Delphinus delphis, also known as the "saddleback" or "common dolphin."
Photo by William A. Watkins. Courtesy of Woods Hole Oceanographic Institution.

Marine Mammal vocalizations:
language or behavior?

by Kimberly Amaral

Go to "Marine Mammals' Greatest Hits"

For decades, scientists have been listening to the seas--listening to the otherworldly "clicks" and songs of whales. It began in 1949, when William E. Schevill and B. Lawrence dropped their hydrophones (microphones designed for use underwater) into the Saguenay River of Quebec, making the first recording (~500K) of an identified species--the white or belukha whale--in the wild. Since then, marine mammal scientists have been anxious to accurately record and understand the nature and purpose of these utterances.

Preliminary research focused on whether or not these creatures could "talk." Most people are aware of the high intelligence of these creatures. Could they be using it to communicate with each other much like we use language?

Peter Tyack, a researcher at Wood's Hole Oceanographic Institution (currently working at Stanford University), doesn't think so--at least not yet. He says a lot more research has to be done before we can make such a determination. So far, most research has not investigated the most important aspects of language--is it cognitive and "aware," or is it simply a learned response for survival? Says Tyack, "It may make no more sense to compare these [animal] songs to language than to compare the marks on a peacock's tail to some strange hieroglyphic writing." Instead, scientists are now looking at marine mammal vocalizations as an indicator of behavior, and even for recognition of each other.

First, a primer on some of the different types of whales. Whales, dolphins and porpoises all belong to the taxonomic order Cetacea. All cetaceans more than several meters in length (such as the humpback) are called "whales." But the toothed whales such as the sperm, killer and pilot whales are much more closely related to dolphins and porpoises than to the baleen whales. Dolphins, porpoises and toothed whales are all classified together as odontocetes. As the name indicates, all odontocetes have teeth. Baleen whales do not have teeth, but instead use baleen to filter small prey from sea water. Baleen whales also have two external blowholes while odontocetes have only one. Beyond these physical differences, scientists have uncovered striking differences between the two groups in terms of life history and social organization. This appears to be reflected in their communication.

Marine mammals "vocalize" in a variety of ways, each of them suited to a particular behavior or situation. Dolphins, for instance, exhibit two main types of vocalization: clicks (~80K) and whistles (~120K). The "clicks" are used in echolocation to find food. Each individual dolphin also has a series of whistles (like a Morse code) distinct from any other member of the group called a "signature whistle." This signature whistle distinguishes an individual, providing a way for dolphins to recognize and bond with others.

Marine mammals are also very adept at imitating sounds. Hoover, a harbor seal at the New England Aquarium, imitated human speech well enough to have a recognizable New England accent. Logosi, a beluga whale at the Vancouver Aquarium, was able to imitate his own name. And when one dolphin of the Sarasota community, Nicklo, was carried onto a raft to be measured and recorded, it imitated another's signature whistle. That of Granny--the oldest dolphin in the group, and perhaps the one most familiar with this new situation or most able to help her.

Some studies have shown that male dolphins might be better at imitating sounds than female dolphins. The signature whistles and their imitations of two captive bottlenose dolphins named Scotty and Spray at Sealand, a marine park in Brewster, Mass. were compared. Not only does a male calf in the wild tend to learn his signature whistle by imitating his mother, but also the imitations of adult males appear more precise than those of adult females. The captive male Scotty produced more frequent and more accurate imitations than did the female Spray.

Common dolphins. Photo by William A. Watkins. Courtesy of
Woods Hole Oceanographic Institution.

Signature whistles also differ between the sexes. Female dolphins generally develop whistles very different from their mothers, while male dolphin signature whistles tend to be very similar to that of the mother. Again, this relates to social behavior, as odontocete groups are typically formed of females with their young, sometimes spanning several generations. If females had signature whistles very similar to their mothers, the members of the group would have difficulty distinguishing between the two. Males, on the other hand, leave their natal group when they mature and form juvenile groups, which may also contain juvenile females in some species. In many toothed whale species, adult males may associate with female groups for only a few days at a time. They tend to leave the well defined population boundaries for periods of several months.

Like other toothed whale species, sperm whales also form very stable groups. They produce individually specific sounds--but these are very different from the whistles formed by dolphins. Their specific sound takes the form of a short series of clicks, called codas (~180K). Also like dolphins and their signature whistles, sperm whales can mimic the codas of others.

Killer whale family groups, called pods, are the most long-lasting of the odontocete groups. Individuals around Vancouver Island, British Columbia have been identified as members of about 30 groups for over 13 years. These groups seem to be made up of related individuals. With low birth and death rates, group composition often does not change for several years. Killer whale groups are so stable, they have been found to produce a dialect for a sound specific to their group. Each killer whale group has a different repertoire of calls, and apparently each individual within the group produces each call. "Each pod has about a dozen...calls that thay use over and over," said John Ford in a recent televised interview. Ford is the premiere scientist studying killer whale pods in Puget Sound. He has found that pods "share" calls, such as when the "L pod" uses a variation of a "J pod" call.

