Regularity and Variation in the Songs of Humpback Whales
“What an odd thing it is to see an entire species — billions of people — playing with, listening to, meaningless tonal patterns, occupied and preoccupied for much of their time by what they call ‘music.’”
— Oliver Sacks, Musicophilia (2008)
Regularity and Variation
Humpback whale (Megaptera novaeangliae) songs, as part of their sexual allure, emphasize both regularity and variation at the same time and weave them together in intricate patterns of sound. These two simple perceptually opposed qualities — order and disorder — have not been given enough consideration as factors in the decisions animals make and the structure of aesthetic phenomena. Along with other simple opposites like high- and low-pitch, continuity and discreteness, and sound and silence, these constitute common, essential features of all songs and music in humans and other singing animals. A sound sequence with perfect regularity or order over time, like an alarm, is not a song, and neither is acoustic chaos. Order and disorder interact over time on every scale in whale songs, and, therefore, in the mating decisions whales have been making since they started singing millions of years ago.
Payne and McVay (1971), in a seminal publication on humpback songs, say they first impress one with “almost endless variety,” but that spectrographic analysis reveals “fixed sequences” and repetition with “considerable accuracy.” They broke songs into a hierarchy consisting of units, phrases, and themes, which has become the standard method of analysis.
Units, phrases and themes each vary throughout a song (sequential variation) and each tends to change over time (temporal variation), so much so that it may become impossible to relate new sounds back to their progenitors (Mercado 1998). Each also varies between individuals and populations (Cerchio et al. 2001). Units are like notes in human music. They’re aurally continuous with frequencies ranging from 30 Hz to more than 10,000 Hz. Phrases are defined as various distinct patterns of notes, which may contain 2 to 20 or more and last between about 5 and 30 seconds. Themes usually consist of a series of similar phrases. Theme sequences sung repeatedly are considered distinct songs (Cerchio et al. 2001). Each level of song structure is simultaneously static and dynamic.
Units may split up into several frequency bands throughout a song (Cerchio et al. 2001). Length of time between units may change throughout a phrase (Cerchio et al. 2001). Units may be added, deleted or modulated to make up different phrases (Mercado 2003). Phrases can rise or fall in frequency or be modified in frequency range or duration (Cerchio et al. 2001). Their number per theme can vary in successive songs (Payne 1971), and it appears that most phrase types will tend to change over a period of a few months (Cerchio et al. 2001). Themes may differentiate into multiple themes over time (Cerchio et al. 2001), vary in how phrases within them change, and they’re left out of songs with different probabilities per theme (Cerchio et al. 2001). Of six themes recorded over a three month period (Payne and Payne 1985) four were found to be “static” (phrases remain the same throughout the theme) and the other two were “shifting” (phrase type changes progressively over the course of the theme). Songs change continuously, but rate of song change may vary (Darling et al. 2014). Noad et al. (2000) report the evolution of a completely new song over a two-year period.
Mercado et al. (2003) use the term sound pattern to refer to “any sound or set of sounds that is consistently repeated within a song session.” They found that some sound patterns recur across populations and can last for decades. Among sound pattern types that ranged in number of units from 2 to 12, only two were consistently recorded over a four year period. All others were modified in terms of the number and characteristics of the units making them up. A “song session” is made up of a continuously repeated song. Consecutive songs in a session typically differ in duration and repetition of phrases, and sometimes in theme composition (Mercado 2005).
Thus it’s evident the songs of humpback whales are both stable and dynamic, both repetitious and variable, at every level within the songs themselves (units, phrases, themes, songs, sound patterns and song sessions), over every period of time (days, months, years and longer), and between individuals.
Are humpback “songs” really songs?
Payne and McVay (1971) justified their use of the term “song” to describe humpback whale vocalizations by reference to definitions given by Broughton (1963). Broughton gave three definitions of song: Sensu latissimo or “sound of animal origin which is not both accidental and meaningless,” sensu stricto or “a series of notes, generally of more than one type, uttered in succession and so related as to form a recognizable sequence or pattern in time,” and sensu strictissimo or “a complete succession of periods or phrases.” Payne and McVay particularly cite the “median,” sensu stricto, but say that Humpback whale vocalizations also follow Broughton’s other two definitions.
