DOYLE: We found that the complexity of the dolphin whistles, as far as we could measure, was compatible with the complexity of human languages.

COOMBS: Here’s why. Linguists have demonstrated that all human languages follow a distinct pattern of sound distribution. They have even put a name on this, it’s called Zipf’s law.

DOYLE: Zipf’s law is a linguistic study that shows that virtually all languages from Chinese characters to Arabic characters and human words and alphabetic symbols and so on all produce this balance between diversity and repetition.

COOMBS: Simply put, Zipf’s law means that the most common sound, for instance, in English the sound made by the letter E, occurs ten percent of the time. The second most common sound, the sound made by the letter T, occurs eight percent of the time. By the time you get to the 25th most common sound, in English, that would be the letter Q, that sound occurs only once in a thousand times. Although the sounds may very from language to language, the patterns of sound distribution always remain the same. To Doyle’s surprise, this same distribution of sound seems to occur with dolphin whistles.

DOYLE: The most frequent whistle, two, is used about ten percent of the time and so on, down. So dolphins show the same linguistic distribution that is argued to be uniquely human, and it isn’t. And that’s news - because recently a lot of studies have been going on in human linguistics that show that one of the reasons human linguistics is unique is because it has this distribution.

COOMBS: By the way, here is the most frequently used dolphin vocalization… the whistle number two Doyle mentioned.

[DOLPHIN WHISTLE]

COOMBS: Doyle, a physicist by training, admits the findings don’t show that dolphins have sentences with subjects and predicates, or that dolphin whistles have the same meaning inherent in human words. But, he says, his research has demonstrated that dolphin calls follow a pattern analogous to human syntax, or sentence structure.

DOYLE: Linguists out there are having a conniption, I’m sure, because when we say “syntax,” it’s because we don’t know what else to call it. We are not implying the advanced concepts of grammar that syntax implies - but there is no term. That there’s some kind of structure, that signals depend on preceding signals, That the signals have a context… that is all we are saying at this point. And so far as we know, the dolphin repertoire is on track for being as complex as the human communication system is.

COOMBS: Because dolphin whistles follow human patterns, Doyle speculates that dolphins can transmit complicated thoughts and ideas. His theory is, however, controversial among linguists and communication scholars like Leonard Hawes at the University of Utah. Hawes doesn’t believe that sound patterns necessarily constitute content or meaning. He says that the dolphins may be communicating, but this doesn’t mean they have a complex language.

HAWES: The question isn’t can two animals or two humans emit patterns of sound that produce particular responses in others of their kind, whether it’s human or gorilla or dolphin. The answer is yes, of course they do, we see it. I suppose the answer to your question, does this demonstrate a language, obviously depends on how restrictive your definition of language is. Does it mean they have an elaborated syntax and vocabulary and grammar and—how you would ever ultimately validate that is more problematic.

COOMBS: That’s because dolphin language may be so different from our own, it might not even be translatable. After all, Hawes says, there are translation problems just going from one human language to another.

HAWES: But at least you can approximate German words for English words or French words. What we are less certain of is can we translate dolphin sounds or chimpanzee sounds or bird sounds into something that looks like human language. Can we put it into “don’t come any closer,” or “get away,”— can that be translated into words? At best only approximately, and the problem is that we can’t ask the dolphin if we got it right.

COOMBS: To make matters more complicated, there may not be a single whistle that means things like predator or food. It’s more likely that communication occurs through a complex whistle sequence. This makes it nearly impossible to identify the meaning of dolphin vocalizations. After 30 years of research scientist have only been able to conclusively identify the meaning of just a couple of whistle sequences. One of these is the thunk call, a pattern of short knocks that sounds like a person drumming on a piece of wood.

[THUMPING SOUND]

DOYLE: When the mother thunks, the baby comes right to her side.

COOMBS: Again, Laurance Doyle.

DOYLE: It means danger, or come here, or something, but we have seen baby dolphins drop everything and go immediately to the mom’s side when she produces a thunk call. Now, you have to keep in mind, they are probably talking about things that aren’t present. Consequently, trying to correlate objects with what they say hasn’t been very successful. When you are talking about a symbolic representation, trying to get at the meaning of something is much more difficult for another species…

COOMBS: Although Doyle doesn’t yet understand the meaning of the vocalizations, he cites another startling discovery that indicates the human-like sophistication of dolphin communication. Scientists recently found that dolphins learn to speak in a way very similar to human babies. Babies make 869 different sounds and tones. As they acquire language, they eventually use only the 45 sounds that make up adult language. This happens in every culture. It also happens in dolphin communities.

DOYLE: As we studied comparisons between dolphin whistles and humans we found that like human babies babble, dolphin babies also babble their whistles. They make all possible whistles and sounds and noises and stuff, all with about equal frequency.

COOMBS: Like humans, baby dolphins begin with hundreds sounds. By imitating the whistles made by their parents, the babies eventually narrow their sound repertoire to about a hundred whistles.

DOYLE: We found that at about two months they start to get very redundant and that lasts until 8 months. Now they are saying a few whistles that are good and they are being encouraged to use those whistles. So they are learning a vocabulary. And eventually, they wind up with the same distribution that human speech has.

COOMBS: Despite the difficulty of identifying meaning in dolphin whistles, Doyle believes that his research demonstrates the profound intelligence of this creature. Given that, he says we have an even greater responsibility to protect this mammal.

DOYLE: Clearly, we have underestimated their intelligence. Is that enough to save dolphins - that they actually talk to each other with meaning and syntax? I’d say, that’s enough to take a few steps back and reconsider the animal community and the degree of intelligence they actually have. And yeah, no acceptable dolphin deaths by accident. I mean, I consider it like human deaths at this point.

COOMBS: Doyle is currently partnering with scientists at the University of California at Davis to search for complexity in bird, primate and whale calls.

Doyle says he’ll continue to look for similarities between dolphin vocalizations and human language. He and his colleagues will also begin to compare dolphin calls to those of the humpback whale, the only creature Doyle suspects that might have a communication system more complex than the dolphin’s.

For Living on Earth, I’m Amy Coombs in Mountain View, California.

[MUSIC: The Orb “Oxbow Lakes” Orbus Terrarum Polygram (1995)]