Study: Elephants Have Learned To Tell Certain Languages Apart

sciencehabit writes "Whether we realize it, African elephants are listening to us. The pachyderms can tell certain human languages apart and even determine our gender, relative age, and whether we're a threat, according to a new study. The work illustrates how elephants can sometimes protect themselves from human actions. The work may be helpful in preventing 'human-elephant conflicts where the species co-exist,' says Joshua Plotnik, a behavioral ecologist at Mahidol University, Kanchanaburi, in Thailand. For instance, elephants might be deterred from entering farmland or encouraged to stick to the corridors designed for their use. 'The trouble is elephants are too smart to be fooled by us for long.'"

Comments

I, for one...

[Anonymous Coward]

... welcome our new pachyderm overlords!

How fine is this distinction?

[gman003]

I would expect that they're either keying off certain words, or that they're going off phonology (the sounds that are used in a language). It might be a good follow-up study to figure out what method they use to make this distinction (TFA does not say, and the paper is paywalled).

I also wonder how fine a distinction between languages they can make. How close are the Kamba and Maasai languages? If they're relatively distant (like, say, English and Maasai), how do they deal with closer languages (like English and German, or even Spanish and Portuguese)? Are they able to distinguish accents?

Re:How fine is this distinction?

[mmell]

On the other hand, they may be more intelligent than we'd previously thought, or at least possess abilities we've previously overlooked. It may not take a brain the size and configuration of ours to have a circuit capable of discriminating or parsing speech. Conceivably, such an organelle of the elephant brain need not even (grossly) resemble its analog in the human brain. Think of it as A/D on two different chip architectures - they may perform equivalent functions in entirely dissimilar ways, even though both are implemented using the same underlying chip manufacturing techniques.

Re:How fine is this distinction?

[Dutch Gun]

I would expect that they're either keying off certain words, or that they're going off phonology (the sounds that are used in a language). It might be a good follow-up study to figure out what method they use to make this distinction (TFA does not say, and the paper is paywalled).

I also wonder how fine a distinction between languages they can make. How close are the Kamba and Maasai languages? If they're relatively distant (like, say, English and Maasai), how do they deal with closer languages (like English and German, or even Spanish and Portuguese)? Are they able to distinguish accents?

Probably the same distinction all sorts of co-habitating animals of different species make when distinguishing between, say, the chattering of harmless monkeys or jungle birds versus the growl of a predatory animal. It makes sense to me, since it seems like the ability to distinguish between animal languages (or even different types of sounds within the same species language) would be a valuable evolutionary trait.

I'd be surprised if they could distinguish fine accents. If you gauge your own ability, you can typically tell when people are speaking different *major* languages, but not between regional differences of the same language, for instance. Or, very closely-related languages are also hard to distinguish for most people. I'd be surprised if elephants were able to distinguish any better than us.

Re:How fine is this distinction?

[gman003]

If they do it a completely different way than humans, that's even better because it tells us there's more than one way to do it. Perhaps their way works better given some constraint - a constraint that might be similar to an artificial intelligence's?

Re:How fine is this distinction?

[rmdingler]

Man's hubris is large enough to obscure vision and good judgement.

When we were little more than barely civilized, our insecurities probably collectively led us to this massive overcompensation that skewed our judgement of the other mammals' intelligence.

We are now, he said hopefully, so much beyond that infantile assumption that we may one fine day be caught saying, "You are welcome for the fish."

Re:How fine is this distinction?

[mikael]

If you draw a graph of brain size vs. number of words an animal can learn (parrot = 200, cat = 50, dog = 1000), an elephant should be able to learn hundreds of words. A wild animal like an elephant is going to have to be aware of every possible sound from every possible creature (crocodiles snapping, toads croaking, hyenas fighting, vultures crying, lions fighting, as well as watery sounds like thunderstorms, rain, waterfalls and rivers. Then they can also hear infra-sound as they communicate using low frequency.

Some brain scans were done of elephants and it seemed they had larger brain regions related to hearing.

Re:How fine is this distinction?

[relisher]

They had a great BBC interview about this and they mentioned how the elephants could tell apart the Maasai and Kamba languages. The online article on the BBC website also mentions how the elephants can tell apart gender by recognizing changes in the pitch and frequency of the voice. http://www.bbc.com/news/scienc...

Re:How fine is this distinction?

[ahabswhale]

It's because man always judges intelligence based on human standards, which is completely idiotic. It would be like me (as a programmer) judging an English professor as an idiot because he doesn't understand code. I'm sure from the perspective of elephants, we're pretty fucking stupid at being elephants. Our intellect isn't well suited for their life style and vice versa.

In short, you're right...human hubris is nearly unbounded. It's very convenient though; we don't have to respect other life on this planet so we can exploit it without regard.

Re:How fine is this distinction?

[schlachter]

On the other hand, they may be more intelligent than we'd previously thought,

I take it you're not in the field of comparative cognition or animal cognition. Elephants have been known for some time to be very smart. Among the smartest animals after Dolphins and Chimps, respectively. They have complex language, tool usage, learning skills, and social relationships.

I don't think this study is meant to be a breakthrough understanding of Elephant abilities as much as proof that they possess a specific ability.

We're doomed

[digitalPhant0m]

'The trouble is elephants are too smart to be fooled by us for long.'

We're doomed.

Racist Elephants!

[Etherwalk]

So... the elephants make decisions about danger based on age, gender, and language?

Pink elephants

[MindPrison]

I knew it, I wasn't crazy!

Those pink Elephants have been talking to me for YEARS. All it takes is a few beers, then some more...and there they are, floppy ear pink bastards!

Elephants are smarter than you might think

[koan]

elephants have a total of 257 billion neurons, three times more than humans.[1] The elephant's brain is similar to that of humans in terms of structure and complexity—such as the elephant's cortex having as many neurons as a human brain,[2] suggesting convergent evolution.[3]

https://en.wikipedia.org/wiki/...

Their trunks are also quite dexterous, I'm actually surprised there isn't more tool use amongst them.
https://www.youtube.com/watch?...

My cats know when I'm up to something.

[billstewart]

Sure, they're nowhere near as smart as elephants, but my cats generally know when I'm up to something, whether that's something that could be used to talk me into giving them treats, or something that might get them locked up into the bathroom and maybe shoved in a box and taken to the vet. One of my cats is better at figuring out treats, and usually pretty dumb about being herded somewhere, while the other one's better at figuring out potential bad stuff, but most cats have at least some clue.