Weilgart and Whitehead named this the “acoustic clan,” similar to what Ford and Spong found with orcas, but with clicks, not cries. Figure 31 shows the kinship chart that shows how the initial twenty units fit into three basic clans, distinguishable by sound.

Although it resembles a fingering chart for a flute, this diagram shows which particular patterns are favored by which units of whales. Some units are acoustically identical, but they were found separately. The filled-in dots reflect codas that make up more than 10 percent of that unit’s repertoire.

The most basic analysis shows that the units on the left favor regular rhythmic codas, mostly 5 beats, 6 beats, 7, or 8. These were named the “regular clan.” The next set of units favors a little unevenness in their phrases, especially 4+1, 5+1, and the syncopated 1+7. This was named the “+1” clan. The last unit doesn’t fit in anywhere, but it turned out to relate to other units found thousands of miles away.

The clans span vast areas of deep ocean water, and in any one area two or more clans may overlap. Further analysis revealed that sperm whales tend to associate only with other units in their clan. Each clan has other distinguishing behaviors besides sound: the regular clan, found in the eastern tropical Pacific and also off of Chile and Ecuador, tended to travel tortuous paths and dive synchronously. The +1 clan, found mostly around the Galapagos, tends to swim more directly, and they do not synchronize their diving. Eventually Weilgart and Whitehead were able to identify four distinct acoustic clans by combing through all their South Pacific data.

What are the implications of all this coded behavior? That sperm whales, who learn these clicks at the level of the clan, group, and unit, possess a form of culture. Whitehead believes he can explain why the sperm whale has so large a brain: “I don’t think it’s for making clicks,” he grins. “The case and junk are big enough for that. But the brain may be for processing clicks, because it has to detect the click out of a lot of noise. As a percentage of body weight the sperm whale brain is not terribly big—the bottlenose dolphin is second only to humans on that statistic.”

He makes an even wider speculation: “Perhaps the sperm’s brain must be large to deal with social and cultural information. They live in complex societies. A lot of what they do depends on learning from each other. Social learning is something that we as humans do all the time. Some of us are a lot better than others, and I suspect the same thing happens in sperm whales. There’s an enormous selection process to do well, so maybe that’s how the brain got so big.”

“I think they do. It may be as simple as ‘we’re best buddies,’ but it may be something more. The mother of the calf has a somewhat different coda, maybe they have to do with raising the offspring. We don’t really know. The coda is a reaffirmation that we are from the such-and-such a clan. But why would they need more than one coda for that? This is what we need to work on next.”

Whitehead first published his hypothesis in the journal Behavioral and Brain Sciences in 2001, with an article written together with his student Luke Rendell. The notion that sperm whales could have different cultures, based on a barely perceptible difference in the tiny rhythmic clicks of separate groups, sent a revolutionary challenge to the world of animal behavior studies. Differences scientists could hardly perceive in the field were being put forth as evidence of learned, social behavior in a huge marine mammal that we had previously only taken seriously as a source of high-grade oil. Whitehead and Rendell suggested that those who study marine mammals might consider their work more like ethnography than zoology, where each population studied is a distinct society with its own unique, defining behaviors—like a tribe.

The argument for culture in cetaceans is quite simple. Culture is basically defined as a set of behaviors that are learned by and maintained by a group of animals, independent of or not determined by genetic inheritance. If a pod or clan of whales learns certain ways of living from the others with whom it associates, and maintains those qualities independent of environmental factors, then those attributes are culturally determined.

Some groups of killer whales hunt mammals in packs, others more calmly eat fish. They live in the same areas, but with different lifestyles. They are the same species, but they have learned and more importantly maintained these differences over generations, even though they interact with and acknowledge the other populations of whales and their ways of life. Pacific humpbacks sing their songs while suspended fifty feet under water, while humpbacks in the South Atlantic, as observed off of Brazil, stick their tails in the air and wave them gently back and forth as they perform their great serenades (makes them a lot easier for researchers to find).

