Why does the chimpanzee vocal repertoire remain poorly understood and what can be done about it?

Abstract

After decades of research, the origins of human speech remain little understood. One potential problem is that the vocal repertoires of humans' closest living relatives, the apes, remain poorly described. Given that the evolution of language has left few fossils, many researchers interested in this question adopt a comparative approach, examining differences and consistencies between human and animal communication. However, comparisons will remain limited in the absence of a comprehensive analysis of the vocal repertoires of the other extant apes, especially of our two closest living relatives, the chimpanzee and bonobo. After years of observing and conducting field experiments with both western chimpanzees in the Taï Forest, Ivory Coast, and eastern chimpanzees in the Budongo Forest, Uganda, I posit several reasons why a comprehensive analysis of the chimpanzee vocal repertoire has not yet been completed, in spite of 45 years of research, and what can be done to remedy this situation. I raise four main problems and later present potential solutions: a) How do we categorise graded calls into information-specific units? b) Why are most call types emitted in most contexts? c) If different call types are combined flexibly within the same context, can calls be tied to emotional states? d) The constraintsand advantagesimposed by dense forest habitat. I also tabulate cross-site consensus in call categorisation, associated contexts of usage, and potential call functions. I also note cross-site variation in presence and absence of vocalisations. Anecodote After an intense three years observing Taïi chimpanzees and listening to their vocalisations for my PhD, I was not to return to the Taï Forest for 17 years. On my first day back with the North Group, after walking through the forest in the dark, I arrived at a group of females as dawn was breaking. I heard through the forest a pant-hoot so familiar, a shiver of pleasure ran down my spine. This was the voice of Narcisse, one of the youngest mothers in the North Group 17 years earlier. Three minutes later, Narcisse walked into view with a baby on her back, and with her adult daughter, granddaughter, and adult son close behind her. This whole remarkable family had survived the devastating epidemics that had ravaged the group in the intervening years. The pattern of her distinctive pant-hoot had not changed in 17 years, likewise those of the other three adult females I had known 17 years earlier, Mystere, Perla, and Belle. Even though the pant-hoot is emitted in a number of contexts, shows greater within than between individual variability, and shows group and population differences, something in the quality and the rhythm of the call is retained year in and year out and as with human voices, is a robust marker of individual identity. This begs the question, why do chimpanzees have such a robust marker of individual identity? Why do they, nonetheless, express thisand other calls-with such extensive acoustic and contextual variability?

The development of communication in alarm contexts in wild chimpanzees

“…2 a Alarm call produced by individual JS (juvenile). Alarm call of young individuals resembles hoo alarm calls produced by adults: they are soft, tonal and low in frequency (Schel et al 2013; Crockford 2019). In adults, repetition rates for hoo alarm calls are relatively slow (Crockford et al 2018).…”

“…Chimpanzees ( Pan troglodytes ) produce alarm calls to a range of disturbances, but previous studies have mainly focussed on a single class of threats—snakes. Chimpanzees emit three alarm call types to snakes: acoustically variable hoo alarm calls (different from hoos in non-predatory contexts; Crockford et al 2018) (Schel et al 2013), waa-barks and SOS screams (Goodall 1986; Zuberbühler 2000; Crockford and Boesch 2003; Schel et al 2013; Crockford et al 2018; Crockford 2019). The current literature suggests that these three call types are not strictly predator-specific, as demonstrated in various monkey species (e.g.…”

“…Young chimpanzees show somewhat different vocal behaviours to dangers, by producing ‘hoo’ and ‘whimper’ calls (Plooij 1984; Plooij et al 2015). Although the vocal behaviour of chimpanzees in threatening contexts has been the topic of much recent research (Crockford and Boesch 2003; Crockford et al 2018; Crockford 2019), we are not aware of any systematic efforts to investigate the vocal behaviour of immature individuals. In adult chimpanzees, alarm call production is known to depend on a number of social parameters, such as presence of specific group members and their attentional and epistemic states (Crockford et al 2012, 2017; Schel et al 2013), suggesting that social competence (i.e.…”

Chimpanzees use numerous flexible vocal sequences with more than two vocal units: A step towards language?

“…Vocal sequences make up towards 50% of their call output and regularly involve sequences of more than two different types of vocal units 40,41 . Chimpanzee single vocal units have a relatively high degree of context-specificity across a relatively broad range of contexts: alarm, recruit, hunt, food, submissive greeting, rest, travel, nest 42 . Whilst combinations of two meaning-bearing vocal units are not uncommon across mammals and birds, these tend to be limited to alarm or recruitment contexts 24,28 .…”

“…We also assessed the accuracy of the classifications using acoustic analysis on a subset of calls and discriminant function analysis. To limit classification problems arising from the highly acoustically graded chimpanzee repertoire, we chose to classify broad call categories that show consensus across studies and chimpanzee populations (reviewed in 42 ; Table 1, Fig. S1) and the acoustic properties of which discriminate well in discriminant function analysis (Grawunder et al In revision).…”