Strange seal sounds: Claps, slaps, and multimodal pinniped rhythms
- 1Max Planck Institute for Psycholinguistics, Netherlands
Beyond vocal communication
Within mammalian bioacoustics, vocal communication has received much attention. Efforts to understand sound production often focus on sounds generated via apparatuses that specifically evolved to phonate, such as the larynx. However, while mammals mostly perceive sounds via one organ, the ear, they can produce sounds via limbs, tails, flippers, tools, and several other mechanisms which, at first, may not seem to have primarily evolved for sonation (Frankel, 2009; Tyack & Miller, 2002; Clark, 2016). For example, kangaroo rats drum their foot to communicate (Randall, 1984), while non-human primates drum using artificial tools (Remedios, Logothetis, & Kaysers, 2009), resonant surfaces (Ravignani et al., 2013), and their hands (e.g., Dufour et al., 2015). Also aquatic mammals can produce a variety of non-vocal sounds (such as whistles, snorts, and others; Tyack & Miller, 2002). These sound production modes may enable communication even when laryngeal phonation is ineffective or impaired (Partan, 2017; Munoz & Blumstein, 2012). Research on sound production beyond phonation is key to properly characterize the richness of animal communication.
Recent exploratory work (Hocking et al., 2020) provides an example of non-vocal sound production in a pinniped: grey seals (Halichoerus grypus) clapping their fore flippers underwater, a signalling behaviour previously attributed to vocalizing. While Hocking and colleagues’ observation is limited to few events, it is reminiscent of previous, seemingly unrelated work reporting water-slapping behaviour in other species, including a close relative, the harbour seal (Phoca vitulina) (Venables & Venables, 1957; Hanlan, 1998; Hayes et al., 2004; Newby, 1973; see also humpback whales: Dunlop, Cato, Noad, 2010). The preliminary data reported by Hocking and colleagues naturally invite a host of questions, whose answers rely on a characterization of when, how often, and under what circumstances these claps occur. Are they a frequent or seasonal phenomenon? Are they modulated by social context? Furthermore, the mechanism of knock production should be considered: while clapping the fore flippers can generate loud knock-like sounds, other mechanisms have also been proposed (e.g., in walruses: teeth clacking, tongue movement, or suction; Sjare & Stirling, 1981; Reichmuth, Mulsow, Schustermann, 2009; Sjare, Stirling, & Spencer, 2003; Larsen & Reichmuth, 2012). This is a good starting place for future research, though, clearly, more observations are required to answer these questions. In particular, the field should perform more empirical, foundational work. This should 1) provide robust observations and descriptions in addition to anecdotes, 2) evaluate context, timing and seasonality in the production of percussive sounds, 3) determine which sex produces these signals and potential sexually dimorphic characteristics, and 4) design rigorous experiments that test potential function of percussive sounds. Once these absolutely necessary foundations are established, we suggest to explore more complex topics related to bioenergetics, signal evolution, multimodality, and rhythm production/perception. In this Opinion piece we discuss these more hypothetical research directions, which however can only be performed after more thorough biological descriptions of the basic phenomena.
Mechanisms for (dis)honest signalling?
If replicated, this research may encourage to re-evaluate previous evidence. Could previously recorded “vocalisations [which] sounded like a loud piercing clap” (pg.61, McCulloch, 2000) and “knocks” (pg.2213, Asselin et al., 1993) in grey seals have been actual underwater claps? A hypothetical reassessment of the production mechanism underlying a sound would entail several implications. Awareness of the sound source could be helpful to test potential sound-body allometric links.
One testable hypothesis is that claps may be a partly dishonest signal as they give away limited information about body size while their source level is surprisingly high, especially compared to captive individuals (Wahlberg, Lunneryd, Westerberg, 2002). This hypothesis dovetails with some empirical evidence of water claps functioning as aggressive and territorial behaviours in harbour seals (Hayes et al., 2004), since claps have been so far observed mostly in males and in presence of other seals (for example in Weddell seals: Russell, Purdy, and Davis, 2016). Furthermore, limited underwater visibility, as reported by Hocking and colleagues (2020), would promote a signalling strategy concealing body size. To test whether clapping is a dishonest signal, field studies should investigate the conditions under which underwater claps take place (e.g., water visibility, social context).