Cetaceans can imitate the behavior of other animals. A wild dolphin can hear another dolphin’s signature whistle and copy it right back. Individuals of many whale species can pick up new behaviors and then teach them to other members of their pods. A new humpback feeding behavior was observed off Maine, and scientists were amazed to see it picked up by all the whales in a matter of weeks.

The article was extraordinary enough to be published with thirty-nine separate commentaries and critiques. Anthropologist Jerome Barkow went so far as to say that cetacean and animal culture, with its rather simple learned behaviors maintained by a social group, could be called “normal” culture, while human culture, with its vast array of learned traits, is really a kind of “hyperculture” unmatched in the natural world. Evolutionary psychologist Robin Dunbar disagreed. He doubted that cetaceans can share meaning in a cognitive way, so he doesn’t think “culture” is the right word to describe how they pass behavior on from one generation to the next.

More supportive comments tended to suggest that there is much more culture in the animal world than Rendell and Whitehead realized. Parrot expert Irene Pepperberg noted that her birds (especially the late, great African grey parrot Alex) certainly possessed social learning ability, in the wild and in captivity. The astute bird scientist Peter Slater pointed out that the question is not whether or not whales have culture, but why certain species have learned social behavior and others have not, since natural selection doesn’t seem to adequately determine this quality. Primatologists David Premack and Marc Hauser argued that we need to figure out which differences in the behavior of different groups of whales are trivial and which are not. What do the differences in humpback song and sperm whale clicks actually mean? That’s what we should focus on, to separate arbitrary variation from real differences in culture, like the orcas’ group choice to hunt either mammals or fish.

If cetaceans really possess culture, then perhaps human cultural approaches might actually help us understand them. Sperm whales certainly click, often in regular rhythms, and sometimes in irregular rhythms. Individual differences in their sounds have been mapped out by Whitehead, Rendell, and Weilgart. But most of the time these clicks overlap upon each other, forming a confusing jumble of rhythms, difficult for us to make sense of. Difficult for scientists, at least. Not so tricky for someone skilled at following a jumble of beats going on at the same time. Who might that be?

While Whitehead and his team were reporting the amazing range of sperm whale rhythms, French scientist Michel André, working in the Canary Islands, had some questions. André, while listening to sperms clicking throughout the day, noted that codas form a very small portion of their repertoire. Most of the clicks are regular beats, what Whitehead and most researchers tended to call echolocation clicks.

But surely, thought André, the whales would need to communicate to each other throughout the day, not just occasionally. The regular clicks, he surmised, ought to have a communicative function as well. These more prevalent regular beats might be more than echolocation clicks. They are regular beats at differing tempos, overlapping rhythms, all at different speeds, fitting together. Listening, he could make no sense of it. Mathematically analyzed, it was still too confusing.

André remembered that European musicologists, when first visiting Africa, could not understand how a single musician in the midst of large groups of drummers in countries like Senegal could keep their own part going in the presence of so many other contradictory beats. In fact they’ve been maintaining their individual signature rhythms in the din of the crowd since childhood. From years of practice, each drummer knows how his pattern sings out in the spaces between all the other patterns. One must be an expert in the discernment of rhythms to successfully play this music.

With this in mind, André invited Senegalese drummer Arona N’-Diaye Rose, one of the many sons of the great Doudou Rose who intuitively grasped the signature whistles of dolphins a few years before with André’s colleague Cees Kamminga in the Netherlands. Arona Rose listened to a recording of a four-member unit of vocalizing sperm whales. The sabar master was immediately able to distinguish the beat of each of the four clicking whales from the others. He also believed that what the scientists heard as cacophony was actually an organized rhythm, based on a dominant beat coming from one of the whales, which Rose felt was analogous to the signature rhythms marking the social structure of an African tribe.

“I couldn’t believe it,” said André to me at a café in Barcelona, close to his laboratory. “We knew there were four whales because we took notes during the recording, but all we heard was a confusion of clicks. I asked Arona how he could tell there were four different animals. He said ‘I don’t know how, but I know.’”

