Do non-human animals see other members of their species as individuals?

I began my last piece in this newsletter with the question “what’s in a name?” Of course, the meaning of a name reflects in part what is being named. Thus, for names of conditions, a name can carry significance through its associations, as illustrated by the names of several mental conditions. In contrast, for “proper names”, that is names of individuals, the name carries little intrinsic information; it is just a signifier denoting the person.

Of course, if you know something about the person referred to, hearing the name conjures up a lot of knowledge about them. Thus, in greeting a friend by their name, you are already envisioning the kind of interaction you expect or hope to have. This is not only a very human thing to do but seems a specifically human thing to do. While an animal such as a cat or a dog will learn their name from its owner and often (not always!) come when called, they don’t seem to have the general idea of proper names¹. Animals will recognize each other by sight or smell or other clues but they never seem to address each other by name. Or do they?

Some recent work suggests that elephants have the equivalent of bestowing and responding to names amongst themselves. Elephants are highly sociable animals, as long recognized, but names amongst themselves? Really? In fact, there is earlier evidence that certain other animals also have name-like vocalizations which apparently elicit appropriate responses from the conspecifics being called, as we will see. The elephant data, however, are the newest and perhaps the most interesting.

The implications are profound for how we think about animal mentalities. We will return to those issues but first let us take a quick look at the evidence that certain kinds of animals have and use the equivalent of individual names amongst themselves. We begin with the elephants, as revealed in a report published in July, 2024.

The study, carried out over the period 1986 to 2022, involved recording and analyzing the calls that wild African elephants, living on the great savannahs of sub-Saharan Africa, made to each other. Elephants are amongst the most social of animals and the calls they make to one another are part of their daily social intercourse. In all these cases, not only was the call recorded and then analyzed but the calling elephant and the intended recipient of the call were identified. Calls with unknown or multiple recipients were excluded from the analysis.

These “calls” are actually a kind of rumble. It is a low-frequency sound and harmonically complex. Most importantly, it is highly individual, as shown from the analysis of recordings. There are three kinds of elephant calls, defined in terms of their communicative purpose. The first is the “contact rumble”, used when an elephant is more than 50 meters away from other elephants and trying to get the attention of one or more of the latter and/or re-establishing contact. Greeting rumbles, in contrast, are used at much closer range, often within touching distance, and seem to be the equivalent of saying “hi, there”. There is a third kind, affiliative rumbles, which seem the most restricted in use; they are made by mothers interacting with their calves. All three kinds of rumble-call involve interactions amongst elephants who already know each other.

The observers were able to analyse the acoustic features of the calls and identify those elements specific to the intended recipient. A model specific to those features was constructed and then examined as to how successfully it identified a particular elephant. In the 469 calls analyzed, 27% contained a sound signature identifying correctly the recipient. This presumably was close to the frequency in the calls with which the recipient was identified vocally by the caller. (Recall that in many human encounters, one does not always address the person by name.) This frequency was far more frequent than achieved by models with scrambled acoustic features. Significantly, the receiver’s identity as revealed by the model matched the reality more than 87% of the time. This gives confidence in the accuracy of the model to replicate the key feature of individual identification.

Another significant feature was that the acoustic identification did not mimic any sound that the recipient themselves made. Hence, this vocalized identification mark had the same element of arbitrariness that human names have. (This is in contrast to the few previous examples of apparent vocalized identification used by other animals, which do involve mimicry; see below.) Nor did the acoustic identification mark relate to the relationship between calling and receiving elephant. All of the evidence, therefore, indicates that these rumbles, directed at known individuals in the group, are made to attract the attention of the individuality of the elephant being “addressed”.²

This work demonstrates that one specific kind of animal, certainly one of the most intelligent on Earth, uses the equivalent of names – vocalized specific sound signatures – for the conspecifics in its group. This research is impressive in several respects: in its methodologies (not described in detail here), its ambitions (reflecting a project of more than 30 years duration), and its conceptual significance (demonstrating an ability long regarded as specific to humans).

More than a decade earlier, however, something similar was reported for bottlenose dolphins, another highly intelligent mammal with many complex social interactions. These animals too have a range of vocal capabilities for communicating with each other. One of these serves to identify them individually and is unique to every one of them. It is a whistling-like sound that they make at the surface, when in groups, and is termed the “signature whistle”. It involves a carefully modulated frequency pattern and is a self-taught ability. For dolphins living alone, as in the big marine parks, the signature whistle seems to be the sole sound they make. In the wild, however, with dolphins living in groups, it is estimated to comprise 38-70% of the vocalizations, the remainder being other and shared calls. In the ocean, groups of bottle-nosed dolphins meeting up exchange signature whistles before melding groups.

