*A version of this article originally appeared in the August 2024 Bush Telegraph newsletter. You can read our recent newsletters and sign-up to receive these in your inbox on our Bush Telegraph newsletter page.
If you’ve ever watched wild elephants interact at close range, you’ll have both heard and felt their signature rumbles: as impressive as any lion’s roar, the deep, echoing sounds reverberate through your entire body. It’s long been known that that these rumbles were more than mere noise and that elephants communicated at a sophisticated level. Now, a remarkable new study by scientists at Colorado State University has found that African elephants call each other by individual names.
The study, which included 14 months in the field, involved analysing 469 elephant ‘communication’ calls recorded in Kenya’s Amboseli National Park and Samburu and Buffalo Springs national reserves. As elephant rumbles are extraordinarily complex, the team trained a Random Forest – an artificial intelligence algorithm – to analyse them.
“We found that elephants used different calls to address different individuals. Not only that, but different elephants used similar calls to address the same individual elephant. The calls weren’t just specific to the caller; they were also specific to the receiver,” lead study author and behavioural ecologist Michael Pardo explains.
Incredibly, this is the first time a non-human animal has been found to use names that don’t involve mimicry.
“Both Bottlenose dolphins and certain parrot species have distinct signature calls -unique names – that they make to introduce themselves. But sometimes, they’ll imitate another individual’s signature call to attract that individual’s attention.
“But we found no evidence of elephants using signature calls or of them imitating the calls of elephants they were trying to address. Their address calls were no more similar to their intended recipient’s calls than any other elephant’s calls. This suggests that elephants, like humans, have arbitrary names for each other, which, in turn, indicates elephants have the ability for abstract thought.
“But the most exciting part for me was the playback experiment, where we played recorded calls to the elephants. It’s one thing to find patterns with machine learning that support your hypothesis, but you still wonder, “Is this meaningful to the animals themselves?” The experiment showed that the elephants could tell just by hearing the call if it was meant for them or someone else.”
On average, the elephants that the playback call was intended for approached the speaker 128 seconds sooner, vocalised 87 seconds sooner and produced 2.3 times more vocalisations in response to the call.
“I was a little surprised by the results. Not because I didn’t think it was plausible that elephants had names, but because we could only do the playback with 17 different elephants, and we didn’t know at that point which, if any, of the calls we were using had names in them. Getting such significant results was the strongest evidence that ‘this was real’, and that’s when I thought, ‘Wow. Maybe we really have something here’.”
The team also found that while not all calls contained names, those that did were more likely to be long-distance calls, where elephants were out of visual contact, and caregiving calls from mothers to calves. Additionally, calls by adults were far more likely to contain names and have the receiver correctly identified than calls by juveniles.
“Interestingly, this suggests an elephant’s use of names may take years to develop,” Michael said.
Unlike dolphins’ signature calls, elephants’ names are just one tiny piece of the information conveyed in a rumble.
“We still need to figure out exactly which part of the rumble is the name. If we can do that, we can ask questions like, ‘Do elephants refer to other elephants in their absence?’. It would be exciting if they do because displacement – the ability to talk about something that isn’t present at that moment – is a key feature of human language and very rarely seen in non-human communication.”
That knowledge might also help answer the most intriguing question of all.
“We’re really interested in how elephants develop names: do they learn them or create them themselves? Mothers often produce repeated coo-rumbles to their calf shortly after birth, and I wonder if that’s the moment a calf is named. It would be really exciting to find that out.”
If you’ve been inspired and want to find out more, give us a call or enquire now to speak to an expert.
