cetalingua

There have been two interesting studies that further support the hypothesis of somewhat complex communication system. One was done in Florida with wild dolphins who hunt cooperatively. One dolphin drives the fish,while others act as a barrier. Such cooperation requires sophisticated acoustic communication. Another study was an experiment with dolphins in managed care. Dolphins were asked to push two separate buttons at the same time, but they could not see each other. So they had to communicate acoustically to achieve the perfect timing and push the buttons together.

Actually no, orcas echolocate a lot and also exchange quite a few calls while hunting fish (based on acoustic tag data).

We do, actually, for dolphins at least. Researchers did an experiment where dolphins had to coordinate their actions (push 2 buttons together) without seeing each other. So they coordinated their actions acoustically by communicating when they should push the buttons.

Motivation could be an issue. What if they do not want to communicate with us? What if we are too slow for them? Dolphins produce burst pulses that have interclick interval of just few milliseconds. We cannot even hear these individual clicks and for us the burst pulse sounds like a creaky door. But they, supposedly, can hear individual clicks. We do not understand or use echolocation, they do, so their "dolphin-socrates" probably would have their own allegory of the cave that is not based on what individuals can see.

No, it is not that simple. Humpback whales have been observed to interfere and mess up with orcas protecting gray whale calves from being killed. It was observed several times and researchers think that humpbacks just really cannot stand orcas harming other whales, read more here: https://www.smithsonianmag.com/science-nature/what-humpback-...

Humpback also have spindle (mirror) neurons in their brains, those are thought to be related to cooperation, empathy, etc. Humans have them too, but not many other species (https://www.sciencedaily.com/releases/2006/11/061127111607.h...).

It is good to not anthropomorphise, but we really should finally move away from Descartes' view on animals being simple automatons.

The obstacle is that we still do not fully understand nor are able to replicate their sonar, even the Navy that has been studying dolphins for decades still cannot duplicate biosonar (read more here: https://www.nationalgeographic.com/news/2019/5/140328-navy-d...)

The burst pulse is extremely complex, some can have 400 single clicks in one pulse (our ear cannot even hear these single pulses, they merge and for our human ear it sounds like a creaky door), and the pulse duration is like a few seconds. Each click is broadband (can go up to 100 kHz and beyond), it is frequency modulated with varied peak frequencies, center frequencies, RMS, some clicks can have 2 peaks, etc. It is super fast and super complex, we we cannot just generate one, only dolphin's sound producing mechanism can.

https://www.zooniverse.org/projects/cetalingua/dolphin-chat

It is not active yet, it will be in a month or so. You can also sign up for our newsletter here: cetalingua.com , we will send notification when this project goes live.

It is not easy, we have not yet cracked the code, plus will they even have the motivation or any interest to communicate with us?

What we do know is that they produce a variety of signals, some very complex and their communication is super fast, which is another obstacle for possible "communication". For example, we cannot generate a burst pulse with our current technology, we can only record theirs and play it back.

But many research groups are trying. We are starting Dolphin Chat citizen science project on Zooniverse in a month or so, to classify and prepare a large dataset of bottlenose dolphins' vocalizations for out deep learning model. You can check it out and even participate, it will definitely help to appreciate how complex their vocal repertoire is (and how "chatty" they are).

We are trying to do something similar (citizen science + AI) but for the acoustic data,our first model can identify manatee calls and mastication (chewing) sounds. https://manatee-chat-demo.appspot.com/

Humpback whales have mirror neurons (the ones implicated in "theory of mind") so they might show "compassion". There is so much we still do not know about marine mammals...

Humpback whales appear to show empathy towards other species. We have 115 documented cases of them doing so (and it is a lot given how hard it is to observe them).

http://oceanwildthings.com/2018/06/what-humpback-whales-can-...

They also have mirror neurons that have been an important component in theory of mind.

https://www.sciencedaily.com/releases/2006/11/061127111607.h...

Thank you!

any suggestions for further reading?

"Deep Thinkers" is a good recent book, you can check it out (https://www.press.uchicago.edu/ucp/books/book/chicago/D/bo27...)

