Learning the Alien Language of Dolphins

By Kyle Munkittrick | February 18, 2011 9:47 am

This set of jumps and twists translates roughly to "so long, and thanks for all the fish!"

Humans and dolphins are inventing a common language together. This is big news!

In all the hoopla over the world ending due to being asteroid-smashed, man becoming immortal thanks to the singularity in 2045, and Watson the trivia-machine winning Jeopardy! the story of budding interspecies communication got under-reported. Denise Herzing and her team with the Wild Dolphin project has begun developing a language to allow humans and dolphins to communicate. If successful, the ability to communicate with dolphins would fundamentally change animal intelligence research, animal rights arguments, and our ability to talk to aliens.

Herzing and her team faced two huge problems when it came to talking to dolphins. The first problem is that the current state of animal language research creates an asymmetrical relationship between humans and the animals with whom they wish to communicate. The second problem is that (save for parrots) animal vocal cords cannot replicate human speech, and visa versa.

Most, if not nearly all, animal language research involves either studying how animals communicate with one another, or teaching them a human language to see if they can communicate with us. There is a problem with both methods–humans don’t learn much (if any) animal language in the process. Think of it this way: how many commands does the smartest dog you’ve met know? Some border collies, like Chaser, can learn upwards of 1000 words. Now how many words do you know in dog? Or parrot? How about gorilla or whale? Know any corvid? I bet you can at least read cuttlefish patterns, right? No? Of course, I’m being facetious, but with a purpose: up to this point, humans have always attempted to understand animal language by teaching animals how to talk to humans. The glaring flaw in this process of teaching animals to use human language is that it is nary impossible to prove the animal is using language, not merely playing a very complex game of repeater.

There is a second, equally interesting problem. Think about your favorite science fiction series populated by aliens (for me, that’s a toss up between Star Trek and Mass Effect). At some point in that series, an alien has introduced itself as having a very un-alien name, like “Grunt.” The reason? “My real name is unpronounceable by humans.” That is rarely an actual problem, because as it always works out the other alien species (why do we refer to aliens as “races” btw?) can pronounce our human words. One of the only films I can think of that doesn’t have this common sci-fi fallacy is District 9. Humans and prawn seem to be able to understand the other’s language in a rudimentary way, despite neither species being even remotely able to reproduce the other’s sounds. Cetaceans pose the same problem: humans cannot whistle, squeak, chortle, or pop the way a beluga or bottle-nose can. Further, the higher squeals of some dolphins and the low rumbles of some whales are beyond the human auditory spectrum. Dolphins can’t say a word in human languages and we certainly can’t do more than parody the spectrum of cetacean sounds.

Which presents quite a question: How in the heck did Herzing figure out a way to both not teach the dolphins an anthropocentric language and ensure the language was speakable by both species?

Herzing’s team developed a communication system with a sprig of technology and a heaping helping of ingenuity:

Herzing created an open-ended framework for communication, using sounds, symbols and props to interact with the dolphins. The goal was to create a shared, primitive language that would allow dolphins and humans to ask for props, such as balls or scarves.

Divers demonstrated the system by pressing keys on a large submerged keyboard. Other humans would throw them the corresponding prop. In addition to being labeled with a symbol, each key was paired with a whistle that dolphins could mimic. A dolphin could ask for a toy either by pushing the key with her nose, or whistling.

Herzing’s study is the first of its kind. No one has tried to establish two-way communication in the wild.

Amazing! Herzing’s method is effectively the same as that used in Close Encounters of the Third Kind. The keyboard allows for dolphins to teach humans as much as the humans teach the dolphins. Furthermore, the matched whistle will allow for a more natural integration of communication into the dolphin’s speech. Given the early stages of the project, it seems to have tremendous potential already.

Yet Wired found it necessary to frame Herzing’s breakthrough within the search for extra terrestrial intelligence. As an analogy, I totally understand the reference to aliens. That’s why I used the examples I did above. What is frustrating is that the article seems to see Herzing’s research only as important when in the light of alien communication. No disrespect to NASA (pace the Bad Astronomer), but I’d rather we dumped the funds from our space exploration and focused instead on the oceans of Earth. We’ve got enough aliens and unexplored frontier right here on two-thirds of our pale blue dot.