The signature whistle also functions in social cohesion within a group where the members know each other well. When they want to get the attention of a specific member of the pod, a dolphin will produce that individual’s signature whistle and the recipient, recognizing it, gives it back. In effect, each bottlenose dolphin’s signature whistle is their “name”, which they know personally and which is recognized by others in the group as that individual’s vocalized identity label. Presumably it is also used to help organize group activities (hunting?) and such things as mothers keeping track of their calves. The intriguing key feature is that bottlenose dolphins, like elephants, recognize each other as individuals. Highly intelligent mammals, at the very least, have individual personalities and recognize that fact in dealing with their conspecifics. The old behaviorist school of animal behavior, which dominated research in this field, ethnology, for decades, would have been unable to cope with this idea.³

There is, at least, one other group of animals that have been shown to have individuating vocal labels in dealing with members of their own group, members of the parrot family. Several parrot species that live in groups have individual-coding in the acoustic properties of their contact calls to conspecifics in their groups. In one detailed study, in monk parakeets, reports that individual coding is embedded in several different kinds of call, as found in several experimental tests. This indicates that it matters to these birds specifically, in several different situations, whom they are interacting with, not just that they are doing so with fellow parakeets.⁴

To anyone interested in animal behavior, these findings, indicating awareness of the individuality of other members of their species will be interesting in itself, as a fact and a phenomenon. Yet, it also has significance in three more general respects: linguistic, sociological, and psychological.

The linguistic dimension has to do with the primary importance of words/terms/names in language. It is believed that only humans have language, as we understand that word, and that is surely correct. However, it is often then assumed that there are no animals with even the rudiments of language – as opposed to simple vocalizations – and that increasingly looks false. There are a couple of instances of claimed syntax in a species of monkey. The material discussed here, the apparent use of names for individuals in certain animal species, would also be a strong example of an element of language – vocal labeling of specific entities, in this case other individuals in one’s species. There is also the case of a chimpanzee being taught 300 words and a border collie that learned 1000 (!) word-object associations. This surely indicates a much broader potential amongst animals for learning and understanding words. Stay tuned: there are bound to be more examples discovered of elements of language in animals. Learning about them should help us better understand the evolution of language.

The sociological implications of these findings concern the nature of sociality in intelligent, highly social animals. If some species of this kind treat members of their group as individuals, it must mean that their social interactions are deeper and more interesting than had been thought. The evolution of the science of animal behavior, ethology, over the last 50 or more years, away from the behaviorism of B.F. Skinner and his disciples, has been to appreciate, increasingly, the complexity and individuality of many animal minds. (The concept of “mind” was itself dismissed by this school as unfit for discussion.) This broadening has contributed not only to deeper work in this field but to a greater appreciation of animals and the need for better relations between humans and our animal relatives.

Finally, and not least, the fact that certain species of advanced animals – the ones discussed in this article – use vocal labels for individual conspecifics strengthens the case for a “theory of mind” (ToM) in these animals (see ‘‘Theory of Mind’: which animals have it?”) After all, if you see your conspecifics as individuals, you necessarily see them as having their own desires and thoughts and a degree of agency. My feeling, which is just an intuition with no value beyond that, is that all animals exhibiting some degree of social interaction with other animals, of their own species or others, must have ToM (to some degree). This point is, of course, intimately connected to the sociological one, above, about a deeper complexity of social interactions amongst animals than had been imagined earlier.

I plan to revisit all these topics, in greater depth, in later articles.

1. At least I used to think this way about it. Now, I am not so sure. My dog Wolfie is highly sociable and amongst my friends, there are two that he especially loves, as judged by his jumping around, tail-wagging and “smile” (far more pronounced than the typical dog smiley face), when they show up. If, however, I just mention their first names, without any emotional emphasis, when they are not around, he immediately looks from side-to-side, starts tail-wagging, and looks excited and even agitated. He clearly associates their names with them specifically.

2. The work is described in Pardo, M.A. (2024). African elephants address one another with individually specific name-like calls. Nat. Ecol. & Evol. 8: 1353-1364. A short news account is given in Conroy, G. (2024). Do elephants have names for each other? Nature doi.org/10.1038/s41559-024-02420-w.

3. See King, S.L and Janik, V.M. (2013). Bottlenose dolphins can use learned vocal labels to address each other. Proc. Natl Acad. Sci. 110: 13216-13221.

4. One of these published reports on parrots using vocal information in addressing their conspecifics is Smeele, S.Q. et al.(2023) Evidence for vocal signatures and voice-prints in a wild parrot. R. Soc. Open Sci. doi.org/10.1098/rsos.230835