Given our depth in knowledge in acoustics, I'm hoping you can expand a little on this

Our challenge is that we are visual species who are trying to understand “acoustic” species. (see ”What it is like to be a bat? https://warwick.ac.uk/fac/cross_fac/iatl/study/ugmodules/hum...) Yes, we can describe their vocalizations in terms of peak frequencies, RMS, center frequencies, duration, intensity, inter-click-intervals for burst pulses or sperm whale codas, but what does it all mean? The biggest struggle is not even that, but how to conceptualize their communication system. Naturally, being humans, we want it to have semantics, grammar, pragmatics and phonetics, and if it does not, they must be dumb (i.e. Chomsky’s argument). So we search for the smallest unit (a phoneme) and naturally we cannot find it.

Additionally, we know that dolphins have at least 2 sound generators (and possibly up to 4) that could work simultaneously and independently, potentially producing very complex utterances. On top of that, we have an animal that “sees” the world through sound, it must have at least some contribution to how the communication system is set up. The conventional belief as of now is that echolocation is separate from communication, although Dr. John Lilly was convinced that dolphins could use echolocation to describe things in their environment (to be clear, we have no scientific support for this hypothesis).

That is why describing dolphin whistles and other sounds terms of duration, peak frequencies, infliction points, etc., does not help at all, as it does not help us to understand how the communication system is set up and what this whistle actually means (and why is it used). The only positive thing in all that is that there must be some sort of communication system. If you ever encounter a pod (especially the offshore species) you would be amazed at their constant chat. So much energy is spent on all these utterances they must be important and must have at least some meaning.

Oh, yes, Dr. Lilly and his NASA funded study. All acid, sex and other things aside, some of the stuff he did was indeed interesting, if only to get to know dolphins in a way impossible in the wild. There was one NPR interview with Margaret some time ago, she mentioned that Peter (the dolphin she worked with) was absolutely captivated by her knees. He would echolocate on them and investigate them constantly. If you think about it, dolphins must know each other's anatomy pretty well due to their echolocation abilities. So a human knee (having no analog anywhere in a dolphin's body) was the constant focus of Peter's interest. But wild dolphins do not seem to be captivated by human knees, so this difference in their motivation is very interesting. In fact, the question about curiosity and motivation to communicate is a important one, even if we do a playback and use all correct sequences, if the dolphin is not motivated to reply in any way, the experiment will fail and we will never know why.

The meaning of a lot of human languages or alphabets is completely lost even if they were conceived by humans almost identical to us biologically. Hoping to understand alien languages without the aliens helping us is a pipe dream.

There has been limited success with parts of Voynich Manuscript (some plants names?) and some interesting developments with rongorongo, but overall, yes, super hard. Hopeflly, Natural Language Processing might help, some promising results were reported for Linear B. (https://neurohive.io/en/news/researchers-use-machine-learnin...)

We know a lot about the language of large ocean mammals but it's just not as one to one with how we would translate it.

Well, not exactly.In fact, we still know surprisingly little. For the humpback song, one of the most studied topics up to date, there is still a major disagreement if it is a reproductive display or something else.Humpbacks make tons of other sounds as well, those have not been studied as much as their songs. For bottlenose dolphins, the most studied species, we sort of know the function of the signature whistles (again, not everyone agrees), but know very little about the function/meaning of other whistles and other sounds like burst pulses or LFNs.Echolocation has been studied a lot, still, we do not know if it is used in communication in any way.

Speaking of dolphins (and whales for that matter), there is a reason why it has been so hard to not just "communicate" with them but to even understand the function and the meaning of a number of calls they make.It appears that it is very hard for a human mind to understand a consciousness so different from our own. We are visual species, and even though our language is acoustic, it reflects our nature as visual species. Dolphins' primary modality is sound, so if they have any communication system, it will reflect that. Will we be able to crack their code someday? Hopefully, the answer is yes.

Yet, after nearly 5 decades of research we are still unable to crack their code. We do know however that some sequences they make do not seem to be random.Our problem is to penetrate the communication system that is not visually based,and is purely acoustical.

Shameless self-plug: we are working on it, combining citizen science and AI. Our plarform goes live in a few weeks.

To add to the mystery, why do people get less motion sickness when they are not just passengers, but drive their car or a boat?

