We thank Rendell and Whitehead for their kind words on our study, and we welcome the opportunity to discuss how sperm whales communicate with clicks.
When dedicated soccer fans of Manchester United and Liverpool enjoy a lunch at home with their families before a game, their quiet conversations to mediate family unit social interactions have a small active space of a few tens of meters to address the nearby audience around their kitchen tables. When the same fans from each club later join each of their fan groups to walk to the stadium for the Sunday match, they employ a very different active space of many hundreds of meters by loudly announcing their football clan affiliation with easily identifiable club-specific songs passed down through generations. Such clear culture-specific vocal displays of ‘us’ versus ‘them’ have likely been with us for as long as we have been humans and beyond. In their paper from 2003, Rendell and Whitehead (2003) lay out a similar scenario for sperm whales by proposing that the Morse code-like click patterns of sperm whales, so-called codas, serve as acoustic markers of cultural affiliation that are communicated between social units to inform decisions of whether to join. Specifically, based on the idea that codas give sperm whales a cultural identity, they write, ‘We suggest that coda dialect performs a signature function in this context, allowing units to identify other units of the same clan within a highly mobile sperm whale society…’. In our recent paper on the active space size of sperm whale codas (Jacobs et al., 2024), we motivate parts of our testable hypotheses by that statement when writing ‘…the proposed long-range coda communication for broadcasting clan identity beyond the unit (Gero et al., 2016a,b; Rendell and Whitehead, 2003).” Specifically, if sperm whales seek to communicate their cultural identity beyond their social unit as proposed by Rendell and Whitehead (2003) to mediate the social assortment based on clan, it seemed obvious to us when citing Rendell and Whitehead (2003) that the active space of codas must extend beyond that of the separation of animals within a unit.
However, in their reply to our paper, Rendell and Whitehead correctly note that they never explicitly wrote that codas serve as ‘long range’ identity signals, and proceed to present a hypothetical scenario of how codas could mediate between-unit recognition without requiring a median active space of codas larger than ∼4 km. Specifically, they speculate that when a unit hears the loud echolocation clicks of foraging sperm whales (Watwood et al., 2006) from another unit, it may then approach within the detection range of coda clicks, with individuals in the outskirts of the two units being able to hear codas or, perhaps, exchange codas, to allow for identification of their clan affiliation when they are close enough to hear one another.
However, because the vast majority of codas are not produced when whales in a unit are separated for foraging (Oliviera et al., 2016), whales foraging in the outskirts of each of two social units are unlikely to produce enough detectable clan-identifying codas within the time frame of the two units passing each other: sperm whales tend to move horizontally about 4 km per hour (Jaquet and Whitehead, 1999), so for two units of foraging sperm whales passing each to produce and hear enough codas to actually be able to assign clan membership from them requires the two groups to be very close. Thus, our findings and the additional supporting evidence for a small active of space codas provided by Rendell and Whitehead (Weilgart and Whitehead, 1997; Whitehead, 2003) make us agree that sperm whale codas can be used to exchange clan-identifying coda information between units, but only if the units are very close together, as codas offer an ineffective way for a unit of sperm whales to radiate clan affiliation much beyond the footprint of the unit.
Irrespective of how stimulating discussions of such hypotheticals may be, our paper and the Correspondence by Rendell and Whitehead must face the fact that we have few quantitative data on the spread amongst individuals within a sperm whale unit as they navigate together and even less on the typical distances and movements between units over larger scales. To further compound that problem, both of these distances or spreads likely vary between clans and ocean basins based on the differences in group sizes and movement behavior which have been demonstrated (Whitehead and Rendell, 2004; Whitehead et al., 2012; Vachon et al., 2022). In part, as Rendell and Whitehead point out, this problem arises because these spatial scales are larger than any one research vessel can address.
While our study demonstrates that most codas only reach a small audience within a given social unit, it also suggests that sperm whales may be biomechanically capable of producing these codas much louder. This begs the question of why they don't, if this is important for mediating clan affiliation and social dynamics between units as they encounter one another. In turn, as Rendell and Whitehead write in their Correspondence, how do sperm whale units of different clan dialect ‘almost never, if ever, form coordinated groups’ if they are not using coda information specifically to make decisions about with which units to associate? For us, this is the crux of the science problem at hand and we posit that our study takes a critical step in answering this. It is still unknown what features of codas might encode identity or how they mediate social interactions, nor do we understand how many codas are required for recognition of individual, unit or clan, if exchanges between nearby units are even needed, or if recognition and social decisions can made passively based upon only hearing distant codas from other units.
Based on the current approach to communication research in this species, it seems unlikely that we can truly quantify the rate of encounters between units, or the distance over which social decisions are made with regard to in-group biased associations which produce the emergent patterns of social segregation between clans. Research conducted on a single vessel, with acoustic equipment which likely can only detect the louder echolocation clicks across less than 10 km, cannot study these patterns across a biologically relevant scale. Particularly given that most acoustic tracking protocols are biased towards staying with the loudest echolocation clicks during a ‘group follow’, especially when tracking overnight when no other cues are available, that means that the research vessel is likely to remain closer to the center of a unit. This biases our observations to remain with the focal unit and dramatically increases the chances of missing a potential ‘swim-by’ of other animals even within range of detecting their echolocation, or perhaps even the active space of codas; this creates an apparent lack of interactions without any real empirical studies to directly test for it. Research involving wide-scale hydrophone arrays, multiple coordinated research platforms or concurrent long-term GPS and sound recording tags deployed on multiple sympatric units is needed to study the movement and distribution of sperm whale units within the spatial and social context discussed here.
In conclusion, the last 25 years of research has provided strong evidence that codas function to mediate social interactions between sperm whales, but how and at what scales physically or within the multilayered social structure they are exchanged remains to be understood. Our study and the present exchange of views with Rendell and Whitehead will hopefully motivate future coda playback and large-scale tracking studies to provide a more complete understanding of the functional use of codas and the socio-spatial dynamics of these top predators in ocean-scale ecosystems.
