A new look at sperm whale communication

May 8, 2024 @ 11:20 am · Filed by under Animal communication, Data bases, Language and animals

« previous post | next post »

For as long as I can remember, I've been aware that whales, dolphins, porpoises, and other large mammals of the seas (the cetaceans) make whistles, clicks, calls, groans, songs, and other sounds / noises.  These vocalizations are manifestly complex and nuanced, leading people to believe that they are communicating content, emotions, and so forth.  What exactly they are conveying and how they do it have remained a mystery, but researchers never stop trying to figure out cetacean "language".  A new study at MIT claims to have made progress in analyzing sperm whale sound systems.

Scientists document remarkable sperm whale 'phonetic alphabet'
By Will Dunham, Reuters (May 7, 2024)
[with 2:58 video]

I was hesitant to read this article at all because of the mention of a "phonetic alphabet".  Even with the quotation marks around it, attributing this ability to sperm whales was a bit much for me.

Yet, since it was "scientists" doing the documenting, I forced myself to read the first two paragraphs:

The various species of whales inhabiting Earth's oceans employ different types of vocalizations to communicate. Sperm whales, the largest of the toothed whales, communicate using bursts of clicking noises – called codas – sounding a bit like Morse code.
 
A new analysis of years of vocalizations by sperm whales in the eastern Caribbean has found that their system of communication is more sophisticated than previously known, exhibiting a complex internal structure replete with a "phonetic alphabet." The researchers identified similarities to aspects of other animal communication systems – and even human language.

At that point, I was about to click the close button on this article as being another tiresome attempt to humanize animal communication.  But my eye caught the next two paragraphs:

Like all marine mammals, sperm whales are very social animals, with their calls an integral part of this. The new study has provided a fuller understanding of how these whales communicate.
 
"The research shows that the expressivity of sperm whale calls is much larger than previously thought," said Pratyusha Sharma, a Massachusetts Institute of Technology doctoral student in robotics and machine learning and lead author of the study published on Tuesday in the journal Nature Communications.

Well, now, since the Reuters article cites a paper in Nature Communications, I thought I should at least take a look at that — Pratyusha Sharma et al.,  "Contextual and combinatorial structure in sperm whale vocalisations" ,  Nature Communications:

Abstract: Sperm whales (Physeter macrocephalus) are highly social mammals that communicate using sequences of clicks called codas. While a subset of codas have been shown to encode information about caller identity, almost everything else about the sperm whale communication system, including its structure and information-carrying capacity, remains unknown. We show that codas exhibit contextual and combinatorial structure. First, we report previously undescribed features of codas that are sensitive to the conversational context in which they occur, and systematically controlled and imitated across whales. We call these rubato and ornamentation. Second, we show that codas form a combinatorial coding system in which rubato and ornamentation combine with two context-independent features we call rhythm and tempo to produce a large inventory of distinguishable codas. Sperm whale vocalisations are more expressive and structured than previously believed, and built from a repertoire comprising nearly an order of magnitude more distinguishable codas. These results show context-sensitive and combinatorial vocalisation can appear in organisms with divergent evolutionary lineage and vocal apparatus.

That enticed me enough to go back to the Reuters article to see what other insights the researchers may have to offer:

"We do not know yet what they are saying. We are studying the calls in their behavioral contexts next to understand what sperm whales might be communicating about," said Sharma.
 
Sperm whales, which can reach about 60 feet (18 meters) long, have the largest brain of any animal. They are deep divers, feeding on giant squid and other prey.
 
The researchers are part of the Project CETI (Cetacean Translation Initiative) Machine Learning Team. Using traditional statistical analysis and artificial intelligence, they examined calls made by about 60 whales recorded by the Dominica Sperm Whale Project, a research program that has assembled a large dataset on the species.

The facts that sperm whales have the largest brain of any animal and can eat a giant squid really impressed me.  But what are they saying to each other?

"Why are they exchanging these codas? What information might they be sharing?" asked study co-author Shane Gero, Project CETI's lead biologist and Dominica Sperm Whale Project founder, also affiliated with Carleton University in Canada.
 
"I think it's likely that they use codas to coordinate as a family, organize babysitting, foraging and defense," Gero said.

I love that:  "organize babysitting"!

Variations in the number, rhythm and tempo of the clicks produced different types of codas, the researchers found. The whales, among other things, altered the duration of the codas and sometimes added an extra click at the end, like a suffix in human language.

Sounds promising, though a less ambitious interpretation would be that the added extra click at the end is similar to Victor Borge's "phonetic punctuation".

"All of these different codas that we see are actually built by combining a comparatively simple set of smaller pieces," said study co-author Jacob Andreas, an MIT computer science professor and Project CETI member.
 
People combine sounds – often corresponding to letters of the alphabet – to produce words that carry meaning, then produce sequences of words to create sentences to convey more complex meanings.
 
For people, Sharma said, "There are two levels of combination." The lower level is sounds to words. The higher level is words to sentences.
 
Sperm whales, Sharma said, also use a two-level combination of features to form codas, and codas are then sequenced together as the whales communicate. The lower level has similarities to letters in an alphabet, Sharma said.

