Language: If whales had it, would we know?

Factual Questions
181

Tangent. Megafauna can mean “Humans or larger”, which implies that white-tailed deer are megafauna or it can mean, “One metric ton or larger” - which I find more plausible for this discussion. Taking the 2nd definition, most bison are not megafauna, though some can weigh as much as 1180 kilograms, with the confirmed record being 1270 kg. Some may have weighed 1400 kg. Grizzly bears typically weigh comfortably below 600 kg. The record for polar bears just crosses the line at 1002 kg, though pregnant polar bears can weigh 1100 kg. Most are below a metric ton though.

Ok, ok: Walruses. Those are megafauna: they can top out above 2000kg. Most weight between 700 and 1700kg.

To be fair and complete, this wiki article actively applies both definitions, and characterizes the polar bear as megafauna:

182

I prefer a more operational definition: “Can it kick your ass? If so, it’s megafauna.” Run away! Run away! :grin:

183

One megafauna that went extinct only ~12,000 years ago that I wish was with us still is the elephant-sized giant ground sloth (megathere). We wouldn’t have to run away from it, a leisurely stroll away would suffice.

Imagine this colossal creature lumbering toward you like a sleepy freight train. Its massive claws scraping the ground, and its beady eyes squinting as if it’s trying to remember where it left its car keys.

Narrator ( David Attenborough ): “And here we have a Homo sapien, the ultimate multitasker, about to engage in a favorite pastime: fleeing from megafauna. Watch closely as he executes the ‘Casual Evacuation Maneuver.’”

You take a sip of your chocolate latte and start slowly strolling away. The giant ground sloth, meanwhile, is still trying to decide whether it wants to eat you or take a nap. It’s a tough call.

Giant Ground Sloth (thinking): “Hmm, should I chase him, or perhaps just lie down and contemplate the meaning of life? Choices, choices.”

Narrator: “Our Homo sapien protagonist maintains a not so brisk pace. The sloth, however, is still pondering its existential crisis.“

Finally, you reach a safe distance. The sloth yawns, revealing teeth the size of your forearm. You nod respectfully. It nods back, as if to say, “Yeah, I could’ve caught you, but I’m on a low-carb diet.”

And so, you continue your leisurely stroll, leaving the giant ground sloth behind. It’s a tale as old as time: humans fleeing, sloths contemplating, and the universe shrugging its shoulders.

Narrator: “And thus concludes our documentary, ‘Sloths: The Original Chill Giants.’ Tune in next time for ‘Sabertooth Tigers: Dental Hygiene Nightmares.’”

184

I wouldn’t count on giant ground sloths being slow and lumbering just because modern sloths are.

Remember, South America was basically like Australia in that it had a fully isolated set of mammals evolving in parallel to the Afroeurasian-North American assemblage*. So you got a new class of beasties, the Xenarthra, who were very diverse and widespread. Anteaters, ground sloths, glyptodonts, armadillos, and more all came from this group.

Modern tree sloths evolved down a path very similar to the koala. They are small, live in trees, exclusively eat very nutrient poor and difficult to digest leaves, etc.

When you live that kind of lifestyle, moving extremely slowly in order to conserve energy, and basically being a hanging furry ball with a fermentation chamber inside, is a necessary adaptation for survival.

But I wouldn’t assume that slowness is necessarily an ancestral trait of sloths as a group. Giant ground sloths were not adapted for this sort of life, and were in fact quite diverse - some were monstrously big, others burrowed, others swam.

Much like the Diprotodon from Asutralia, I wouldn’t assume that just because their closest living relatives are so pathetic that ground sloths would not have been quite formidable.

*Is there a term for this ecoregion?

185

I’m wondering how scientists know that the extinct elephant-size giant sloth moved as slowly as the modern sloth.

[Editing: ninja’d by Babale. Who said it better .Thanks.]

186

Points well taken, though these articles claim megatherium was one of the slowest animals in its ecosystem, with a slow metabolism.

In any case, if we take a time travel trip to visit them, I’ll be sure to walk a little faster than you guys. :laughing:

187

Deer only give 60-70 lbs of meat, so they’re not megafauna, though they’re close. Bison and bears give much more, but you can only carry 100 lbs back to your wagon.

188

Bumping for a news story, more information:

Gero teamed up with artificial intelligence researchers to create Project CETI in the hope of decoding what sperm whales are saying. They analyzed more than 9,000 recordings of Caribbean sperm whales using advanced computer algorithms.

“In using machine learning to detect the clicks, we found that there were so many more clicks than people could manually segment from the data set,” says Rus, who worked on the project.

The team found sperm whales have a large repertoire of clicks, which they’ve catalogued in a sperm whale “phonetic alphabet.” Sometimes they slightly vary the tempo of the clicks in a coda. Sometimes the length of the coda is subtly longer or shorter. Sometimes the whales throw in an extra click. These variations can be matched closely by different whales communicating. The patterns also appear to be based on the context of the conservation.

“They can be predicted by machine learning in the same way you might predict the sequence of syllables or the sequence of words in a sentence,” Rus says. “It really turned out that sperm whale communication was indeed not random or simplistic, but rather structured.”

189

Very cool @thorny_locust!

So returning to my first contribution to this thread:

Your link documents better minds than mine doing the first part and succeeding in recognizing complex variety that is conversation dependent.

You think they might attempt what seemed to me to be a logical next step? Creating a cetacean large language model that responds and engages with whales in the wild even though it has no “understanding” of the meaning?

A complicating factor is that we already know there are very different cetacean cultures (ecotypes) even within the same general geographical space: will those groups have different languages or at least dialects?

Of course now searching someone else has thought about that!

190

Somewhat belatedly: thanks, interesting article.

Seems rather like humans: different language groups, some words and/or phrasings used as family in-language, everybody recognizes a cry of pain or somebody cooing at a baby.

191

In common with human language, humpback whale songs follow Zipf’s law and Menzerath’s law.

Zipf’s law, from Wiki: “When a list of measured values is sorted in decreasing order, the value of the n-th entry is often approximately inversely proportional to n.”

Comment: Zipf’s Law has also been observed in city sizes. I’m curious about whether the researchers included a control in their experiment. Would recordings of jackhammer sounds or ocean waves on the beach follow Zipf’s law? I see that the researchers did validate their study by scrambling the whale noises and running their tests on a set of artificial datasets. But I’d like a more naturalistic approach.

Menzerath’s law is linguistic only. It states that longer sentences tend to contain shorter words. Huh. I have the same question about experimental controls.

Gifted article. Scientific American and New Scientist also covered this.

Science mag:

The findings don’t suggest whales have a language, where combinations of sounds have fixed meaning and join together in grammatical structures, Garland emphasizes. But the research offers scientists an “amazing window” into how this core property of human communication appears in other species.

192

Another update on an LLM approach.