When I first heard this story I had to smile—it is just the kind of wondrous tale that makes me hope that music and biology might work together. I passed it on to my friend Andrew Revkin, science reporter for the New York Times, and pretty soon it appeared in many newspapers around the world. Nature needs culture. Science needs art.

Animals spend a large amount of time making calls back and forth to each other, in pairs, groups, or all at once, in the striking dawn chorus of birds or a lek of chorusing frogs. As common as such behavior is, it has hardly been studied. We don’t have the tools to accurately analyze the relationships between many simultaneous noises, and we tend to focus on the sounds, not the spaces between the sounds. But that’s what rhythm is all about: silence, and the marking of pulses in the wake of the silence.

Musicians whose ears and minds are tuned to hear layer upon layer of overlapping beats as music, not noise, may be well trained to pick out the order behind the overlapping mess of sperm whale codas. Whitehead took the first step and identified the different kinds of codas. Now André and Rose want to study the relationship between each pattern, not with machines, but through an education in polyrhythmic listening.

Since that session with Rose ten years ago, André has been seeking funding to continue this work. But it’s the same story many scientists have told me: it is always difficult to get support for descriptive work. Applied science, especially work toward managing whale “stocks” or populations, is always the easiest to fund. Something that combines biology and music, for all its intercultural promise, is harder to support. Will it be funded by research agencies or cultural exchange groups? Neither wants to touch anything so clearly on the charged border between one approach and its opposite.

Some scientists are suspicious of this news from the Canary Islands that has been heard ’round the world. Anything smacking of music is too subjective to be of much use to science. “Michel André?” Hal Whitehead shakes his head. “I have no idea what he’s talking about. But Luke says he understands him.”

I asked Luke Rendell what he thought of André’s work, and he was far from enthusiastic: “André analyzed just eighty-five seconds of data, producing periodograms of click intervals from just three whales. That’s about a hundred clicks from each whale, from an average per dive production of two thousand. . . . You simply can’t characterize a communication system as complex as the sperm whale’s with so little data.” Plus, nearly all sperm whale researchers are convinced that the “normal” sperm whale clicks are primarily for echolocation purposes, and not for the cultural rhythmic activity Rose hears in them. In fact, this paper, for all the inspiration it has given those like myself and the world media, has only been cited twice within the sperm whale research community. Those in the know have not been impressed.

But André will not give up. Look, he points out, in the Canary Islands, we rarely heard codas when the whales were resting and socializing near the surface. The echolocation clicks were heard constantly, so perhaps they have a social role as well. “The codas, until now defined as the principal acoustic support of communication in sperm whales, actually seem to be only a part of a whole system articulated by the continuous emission of usual click trains, which are the sounds usually made by the whales as they socialize. The famous codas are barely heard!” As to why he only looked at eighty-five seconds of clicking sounds, André says there was enough complexity in even such a small sample to illustrate his point, but the critique is fair. “We’re currently working on these very issues, and the next paper is in the works. I am convinced rhythm is very important to these animals, even in the codas.”

André has continued to publish on the rhythmic structure of codas, suggesting that there is much more variation in pattern than the Canadian team would admit. He is concerned that Whitehead would report two click trains as being “3+1,” three regular and one irregular, even if the interval between the three regular clicks was different in each case, say, one triplet faster than the next. “I’ve talked to Rendell and Whitehead about this, and they realize there is a basic flaw in their analysis.” André has been trying to come up with a more mathematically rigorous way to categorize clicks, as a prelude to being able to study the relationship between clicks coming from different whales with greater accuracy.

It would also be a method that takes account of the precise rhythm going on between the clicks, how they fit into a larger patterned context. This is what Rose heard happening, and André feels we should not ignore it. “We need to study the whales’ perception, not our own perception. Scientists are more used to counting, so we count. We have to learn from the insights of African drumming to perceive the value of rhythm at work in the clicks themselves.” So at the same time as trying wild ideas like working with drummers, he is trying to develop better and more precise mathematical tools.