One thing I will grant the MIT researchers is that, thanks to their project, we know a lot more about sperm whale communication and that it is much more sophisticated and expressive "than was previously thought / believed / known", which has become a sort of mantra for the team, and a reasonable scientific aim at that.

 

Selected readings

[Thanks to June Teufel Dreyer]

May 8, 2024 @ 11:20 am · Filed by under Animal communication, Data bases, Language and animals


7 Comments »

  1. Gene Buckley said,

    May 8, 2024 @ 12:55 pm

    I was embarrassed on behalf of the author, for not understanding that "alphabet" is a written phenomenon, and that /ba/ is not a "sound".

  2. AntC said,

    May 8, 2024 @ 4:08 pm

    AFAICT from their biogs/publications, none of these "scientists" is any sort of Linguist. They're using the same statistical pattern-finding as brought us ChatGPT. How hard would it be for the Computer Science and Artificial Intelligence Laboratory — MIT! — to knock on the door of the Linguistics department?

    Neither could I see a musician, so it's cold comfort they're mis-using musical terminology as much as linguistic.

    more … than previously believed

    By who? (No citations for that claim.) I've always believed cetaceans are super-smart.

  3. Chris Button said,

    May 8, 2024 @ 4:12 pm

    I just saw a report of Aljazeera on this very topic. And I thought to myself, I wonder if this will appear on Language Log…

  4. Victor Mair said,

    May 8, 2024 @ 5:42 pm

    @Chris Button

    We try not to miss anything interesting or important in language and linguistics. Our antennae are always out.

  5. Jason said,

    May 9, 2024 @ 8:24 am

    So sperm whales have the largest brain of any creature, does that make them the most intelligent?

  6. RfP said,

    May 9, 2024 @ 6:35 pm

    @Jason:

    To borrow from an adjacent thread: My grandpappy didn’t raise me to be dumber than no sperm whale!

  7. Yves Rehbein said,

    May 10, 2024 @ 2:56 pm

    @Gene Buckley, I believe "alphabet" is standard terminology in coding theory. For example, the DEFLATE algorithm used in the zip file format is described thus in RFC 1951.

    https://en.wiktionary.org/wiki/alphabet#Noun
    https://datatracker.ietf.org/doc/html/rfc1951#page-6

    Also, "bah!" as in "bah humbug" is a phrase, a word, a morpheme, sylable, mora, and I'm tempted to argue that it is a phoneme because the vocalic auslaut (unfortunately also called "syllable coda", a false cognate) is phonetically conditioned.

    @TFA, the cited linguistics seem to be on point with regards to "combinatorial and contextual structure". Their main source for design features of natural language (supplementary table 2) is Hockett, "The origin of language". The hierarchy of language is useful as analogy as it is simple (supplementary table 4) and they admit that the terminology is provisional. However, their use of the terminology is slightly confusing.

    The click is defined as a "multi-pulse structure", i.e. sequences of monopulses characterized by mutliple features. They find a number of 156 recuring codas, "frequently generated feature combinations", based on the observation that one coda consist of a small number of clicks, five on average. In supplement section 7, "Parallems to Human Language", codas are likened to morphemes/words; sub-coda features are likened to phonetic features or phonemes, quite ambiguous.

    On the other hand, the discussion claims to show "… how a small set of axes of variation (…) give rise to the diverse set of observed phonemes (in humans) or codas (in sperm whales)". Other than that, they don't really use the term phoneme and stick to coda as the basic unit.

    This is acceptable, because they claim "that characterising sperm whale interactions as sequences of fixed coda types overlooks a great deal of information" (p. 2). Replying to the linguistics side of the argument, I am sure that the same may be said of human language, in line with the premises of the introduction.

    A real problem is reproducability. .@Gene Buckley, I believe "alphabet" is standard terminology in coding theory. For example, the DEFLATE algorithm used in the zip file format is described thus in RFC 1951.

    https://en.wiktionary.org/wiki/alphabet#Noun
    https://datatracker.ietf.org/doc/html/rfc1951#page-6

    Also, "bah!" as in "bah humbug" is a phrase, a word, a morpheme, sylable, mora, and I'm tempted to argue that it is a phoneme because the vocalic auslaut (unfortunately also called "syllable coda", a false cognate) is phonetically conditioned.

    @TFA, the cited linguistics seem to be on point with regards to "combinatorial and contextual structure". Maybe they are missing newer trends. Their main source for design features of natural language (supplementary table 2) is Hockett 1960, "The origin of language". The hierarchy of language is useful as analogy as it is simple (supplementary table 4) and they admit that the terminology is provisional. However, their use of the terminology is slightly confusing.

    The click is defined as a "multi-pulse structure", i.e. sequences of monopulses characterized by mutliple features. They find a number of 156 recuring codas, "frequently generated feature combinations", based on the observation that one coda consist of a small number of clicks, five on average. In supplement section 7, "Parallems to Human Language", codas are likened to morphemes/words; sub-coda features are likened to phonetic features or phonemes, quite ambiguous.

    On the other hand, the discussion claims to show "… how a small set of axes of variation (…) give rise to the diverse set of observed phonemes (in humans) or codas (in sperm whales)". Other than that, they don't really use the term phoneme and stick to coda as the basic unit.

RSS feed for comments on this post

Leave a Comment