Conversely, claps could be honest signals, since the strength of the animal or the size of its flipper may determine the intensity of the clap and he perceived loudness. This second hypothesis would comply with allometric scaling: if sound-producing structures scale with body size, honest signalling ensues (Garcia & Ravignani, 2020). Larger individuals should also be able to produce stronger or (visually) larger claps (Partan, 2013). Anatomical observations in, and tests of allometry across, individuals producing underwater claps might contribute in disentangling these contrasting hypotheses.
Exaptation and repurposing of biomechanical processes
This research underlines the process of evolutionary exaptations for communicative purposes (Gould & Lewontin, 1979; Buss et al., 1998). Seal clapping “may be a ritualized version of a swimming stroke” (pg.1, Hocking et al., 2020). In other words, a movement evolved for essential in-water displacement (Fish, 2000; Kuhn & Frey, 2012) may have been repurposed for acoustic communication (see also Clark, 2016). As a parallel, the tree drumming of woodpeckers is now a purely communicative signal, probably repurposed from what originally was a simpler foraging behaviour (Dodenhoff et al., 2001; Garcia et al., 2020; Miles et al., 2018; Miles et al., 2020). Both seals and woodpeckers may provide fascinating examples of the evolution and repurposing of biomechanical processes. At what point in pinniped phylogeny may swimming strokes have been recruited for communication, and did this happen in multiple pinniped species? Communicative non-vocal sounds have been observed in several pinnipeds (Schustermann & Van Parijs, 2003; Purrel et al., 2016 ); more specifically, claps have been reported, to our knowledge, in grey seals (e.g., Hocking et al., 2020), harbour seals (Venables & Venables, 1957; Hanlan, 1998; Hays et al., 2004 Newby, 1973), and walruses (e.g., Reichmuth et al., 2009). Underwater recordings of similar non-vocal sounds in other pinnipeds might contribute to answering this question. With limited evidence it is difficult and unwise to generalize to all pinnipeds. However, one may hypothesize that the swimming style of phocids, propelled by their hindflippers, may free up their foreflippers for communicative purposes, while the swimming style of otariids, using their foreflippers to ‘fly’ underwater, may have hindered their exaptation for communication (Kuhn & Frey, 2012).
Multimodality and energetics
Most work mentioned above advocates multimodal approaches to communication. Multimodality is sometimes neglected, with some research programs only focusing on one production and one perception channel (Slocombe et al., 2011). Grey seals exhibit a communicative behaviour that is motorically produced (bypassing specialized laryngeal neurons) and might be perceived acoustically at long ranges and visually at short ranges (Hocking et al., 2020; Ravignani et al., 2016; Wahlberg et al., 2002). As the larynx is hidden from sight, the act of mammalian vocalization is invisible to the receiver (cf. Fitch & Reby, 2001; Higham & Hebets, 2013; Nowak, 2020). A clap, instead, could potentially reach the receiver visually, acoustically or haptically, also allowing for multisensory integration.