It is easy to choose equations that show you what you most want to see. Whitehead admits his team has been so busy trying to identify the individual coda patterns that they have not spent much time studying the relationship between codas. In his wife Lindy Weilgart’s thesis, there is brief mention of the occasional phenomenon of “echo-codas,” where one whale clicks a pattern, and another copies it exactly, a tenth of a second later. But this was never investigated in greater detail, and they reported it only tentatively: “The ordering of codas within exchanges is to some extent non-random, suggesting conversations, but we do not know what information is being transmitted.”

André wants to spend more time analyzing the spaces between sperm whale clicks. It’s going to take musicians, scientists, and whales spending a lot of time together. “Sure, it’s subjective if a drummer just listens once, but if I ever get to work with Rose, for several months time, learning his perception, and his approach, to analyze the combinations, then I hope to learn something from his rhythmic intelligence that has been passed down through many generations. Yet we still don’t have the funding to bring a drum master onto our team. Rose was certainly on to something, he immediately sensed a musical sense of culture that Whitehead and Rendell had to spend years in the laboratory to uncover.”

The very idea that sperm whale tapping makes sense to both musicians and scientists should clearly signal its importance as a code worth deciphering. At the moment, only about ten people in the world have the slightest understanding of these giant beasts’ code-like tappings. How to figure them out? The problem with whale science is the same problem Brian Eno pointed out ten years ago about the world of digital communication: “It’s got to have more Africa in it.” This is a perfect opportunity for a second generation of musical scientists, and scientific musicians, to delve once more into the mystery of whale sound. We know both these methods can help us cross the border into the cetacean world.

In July 2007 I attended the Third International Workshop on the Detection and Classification of Marine Mammals Using Passive Acoustics, put on by the United States Office of Naval Research and the New England Aquarium. In preparation for this gathering, signal processing specialists, many of them physicists and engineers, not biologists, received a data set of clicking sperm and beaked whales that they proceeded to decipher, using algorithms of their own invention. Who would turn out to be best at separating one clicking whale from another, and tracking their paths through the sea? Would mathematics be more accurate than human observation?

Paul Baggenstoss of the Naval Undersea Warfare Center tried to separate superimposed click trains, “seeking to minimize an aggressive criterion of optimality.” He computed the distance between every pair of adjacent clicks and looked for numeric anomalies. George Ioup, from the University of New Orleans, discovered that there is a common spectral pattern to each coda of sperm whale clicks, making it possible to identify them automatically using algorithms called self-organizing maps, which can scan the data set to figure out which whale is producing which coda, and thereby aid in estimating the total number of whales.

When I mentioned to these two researchers the possibility of using similar approaches on African drumming, and the notion that drum experts might be able to detect click patterns with their heightened sense of rhythmic attention, both scientists were extremely interested, much more so than most biologists I had met. Signal processing analysis is much more like a game or a puzzle, and there is a sense of tinkering and playfulness to the whole field, in military and academic worlds alike. So perhaps the Fourth International Marine Mammal Detection Conference will invite some Senegalese drummers. If they invite clarinetists who jam with whales, it’s going to be active, not passive, acoustics.

Believe it or not, there are a few examples on record of humans and cetaceans cooperating to form an interspecies sense of culture. In the town of Laguna near the southern tip of Brazil, fishermen have developed a unique method of fishing for mullet that works only with the help of wild dolphins. The water is especially turbid, and it is impossible to see the fish. About forty fishermen head out each day, with one to four dolphins assisting them. The men position themselves in a straight line parallel to the shore, each one holding a circular nylon throw net. The dolphins swim several meters farther into the ocean than the fishermen, moving back and forth on the surface. A dolphin suddenly submerges, diving seaward, and the animal comes up in a few seconds, heading back toward the line of fishermen. Just out of net range he dives deep down, and the fish he’s been chasing head straight into the nets. “The men rarely cast without a dolphin’s cue, and the fishing method seems to be initiated by the dolphins,” say Karen Pryor, Scott Lindbergh, and Raquel Milano, who wrote this up in Marine Mammal Science in 1990. After the men have caught what they need, the dolphins eat the addled fish that remain. This behavior, passed down from generation to generation of people and dolphins, has been going on since 1847.