Evidence that humpbacks really sing includes the fact that we have conventionally agreed to call certain of their vocalizations “songs,” and, as noted by Silber (1986) the fact that communicative vocalizations occur throughout the year but singing is much more common or restricted to the breeding season, that songs are rhythmic and continuous compared to other types of vocalization (as in human repertoires), and that, in contrast to other vocalizations, whales produce songs when they are alone. Notice that these same criteria might be used to separate human singing from other human vocalizations. We are conventionally aware of the difference between song and prose. Music and song are more closely associated with pleasure and mating than work, and human singing occurs with relative spontaneity, rather than in direct and immediate response to stimuli.
Green and Marler (1979) distinguish animal sound repertoires into “discrete” and “graded” categories. Discrete repertoires are made up of a set of relatively distinct, stereotyped sounds. Graded repertoires contain sounds with more variation and more continuity between sound types. For most animals repertoires are a mixture of these types (Green and Marler 1979). Discrete repertoires are more useful for long-distance communication than graded repertoires (Mercado 1998:88). That humpback songs fall into the graded category (Mercado 1998:105), and therefore are not optimal as a communicative device over the long distances of sound travel between whales, is a further reason to consider them worthy of being termed songs.
What is the function of Humpback whale songs?
Songs of humpback whales are considered to function in attracting mates (Cerchio et al. 2001, Clark 2004). Cerchio et al. (2001) relate Katharine Payne’s (2000) hypothesis that females may choose male songs (and hence males) based on both their conformity and innovativeness. In this case, males who tend to repeat songs from the population but also add to them some degree of variation would be at a selective advantage, and their songs, or sounds, would tend to increase in complexity over time. Researchers have suggested they contain information that conveys the singer’s degree of fitness (Frankel 1995) in terms of size (Darling and Berube 2001), reliability of information content in songs (Winn and Winn 1978) or the ability of a male to hold its breath (Chu and Harcourt 1986).
However, fitness, or survival ability for a humpback whale might involve ability to locate food, assimilate nutrients, resist predation, resist parasitic infections, swim fast, maneuver, hold its breath, tolerate extreme temperatures, sense features of its environment, live long, learn, or even to make intelligent decisions and probably many other things. One might think of each of these abilities as parameters to be potentially considered, or measured as fitness cues, by females when they choose mates from among a set of males. And for each parameter, intuitively at least, it seems there should be more direct ways for males to demonstrate fitness and/or females to gauge mate fitness than by way of song. Why, for instance, should ability to locate food be communicated by a song when a female can simply observe the size and health of a male to gather the same information more reliably, and without the potential for cheaters? Why would songs become vectors for fitness information that should be apparent to females through so much more direct and reliable channels as physical appearance, agility, or social prowess?
How could the many, objective, observable, fundamental realities concerning male humpback whale fitness, the features that have led to their present existence and promote their future survival, ever come to be encompassed or relayed in a subjective, relatively faith-based, abstract communication channel such as a song? And if songs do convey information, why are they constantly changing? Every alteration should undermine whatever information was previously available to be gathered from the sound pattern before it changed. Assuming the patterns don’t matter, and that females are only using the amount or duration of songs in their decisions completely ignores and leaves to be explained the initial problem of song complexity.
Music in Humans
Human music and song are functional in various ways. We use it in unifying the people of a state (e.g., national anthems) or coordinating behaviors (e.g., marching). The applications we’ve found for music are often thought of as the reason we like it, but a much simpler and more reasonable interpretation of human musicality is musical enjoyment. The rhythm and collective enjoyment of music have probably been co-opted to serve various secondary purposes. The idea of songs helping to keep marchers or people rowing boats in sync doesn’t explain its non-rhythmic elements, and it assumes differential survival. Soldiers marching and rowing without singing would have to die faster than singing soldiers, for instance. Such a scenario would seem to require large portions of human populations to be singing in groups while engaged in dangerous activities for a long time in order to get the result that singing is built genetically into humans, and the most difficult issues would remain unaddressed. Why do we sing mostly for fun? Why does it sound good? Why do we do it mostly in noncombat situations, listen to it alone, and dance to it? Why does it vary and why do thousands of other animals use it to impress mates if it arose in humans for a completely different reason? Also, song has a scale-free character that other communication does not. It’s ordered and disordered over time on every scale, as revealed by the humpback song research referenced above. Human music has the same kind of scale-free character, itself seemingly in need of an explanation having little to do with any survival advantage humans might get from singing in groups.