Thankfully, the flow of information among SETI scientists and marine biologists is two-way. Information theorists like Laurence Doyle (mentioned in the side bar of the Wired article) has used techniques for signal-searching developed with SETI to determine that whales and dolphins use grammar and syntax in their communication. The consequences for genuine communication between a new species is enormous. It would dramatically improve animal intelligence research, as well as make a real case that non-human persons should have limited rights. Success with the dolphins might enable researchers to devise forms of communication with a whole host of other intelligent animals. And, in the far flung future, we might get ourselves a universal translator.

I’m at least hoping for a collar like Dug’s in Up!

Image of impressive dolphins by justthatgoodguyjim via Flickr Creative Commons

MORE ABOUT: Aliens, dolphin, language

Comments (26)

  1. Cmdr. Awesome

    I must be missing something here; I can’t really see what the functional difference is between having them push buttons and whistle, and teaching apes to use symbols and sign language (in fact, if I recall correctly, did not Koko have her own symbol set that she used, one that was not tied to human language? Or am I just completely brainless today?)

    I mean, I see what they’re doing – they’re taking a symbol concept that we have designed, pairing it up with a sound that dolphins can make, and then using those to reference a physical object. But…does the whistle really add anything? Is she using arbitrary whistle noises, or is she using ones that the dolphins make naturally?

    If the whistle noises are arbitrarily chosen and have no natural meaning to dolphins, then that’s really no different than us making funny finger shapes to apes and monkeys and associating that shape with an object. To us those shapes might be sign language, sure, but to a monkey it’s just a funny shape. It’s contextless and meaningless to them.

    If the noises are chosen from sounds that dolphins already make…does that really mean anything useful? Imagine this poor confuse dolphin thinking “Every time I try to warn them about the impending volcanic disaster that’s coming, they throw a ball at my head!”

    I guess what I’m getting at is that it feels to me like they’re going to an awful lot of effort to decontextualize something that has no real context for the animal subjects. I’m guessing I’m seriously overthinking this, or more likely underthinking this. If someone can clarify for me I’d be grateful >.<

  2. Robert E

    @Cmdr. — You are correct, Koko modified some ASL signs and invented some of her own.

  3. Kyle Munkittrick

    Great points, Cmdr. Awesome, you wear your title well.

    Herzing and her team are in the midst of revamping their system. The link to the full academic article is at Acta Astronautica (just search Herzing’s last name). From my understanding, the goal is to avoid precisely the problems you describe. Part of the reason sign language is good but not great is that, ultimately, it is anthropocentric. How do you communicate “this device is for communication” to a species? Close Encounters of the Third Kind offers a pretty good suggestion-just start talking and as they talk back, develop a custom vocabulary.

    Unlike sign language, which required Koko to generate new symbols, whistle language will have words and syntax the dolphins already grasp among themselves. Star Trek played around with this in the episode “Darmok” . Neither Picard nor the Tamarian captain with whom he is attempting to communicate can understand one another, but by being able to make sounds the other could understand, Picard was able to gather meaning. I think, and I may be wrong, but the big advantage is that as the technology progresses, dolphins will be able to teach us their syntax and vocabulary in their own language. The tool is not so much so that dolphins can push a button and get a ball, but so the dolphin could correct us and say “whistle 8 plus a click” is ball, not “whistle 1.” Koko didn’t have a “correct” symbol for kitten, so she could only accept what the trainer gave her in ASL.

    Great questions. I’m not sure myself, but that this project is “open-ended” and founded on the premise that dolphins already have their own language gives me a strong suspicion that Herzing and her team are taking these issues into serious consideration.

  4. Matt B.

    Star Wars episode IV actually had a very sensible system of communication between being of different species. People always spoke their own language but learned to understand the alien’s or droid’s language. No one beeps and whistles at R2-D2 or growls at Chewbacca.

  5. > I’d rather we dumped the funds from our space exploration and focused instead on the oceans of Earth.

    Considering how minuscule a budget our space exploration efforts have, I would say both outer and inner space research could easily be funded. Just buy fewer stealth bombers and invade fewer countries.

  6. Cmdr. Awesome

    Ahhhhhh. Thanks for the info, Kyle. I see where they’re going now and what I missed. That makes considerably more sense to me now!

  7. In “Mind In The Waters” and my recently self-published novel “Wet Goddess” I described an incident where I tried to get a dolphin to mimic my sounds and was successful… then she turned the tables and got me to mimic her!

    All I can say is that when a human does that, the dolphin gets very excited. They’ve finally found a research subject that’s not stupid, after all! It might be worth a try, at least to pique their interest.