An even more remarkable story comes from the town of Eden on Twofold Bay by Australia’s Pacific coast. For more than a hundred years the Davidson family hunted humpback whales from small rowboats with the help of a pod of killer whales, who learned to alert the whalers to the presence of large baleens by flopping their tails rapidly on the waves. The orcas would work together to round up the confused prey, aiding the humans like a second whaling boat would. Together orca and human would make the kill. The orcas, known as the “Killers of Eden,” would be rewarded by getting to eat the tongue and the lips of the humpbacks. The rest was left to the whalers, who used the blubber to make soap, pencils, candles, and crayons. The leftover meat was turned into pet food. This arrangement was taught from generation to generation of humans and whales, all the way until the 1930s, when the last killer whale in the pod, Old Tom, died.

These tales demonstrate that whale culture can aid human culture in a most practical way. Today, whale voices continue to inspire us artistically. Jim Nollman, together with Sam Bower, the director of the environmental art web site www.greenmuseum.org, sent out samples of beluga, dolphin, and sperm whale sounds to about fifty electronic musicians. Seventeen of their pieces, from ten countries, were chosen to be released in 2006 on Belly of the Whale, a compilation from Important Records.

Kim Cascone, who once worked as sound editor for David Lynch’s Twin Peaks, created a mix of rich ambient booms and alternating rhythmic clicks, evoking something far-off and swirling from the source material of whales. The British electronic composer Scanner, who also composed an anthem for the European Union, combined the restless patter of sperm whale clicks with his characteristic grave minor synthetic harmonies, overlaying beluga shrieks that turn into buckets of noise. Finnish electronica wizard Petri Kuljuntausta turned the sperm clicks into a rapidly pulsing loop that resembles the thrum of insects. Nollman himself added a carefully remixed version of his guitar twangs charming up orcas, perhaps his best track yet.

I was honored to be invited to play on this disk too. For my contribution, I decided to build a rhythm entirely out of sperm whale clicks, to provide a musical analogy to what I thought Michel André was talking about—a unique submerged tribal groove—upon which I then played the bass clarinet, trying to enter into the secret polyrhythmic channel of the whale. The version on the record is all instrumental, but the original includes spoken lyrics, intoning the whole story:

In the year 2001, Atlantic Ocean, Canary Islands
two scientists hear the clicks of singing sperm whales
calling to each other down under the sea.

Even with all their machines measuring the sounds
and the times between the sounds
no one can figure it out.

So they call in an African master of drums
He listens to these clicking whales
Knows how to pick out one beat from the next

He says, “I hear the beat
There are many beats
They are all beating ’round
the sum of one rhythm.”

And the master smiled and he said
“Well, there is a way to get inside the whale.”
Inside the whale

Like George Orwell said in 1939
“Face it, you are inside the whale.”
Stop fighting, or wishing, you control it.
Just accept it, endure it, record it.

How does it go? They ask
What do they say? They wonder
Where does it start? They murmur
How does it end? How does it end?

Inside. That’s what it sounds like.
Inside the whale.

“No words on this disk,” said Nollman. “These lyrics have to go.” That guy is always giving me a hard time! Perhaps the less said the better when it comes to deciding what whale sounds mean. What if the sperm whales are making music themselves? This may be wishful thinking on my part. The reason for the rhythms of their clicks is still unknown, and might never be found.

Each click only makes sense in the presence of other clicks. They bounce from whale to squid, surface and bottom and back. There is rhythm in our breath, our movement, and the way the deepest questions repeat in our heads, over and over again. The whales may be wondering, too. The ocean reflects and absorbs their inscrutable rhythms.