Multimodality also entails energetic considerations. While laryngeal phonation is relatively cheap, other modes of controlled sound production may be more energy-expensive. Indeed, research on the energetic costs of communication generally assumes higher expenditure for multi-modal as compared to uni-modal interactions (Partan , 2013). For example, in sympatric wolf spiders, multi-modal displays (as in Schizocosa ocreata) require higher energy levels than unimodal displays (as in S. Rovneri) (Cady, Delaney, Uetz, 2011). Rather than being disadvantageous, such a costly display might serve as an honest signal indicating a male’s good condition (Byers, Hebets, Podos, 2010; Mitoyen, Quigley, Fusani, 2019; Zahavi, 1975). Yet, questions concerning the relative costs and benefits of pinniped sound production via clapping and slapping (Beier & Wartzok, 1979; Gillooly & Ophir, 2010; Wahlberg et al., 2002) remain open, as well as their function (e.g., territorial, reproductive, etc.; Purrell et al., 2016). If they indeed relate to mating, one may expect a variation in clapping/slapping abilities due to ontogeny (Rado et al., 1991), an increase at puberty onset, a cyclical variation entrained with mating seasonality or a decline due to senescence (Soulsbury & Halsey, 2018). The reliable evidence of such seasonal and developmental trends in the vocal displays of seals (e.g., Reichmuth & Schusterman, 2009; van Parijs, Hastie, Thompson, 1999; Galimberti, Sanvito, & Miller, 2008) may invite comparisons with non-vocal displays and facilitate the understanding of their function. By combining quantitative techniques (e.g., Gillooly & Ophir, 2010), allometric considerations (Garcia & Ravignani, 2020), and field observations (Hocking et al., 2020), some of these questions may be addressed.
Communicative rhythms in the millisecond-second range
More observations on the clapping behaviour of grey seals, their characteristics, and context of use are needed: Are these sounds produced occasionally or routinely? Do they contain rhythmic components? The presence of rhythmic features in claps would allow to link Hocking and colleagues’ (2020) finding to research on communicative rhythms and could spur a subfield of ecologically-relevant percussive rhythms in mammals. Recently, cross-species evidence has shown rhythmic capacities, sometimes employed for communication, in pinnipeds (Cook et al., 2013; Mathevon et al., 2017; Ravignani, 2019; Rouse et al., 2016). ‘Rhythm’ is not meant here in its circadian sense, studied for instance in ecology, but instead as ‘temporal structure’ at short timescales (de Reus et al., in press). When little information is encoded in the frequency domain, as in seals’ claps and slaps (Hocking et al., 2020; Wahlberg et al., 2002), this temporal structure could emerge in sound signals and serve to encode information. Within animal cognition and behaviour, evidence for rhythm in pinnipeds is particularly interesting (Ravignani et al., 2016; Wilson & Cook, 2016). In fact, pinnipeds constitute a key taxon to test a cross-species hypothesis which links rhythm and vocal learning capacities (Patel, 2006). Still, work on rhythm in mammals is relatively limited, especially when compared to the richness of rhythm production research, for instance, in insects and frogs (Greenfield, 1994; Hartbauer & Römer, 2016).While these species can produce extremely fast rhythms, the rate and complexity of non-vocal rhythms in mammals may be hampered by the physical limitations occurring when moving a limb (but see, for example, Randall, 1984 on the foot drumming behaviour of kangaroo rats). Mammalogists and comparative psychologists may still be inspired and benefit from decades of work on rhythm and percussive behaviour in arthropods and anurans (Ravignani et al., 2014). In particular, since the 1960's, entomologists and then herpetologists have been measuring the communicative rhythms of their species with almost millisecond accuracy (e.g., Buck & Buck, 1968). Avian researchers followed, while mammalogists and primatologists are slightly lagging behind (de Reus et al., 2020). Applying concepts such as phase resetting and period correction, for instance, to communicative rhythms in apes could also inform the evolution of rhythmic capacities in our own species (cf. Bittman, 2020).
Conclusions and future work
Hocking and colleagues’ (2020) finding, albeit preliminary, can inspire at least six hypothetical strands of future work. First, to establish a base from psychophysics, propagation experiments could test how far the sound of grey seals
Keywords: multimodal communication, pinniped, Seal (Pinnipedia), bioacoustics, Exaptation, Honest signaling
Received: 21 Dec 2020;
Accepted: 02 Jul 2021.
Copyright: © 2021 Verga and Ravignani. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence: Dr. Andrea Ravignani, Max Planck Institute for Psycholinguistics, Nijmegen, 6525, Gelderland, Netherlands, firstname.lastname@example.org