  8. Thomas

    Reminds me of the experiment where a dolphin was supposed to respond to some stimulus with a whistle. Except the dolphin kept increasing the pitch every time until the human no longer could hear it, at which point it lowered the pitch again. That was a pretty cool scientific experiment by the dolphin.

  9. LittleJim

    I recall that Lucy, a chimpanzee, went on to take the signs she was taught and use them to make new signs, or to combine signs in arbitrary ways, e.g. there was no sign for “radish”, so she combined the signs for “cry”, “hurt” and “food”.

    It is also possible to speak/understand dog and cat with varying degrees of success. The thing is, these two languages have a large non-verbal component. Try blinking slowly at a cat and then looking away or bowing to a dog and then looking over your shoulder.

    • Kyle Munkittrick

      Good point Little Jim. Again, I’m not denying there isn’t already rudimentary ad-hoc communication. The big breakthrough here is that the technology will allow for much more fluid and inventive communication. Instead of Lucy’s “radish” being an outlier example, hopefully it will be a standard result.

  10. Idlewilde

    This is really interesting. I got a second hand book about John Lily’s experiments and thought they were worthwhile, even though there are many detractors to his work. I’ve always thought it would be great if someone continued this line of research with today’s technology and ideas. Imagine if this worked with whales too…there’s got to be something going on in those huge brains, regardless of the brain to body ratio thing…..

  11. Read all the John C. Lilly stuff you want, including (I highly recommend it) Programming and Meta-programming in the Human Biocomputer, which is basically his report to NASA on his isolation/LSD experiments.

    But for chrissake, if you’re going to mention Lilly’s name around any marine mammalogists or cognitive psychologists, be prepared to either fight or run. He made their professional lives hell, to hear them tell it.

  12. solitha

    Reminds me of Alex.


    The two biggest hurdles to get past are first, for the targets to realize you’re trying to communicate with them; second, to establish baseline concepts to build upon.

    “There is a second, equally interesting problem. Think about your favorite science fiction series populated by aliens (for me, that’s a toss up between Star Trek and Mass Effect). At some point in that series, an alien has introduced itself as having a very un-alien name, like “Grunt.” The reason? “My real name is unpronounceable by humans.” That is rarely an actual problem, because as it always works out the other alien species (why do we refer to aliens as “races” btw?) can pronounce our human words. One of the only films I can think of that doesn’t have this common sci-fi fallacy is District 9.”

    I don’t get this as a fallacy. “Grunt” may be able to speak his own language as well as English (or whatever language the human is using). However, that doesn’t mean the human could repeat Grunt’s native name; it also doesn’t mean that Grunt’s native name could be translated reasonably into any human language. So unless he introduces himself as Grunt, how are humans supposed to refer to him, or call him?

    This actually happens a lot within human linguistics. Asians in particular may choose to Anglicize their names, or choose a nickname that English speakers can pronounce, when moving to the US, for example.

  13. Orcinus

    I’m sorry, but how is 25 year old research “new and exciting”?

    I’ve worked on three different communications projects over the years, not the least of which was Project JANUS in the 1980’s. This is just a rehash of Lilly’s work [ and most every other research on this since then ]. The idea was good, but computers of the time just weren’t up to the task to properly sample the dolphins sounds. Still there was a working 20+ word vocabulary the dolphins understood and would respond to, despite the poor quality of sound. Matching sounds was difficult as the system only did 512 samples a second. A lot for the day but no where near enough to properly analize the dolphin’s range of frequencies. Modern computers are more up to the task and there are new approaches to try.

    The real problem is the lack of overlap in used frequencies. John Kurt’s [ a co-researcher of Lilly’s ] approach at the time actually got a better reaction from the dolphins than JANUS did with the system he was using. If current technology was applied using a newer method I’ve come up with, which allows each to use their native frequencies, we could be actually talking to them. It’s not so much about creating a new language as it is overcoming the frequency differences, which can be done with current technology. In fact the equipment needed to do this already exists, it just hasn’t been applied to this area of research.

    What might not be so easy to overcome is the social/religious impact that having another intelligent species on the planet would have. It throws a monkey wrench at some long held, deeply ingrained beliefs. I wonder if the human species is ready for that.

    Malcolm is right, Lilly has left a bad taste in some peoples mouths, but he was the first to work this field back in the 1960’s, and yet a lot of current “new research” is founded on some of his basic work.

