Today, animal-translation technologies are being developed that use the same "machine learning" approach that is applied to human languages in services such as Google Translate. These systems use neural networks to analyze vast numbers of example sentences, inferring from them general principles of grammar and usage, and then apply these patterns in order to translate sentences the system has never seen.
One researcher engaged in these efforts to "translate" animal communications is Denise Herzing, research director of the Wild Dolphin Project. According to Ball, Herzing is using "similar algorithms, coupled with underwater keyboards and computers, to try to decode dolphin communications." In 2019, Herzing was interviewed by NPR's Guy Raz. In the interview, Herzing noted that one way dolphins communicate is through whistles. She then explained how she and her colleagues developed distinctive whistles for certain objects that the dolphins they were studying liked to play with. Next, they exposed the dolphins to these whistles in the hope the dolphins would start mimicking them. Later, Herzing and her fellow researchers discovered something interesting. According to Ball:
Herzing and her team at the Wild Dolphin Project used a machine-learning algorithm called Cetacean Hearing and Telemetry (chat), designed to identify meaningful signals in dolphin signals. The algorithm picked up a sound within a dolphin pod [group of dolphins] that the researchers had earlier trained the dolphins to associate with sargassum seaweed--a clumpy, floaty plant that dolphins sometimes play with. The dolphins may have assimilated the new "word," and begun using it in the wild.
Herzing's discovery is quite interesting, but it is hardly evidence of dolphins' having language. The reason for this is explained by philosopher J.P. Moreland in a passage I have cited several times on the question of whether animals possess language:
...the question of animal language cannot be adequately discussed without drawing a distinction between a sign and a symbol. A sign is a sense-perceptible object, usually a shaped thing like the characters "BANANA" or a sound (the utterance of "BANANA"). Now if an animal...comes to experience repeatedly the simultaneous presence of a sign (the visual presentation of BANANA) and the presence of a real banana, a habitual association will be set up such that the animal will anticipate the sense perception of a real banana shortly after seeing this shape: BANANA. In the case of an animal, BANANA does not represent or mean a banana, so it is not a symbol. Rather, BANANA is merely a geometrically perceived shape that comes to be associated with a banana in such a way that the latter is anticipated when the former is observed.
By contrast, real language requires symbols and not mere signs. When language users use the word banana, it is used to represent, mean, and refer to actual bananas. Now the evidence suggests that animals have certain abilities to manipulate and behaviorally respond to signs, but it is far from clear that they have a concept of symbols. One reason for this claim is the lack in animals of grammatical creativity and logical thought about language itself that is present in real language users. (The Soul: How We Know It's Real and Why It Matters, p. 144).
In light of what Moreland says, the most that can be said is that the dolphins Herzing studied had come to associate a certain whistle with the presence of sargassum seaweed. In other words, they had learned to use the whistle as a sign for sargassum seaweed. However, unless it could be shown that the dolphins were using that particular whistle when sargassum seaweed was not actually present, they would not have been using it as a symbol, which is what real language consists of, to a large degree. Moreover, note that the dolphins themselves did not develop this sign for sargassum seaweed. It was introduced to them by humans. This suggests a lack of creativity on the part of dolphins. In contrast, creativity with language is a characteristic of users of actual language, as Moreland suggests when he speaks of "grammatical creativity."
In short, it would appear that dolphins do not possess a capacity for language--properly understood. Interestingly, in her NPR interview with Raz, Herzing herself seems less than certain that dolphins have language: "...The answer to the question - do dolphins have a language? - is maybe, kind of." She seems to be aware that claiming unequivocally that dolphins possess a language would be a bridge too far. On the other hand, she seems unable to give up on the idea. Later in the interview, after acknowledging that dolphins could not have a written language, she speculates: "...maybe they have an oral history that they produce like humans did..." The problem is that even an oral history would require a capacity for real language. Given the lack of evidence that dolphins can use symbols (or engage in abstract thinking), this would appear to be no more than wishful thinking.
Image of a dolphin from Wikimedia Commons