Although killer whales have not yet been found to produce an individual signature whistle, scientists in British Vancouver have found differences in dialect between transient killer whales (who roam the seas) and whales that stay close to shore or in one particular area.

Baleen whales, on the other hand, are solitary animals. The most stable bond is between a female and her calf, and this lasts less than one year. Humpback whales off Newfoundland feed in groups that are seldom stable for more than a few hours. They have several methods for breeding access. They may join in large groups to fight for access to one female. Or, if a male is alone, it may produce a long, complicated vocal display called a song (~800K). The song consists of a series of notes and lasts up to 20 minutes before repeating. Since humpback song is sung by males primarily during breeding season, it is presumed to be a song of seduction. Humpback songs also change gradually throughout the singing season. Sounds may change in pitch, duration and timbre. Specific sounds may disappear from the song entirely, and new sounds may appear. Over a twenty year period, entire songs are slowly transformed. In contrast, signature whistles of dolphins are stable over periods of many years. Tyack compares this change to the sexual appeal of a bird with a large repertoire of songs.

How do they hear that?

Sound travels through water a lot differently than it does through air. Because water is relatively more dense, sound travels through it very easily. So easily, in fact, that it moves five times faster (at a temp of 20 degrees Celsius, sound travels through ocean water at 1,450 m/s as compared to 334 m/s in the atmosphere). Velocity also increases with increasing salinity and temperature. Although pressure increases steadily from the surface of the ocean to the bottom, the generally dropping temperature below the thermocline (the buffer zone between the upper layer of water and the frigid ocean below) more than offsets this. Because of this, an area of low-velocity sound transmission exists at the base of the thermocline. The refraction of sound waves then causes sound to be trapped in the zone, called a SOFAR (sound fixing and ranging) channel. Marine mammals may use this phenomena to a much greater extent than we presently realize as a means of long-distance communication. Other animals in the ocean also take advantage of this phenomena to transmit sound.

Not only does sound travel farther and faster (with plenty of interference and refraction), it also travels in all directions! The way that we may identify the direction of sound in air is completely lost when we enter the ocean. Instead, we hear sounds within our own heads. Marine mammal's "ears" are thus very different from ours.

Because the ocean environment (and the tissue from which sounds are emitted) change the vocalizations, the sounds that marine mammals hear may be very different from how we interpret sound.

"Whatever these animals use sound for, the ocean environment is so different from the air that we speak. It's highly unlikely that they structure language the way we do," says Kirt Fristrup, a WHOI research specialist. "It's under question whether or not 'voice' exists underwater. It could very well be that pressure in the water actually masks any difference in the (voice-making) apparatus--in the tissue that's producing the sound. Animals may not have 'voice' or individualistic sounds...It's probably misleading to expect that the quality of marine mammal sound is anything like the quality of human voice."

Instead, Fristrup says, the mammals identify through those "learned codes" or "learned sequences" of signature whistles and codas. "So that the signature whistle pattern that an animal actually learns and uses it to distinguish. So the pattern carries the identity rather than the sound."

How did we hear that?

In a relatively short period of time, scientists have learned much about the vocalizations of marine mammals. They must rely on many technologies to track and record marine mammals. Their tools include radio tags, acoustic tags (with transponders), satellite tags, hydrophones, and vocalights (developed by Tyack). Scientists can get an even more detailed look at marine mammal sounds using visual representations of sound called spectrograms.

Hydrophones are simply microphones equipped to record underwater. Because it is difficult underwater to tell from what direction a sound is coming from, four different narrow-beam hydrophones (to pick up sound only in certain directions) are usually placed in the water at the same time.

The different types of tags are simply used to track the movement of animals during their dives, when they can't be seen visually or over longer time periods when the scientist cannot be physically in the area. Radio tags use radio waves for transmission, but these radio waves get scrambled underwater. Acoustic tags are linked to receivers, or transponders, that receive information. But scientists must be on site to make recordings. Satellite tags don't require scientists to be on site, but they don't provide as precise location data. And even when a group of scientists has located and isolated marine mammals, it is still difficult to tell which animal is making which noise. That's why Peter Tyack developed the vocalight. This light, when attached to an animal, lights up when that animal makes a noise. Because each light is different, and observer simply calls out the color that was lit up to find out which animal was vocalizing.

Their search for the role of sound in the everyday lives of marine mammals is getting clearer with new technologies, as species and even individuals are identified. And scientists get even closer to the meaning and purpose behind these otherworldly calls.

"In humans, the ability to learn sounds is the key to the development of languages," says Ford. Killer whales, with their ability to learn each others calls, have hinted at that possibility. Though not quite there yet, "They must be on some point toward that road of a true language...And dialect is a good indicator that they are progressing toward that state."

For Further Reading:

Other wet web sites:

Help! I can't get sound to play!

  • Bibliography

  • Go to thesis main menu