That Funky Gibbon Song

A male agile gibbon

We humans learn to speak from our parents, with our mothers playing a particularly important role. It's because of this that we have different languages across the globe, rather than all speaking the same genetically-determined tongue, pre-programmed from birth. Yet, in other animals, communication seems to be primarily genetic, with social learning playing little, if any, role.

There are, of course, exceptions to this. We know, for example, that songbirds raised in isolation produce much simpler songs than those in a natural environment, able to hear, and respond to, other birds of their species. On the other hand, they do still sing, and the songs are much closer to normal ones than a similarly deprived human would be able to do with respect to English, French, or Gujarati.

Well, we're not songbirds, but we are mammals, and more specifically, anthropoid primates. Is there really such an absolute split between us and our close relatives? The calls of many primates can be quite complex, and this surely has something to do with why language became so important to us. Are these calls entirely genetically programmed, or are they, in some way, learned?

Among the most vocal of our fellow primates are the gibbons. Gibbons are apes, rather than monkeys, and live only in the jungles of Southeast Asia. (Cladistically speaking, of course, apes are monkeys, albeit of a special kind, but I'm using the vernacular sense, here). There are several species of them, and together they form one of the two families of apes. We, along with chimps, gorillas, and orang-utans, belong to the other family, that of the "great apes", and gibbons are sometimes called "lesser apes" to distinguish them from their larger cousins.

Gibbons have evolved to be so vocal in part because they live in such densely vegetated areas, and, crucially, do so as nuclear families, not as large troops. They're monogamous, so, apart from any children, there's likely no more than two gibbons in a particular area. This makes it particularly hard for them to see one another from a distance, and, like any happily married couple, they want to keep in touch.

They do this by singing duets to one another through the trees, generally taking it in turns. Even in zoos, where they are generally in larger groups, and not able to travel far out of eyesight, gibbon pairs that spend the most time together and show the most visible signs of affection towards one another, are also the ones that sing to each other the most. The songs are highly complex, especially in the females, with an introductory part followed by a series of "great calls" separated by organising sequences. Indeed, they are complex enough that it's possible to identify an individual gibbon by its call, which, presumably, is part of the point.

Do they somehow learn to sing, or is it all just programmed behaviour? Certainly, infants and adolescents don't sing, instead making relatively simple sounds to alert their parents. Although they practice rather earlier, it's only by the time that they reach about six years of age that they've really got the full repertoire of adult singing down. This, it's worth noting, is about two years before they leave home to find a mate, and, during those years, daughters, at least, are duetting with their mothers.

This, however, does not prove that they're learning from them. In fact, there is clear evidence of a strong genetic component in gibbon calls, which, in part, ensures that different species don't sound alike. One way we've tested this is to look at hybrids born in zoos, with parents belonging to two different, but closely related, species. If the calls are genetic, then hybrid daughters should sound different to either of their parents, no matter how much they apparently 'practice' singing with their mother. And this is exactly what we see.

So, if they aren't learning, why do they bother? To find out, we need to understand more about the nature of these interactions. A recently published paper by Koda Hiroki and co-workers describes the results of the first such study. They looked at agile gibbons (Hylobates agilis), a species native to Sumatra and parts of peninsular Malaysia. (Possibly also Borneo, although these are now sometimes thought to be a different species). To do this, they recorded the calls within six families of gibbon living in the jungles of western Sumatra, each of which included a single sub-adult daughter, along with her parents, and, in some cases, a younger sibling.

You can listen to a short clip from one of the recordings here.

There were considerable differences in the amount of mother-daughter duets between the different families. Perhaps surprisingly, the less a given family sang together, the more similar they sounded, and the more closely synchronised their singing was. That's the opposite of what we'd expect if learning was involved: wouldn't the young gibbons sound more like their mothers if they practised together more often?

Over the long term, yes, but this isn't a long term study. We know that, while their songs are still developing, young gibbons sing more and more frequently with their mothers, and that it's always the mother who starts the duet. However, once they get past the age of about six, and have mastered all the basics of the song, they're also beginning to drift apart from their mothers. They sing together less and less, until eventually they stop, leave home, find a mate, and sing to him instead.

We don't know the exact ages of the young gibbons in this study, although they all appeared to be between six and eight years old. But it seems likely that the families that sang together the least are the ones with the oldest daughters. In other words, what we're seeing is that, the older the daughters get, the more they sound like their mothers. They may be singing together less now, but only because they've already had a lot of practice in the past, and are not only sounding more 'adult', but also, specifically, more like their mother than other random females in the area.

This doesn't prove that they did so by learning, mind you. After all, any daughter shares half her genes with her mother, so if the whole process is genetically controlled, you'd rather expect her to sound like her mum. But, if so, it seems odd (at least to me) that she starts out sounding different, and gets more like her mother as she ages. To know for sure, we'd have to listen to orphans, and see what they do, but it seems plausible that there really is social learning going on here - the daughter is learning how best to sing from her mother.

Another pointer in this direction is that mothers sing differently with their daughters than they do with their mates - at least at first. According to the study, the less they sing together (and thus, we assume, the older the daughter is) the more complex the duets they do sing become. It's as if the mothers start out singing to their daughters in "baby-talk", and move on to more 'adult' songs as their daughters grow up. It's also notable that, as the daughters get older, it's increasingly likely to be them, not the mother, that starts the duet - perhaps they're becoming increasingly independent.

This doesn't contradict the earlier studies that showed the importance of genetics. Those looked at the overall pattern of the songs, which certainly do seem to be in-born. This newer study is looking at the fine detail, the sort of differences that enable gibbons to tell one individual from another, not simply whether they're the same species or not. Perhaps the hybrids in the earlier study were trying to sound more like their mothers, but, due to their genetic limitations, couldn't quite get there. We may never know.

There's also the question of whether fathers do the same with their sons. The songs of males are less sophisticated than those of females, so it would be harder to tell, and so it's not surprising that the first study focussed on the females.

As I say, we'll need more studies to really get the full picture. Following some families over the two period it takes the daughters to mature would, for instance, confirm that the changes really are due to increasing age and experience. But it's a start, and the parallels with "baby-talk" may show some similarities with how human speech develops.

Gibbons are, after all, closer to us than they are to monkeys.

[Picture by Julie Langford, from Wikimedia Commons]