  14. TI’s R&D has pushed and driven my work with speech synthesis for the past twenty six years. I’ve seen great progress in the synthesis, but moreso with the recognition portion. Google and Motorola lead the way, as I see it, with lending research principles, development funds, and getting this tech into the hands of biologists and fans of dolphins.

    It’s unfortunate the author did not mentioned Seaquest DSV. There were mentions of alien-based scifi movies, but not the most relevant scifi tv show. I’m confused when writers jump to popbuzz alien references, instead of sticking with the references to the very animal subjects of their articles. I understand why it’s done, for effect, but it’s easier for me to understand points when one makes mention to the proper entertainment sources that help the points be better understood.

    (I loved this show each week and wanted to be the tech for the talking
    [robot] dolphin. Plus, the dolphin was a good listener and had good jokes.)

  15. tim reynolds

    Koko’s “radish” was “hurt cry fruit”.

  16. bent

    In contrast to this work, there have been many serious attempts at understanding animal communication. More than 50 years ago Hubert and Mabel Frings studied the language of crows. (Frings, H & M. Frings. 1959. The Language of Crows. Sc. Amer. 201 (5):119-131) More recently in his studies of ravens Bernd Heinrich has come to understand many raven calls. Many studies of wolves include interpretation of wolf communications. Etc.

  17. Chris
  18. Dan L

    And after we learn how to communicate with dolphins,
    we’re going to give them jobs. Do you have any idea what
    the unemployment rate among dolphins is?

    I wonder if they’ll ask us to let them out of their cages and
    back into the ocean?

  19. Logan

    As an animal rights person, I don’t agree that “a real case that nonhuman persons should have limited rights” (emphasis on real) could only be made if it were demonstrated that some nonhuman animal was intelligent or had a faculty with language. It would be the final nail in the coffin, I would hope, but numerous philosophers going back at least as far as Jeremy Bentham have cogently reasoned that, if anything, language ability is superfluous to the case for anyone being endowed with certain inalienable rights (like the right not to be owned and the right to life). “The question is not Can they reason? nor Can they talk? but, Can they suffer?” in Bentham’s oft-quoted passage.

    I won’t hesitate to point out that if intelligence or language is the currency by which we weigh moral value, than you inevitably leave out many members of the human community, such as newborn infants and the severely mentally disabled (excluding extremes like embryos and the irreversibly brain-dead). It also poses interesting thought-experiment problems for transhumanists, if unenhanced humans ever had to plead the case for their continued existence and autonomy before skeptical “intellects vast and cool and unsympathetic”.

    To comment on something else, I have also noticed and been annoyed by scenes, such as Han Solo’s dialogue with Jabba the Hutt, in which the human can understand alien and the alien can understand human, and they both can pronounce each others’ words, and yet Han still speaks Basic and Jabba still speaks Huttese.

  20. amphiox

    “The question is not Can they reason? nor Can they talk? but, Can they suffer?”

    The link here is that is it that much easier to show that they can suffer if they can talk.

  21. Very Very Interesting and Thoughtful Post Indeed!!!!!! This writeup has made me spared thought over this highly exciting chapter. I am an animal lover – have been experimenting with my dogs. I have discovered that its not only WORDS that they need to understand to carry out a task assigned by me. It’s about the expressions too! I have learned that when I am happy, my pets seems to be happy and their eyes and facial expressions and body languages mirror their emotions. Contrary, when I am sad… they get dull too. Communication between human and animals, I think, is like an invisible link – like ‘LIGHT’, which we cannot see; we can see things on which the light falls.

  22. Paul

    (Just found this blog, trawling through archives…)

    “but I’d rather we dumped the funds from our space exploration and focused instead on the oceans of Earth. We’ve got enough aliens and unexplored frontier right here on two-thirds of our pale blue dot”

    Sigh. Why space exploration? Why is it always NASA&co that people target when they want their pet project funded.

    Not once have I ever heard a space advocate bemoan the funding of Earth/Ocean Science. Not once have I heard them criticise research into animal language/behaviour. So why is our respect for other research rewarded by making us the target of contempt?

  23. I was thinking about this… i know this is a bit “frankinstineish” but after someone dies what if you took their vocal cords and replaced them with a dolphins… i know that sounds idiotic XD but it might just work… LOL XD ive been having dreams about this…
    -John E

  24. “nary impossible”? Does the author know what the word nary means, or how to use it?

  25. I’ve been saying for years that we should teach dolphins to teach us. They know their language pretty darn well. But since I’m not a dolphin researcher my ideas went no where. Oh well.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!


See More

Collapse bottom bar