When you see it in nature, like in birds and whales, it's about finding a mate.

Not exactly, at least for whales. The humpback whale song's function has not been fully understood. The initial hypothesis that it is a vocal display for females has not been supported, since females do not approach singing whales, and whales sing all over the place, not just at the breeding grounds. So the function of a humpback song might not be solely to find a mate, or might not be related to reproduction at all.

https://www.the-scientist.com/features/the-mystery-of-whale-...

Furthermore, you do need a complex communication system for complex social groups. For example, there are two bottlenose dolphin ecotypes:coastal and offshore. Coastal ecotype is a small fission-fusion society and we see a somewhat simple vocal repertoire. The offshore ecotype lives in big groups and they have very complex vocal repertoire. The bottom line is, at least for cetaceans, we still do not have enough data and understanding to realy argue about the complexity of their communication system one way or another. But it definitely goes beyond of just being "about finding a mate"

Agree!

First email, "you are in! ",second email "ups, actually, you are not". Crushing...

Great movie!

Even with a tag attached to a whale it is still impossible to be completely sure who made a sound. This is a big issue that has been haunting marine mammal communication research for decades. Hard to establish the function of the call when it is not clear who made it and who replied.

We signed up for Stripe account on February 9, but are yet to start processing paymens, will our account be grandfathered in as well? I sent you email yesterday asking about it too. We did not receive Stripe email about changes. Could you please clarify it? Thank you!

Here is a question, how do they determine who is whistling to them if they do not see them? Is whistler's identity encoded in a whistle somehow? And if yes, then how? In cetacean whistles it is believed that dolphin's identity is not encoded in a whistle that is why they need signature whistles.

On a side note, whistling language and cetacean whistling are nice examples of convergent evolution, cetaceans also use whistles for long distance communication, and use other sounds like burst pulses for short distance comunication with animals nearby. Sociality is another common component.

Another amazing thing about whistled languages is that they are still closely linked to their language of origin. If the language is tonal, the whistle will follow their language melodic structure. If the language is not tonal , whistles will try to mimic vowel resonance and convey consonants via how they whistle, i.e abrupt changes.

Denise Herzing used synthetic whistles trying to label things in our human way, pretty much the extension of Herman's line of thought. But we do not know if dolphins indeed label things in their natural communication system.

Signature whistles are tricky, and not everyone agrees that they are actually a thing. There was a recent study done in Israel where dolphins appeared to be using signature whistles of dolphins who died some time ago. It could mean that either signature whistles are not what we think they are, or that dolphins have routine conversations about their dead companions.

The role of larynx in cetacean sound production is still debated, but we do know that they posses at least 2 sound generating mechanisms that could work independently and simultaneously. It appears that right side phonic lip is involved in echolocation and pulsed sounds, the right one is involved in whistling. Two (maybe even 3, if larynx is indeed invoved too) sound producing mechanisms make their communication even more complex, since it greatly increases the sound combinations probabilities.

Many researchers have already done that, i.e. using video and audio for sound analysis, and here we are,60+ years later, still do not know the function of even a single call, apart, of course, from signature whistles and echolocation. This is a non-human communication system, 25 mln years in the making by creatures who rely on sounds in ways humans never will.

When we study cetaceans, we struggle to determine who produces the sound, who responds, even tagged animals are problematic, because the sound could be made by someone next to them. In large pelagic pods, it is a cacophony, they chat all the time and never shut up. It appears that at least in dolphins, whistles could be more of a long distance signals, but pulsed ones, like burst pulses, are more of a short distance signals, for those who are close by.

But make no mistake, some information is being transmitted, we know that. The Bastian experiment with all its flaws was first to demonstrate it,and it has been replicated too. Some researchers are not entirely convinced, but it is definitely a very interesting area to explore.

When she died in 2014 Nellie (a bottlenose dolphin in Marineland) was 61. No one realy knows what is the maximum lifespan in the wild. Some bowhead whales have the lifespan over 200 years, and we know nothing about their self-recognition or when it emerges.

In pygmy sperm whales, magnetite crystals appear to be concentrated in the rostroventral dura, but we know next to nothing about how they function or are used, let alone about how they are being formed.