Discarding direct survival advantages leaves us with the option that people are impressed by singing and tend to mate with those who do it better, like humpback whales, but it’s hard to imagine how our appreciation for music and its ubiquitous use in current times in nonsexual contexts arose from such a relatively direct and confined application as winning a mate. Music evokes strong emotions, and its lyrics are often filled with amorous content, but we use it for much more than sex. In other animals only the males sing, usually, a reflection of the behavior evolving by female choice, but this is hardly true in humans, undermining the case for a mate choice mechanism. Even if human singing did evolve through sexual selection, the question of its structure, the way it couples order with disorder over time and why we prefer it to noise would be entirely unanswered. Most likely something about the fundamental structure of the brain makes us more receptive to musical patterns and less receptive to aural monotony or discord (non-musical sequences of sound).
None of the proposed functions of song explain why a female, or any whale, or any animal would prefer song-like sequences of sound. Perhaps whales inherently experience, and enjoy, mixtures of mildly surprising and familiar stimuli the same way that humans do. Perhaps cetaceans are not so monstrous, and given an evolutionary period of freedom from the threats of predation and starvation, cognitive animals are prone to play, like we do ourselves, with patterns that reflect the complexity of our cognitive organs.
If the reason for the evolution of song in both human culture and humpback whales is choice of sound patterns that are pleasurable (musical), then it’s natural to hypothesize that some common element in the physical structure of human and humpback brains is the ultimate reason. The deep structure of the brain is, after all, one of few common elements other than singing behavior which unites humans, gibbons, whales, birds (about 5,000 singing species), toads, insects (Hartshorne 1973) and mice (Holy and Guo 2005). Deep brain structure is liquid crystalline, a state of matter balanced in a condition of intricately mixed order and disorder, or regularity and randomness, very similar to a song.
Darwin (1871) theorized that certain elaborate male traits, such as, presumably, singing, evolve by way of an intrinsic aesthetic sense. Accordingly, the more recent sensory bias hypothesis holds that, sometimes, female preference for a male trait evolves before the trait itself. Females may have a “preexisting bias” for certain male traits, which are “intrinsically stimulating” because of female sensory system organization. This is supported by sensory physiology studies and behavior, and has been demonstrated in swordtail fish (Basolo 1995, 1998), Tungara frogs (Ryan and Rand 1990) and a growing number of other species.
Mercado (1998) points out that the organization of sounds produced by animals can be functional, and/or may reflect the physiology of the organs that produce them. By way of sexual selection, perhaps they also reflect the physiology of the organs that perceive them (e.g., female humpback whale brains). Wallin (1991) observes that the “dynamic dichotomy in music is similar to that which characterizes organismal systems,” and goes on to propose a “morphodynamic isomorphism” between the form of music and that of its physiological substrate (i.e., the brain).
Disorder and variation appear to be related in a mental category of greater excitement, which also contains several other qualities such as high-pitch, and sound itself, compared to silence. Low-pitched sounds, order and regularity are likewise related in a lower-excitement mental category. Humans are amused by intricate mixtures of these and other more and less exciting things in many contexts outside of music and song. We remove the disorder from our visual environment while at the same time embellishing it with decorations, for instance, and appreciate such mixtures in idiomatic language.
What Wallin called morphodynamic isomorphism is probably closely related to the inevitable, universally liquid crystalline physical state of the brain and sensory system in animals. Due to its liquid crystalline composition, the brain fluctuates on small scales over time between a relatively solid, ordered and regular condition and a relatively fluid, disordered and irregular one. It takes on the former qualities when we sleep and the later qualities when we’re awake. It fluctuates on a daily time scale between a composition with more and then less of the given qualities as its temperature changes with its activity level and our level of arousal. It resists spending long periods in either a more or less excited state, in part by resisting exposure to sensations which push it too far in one or the other physical direction. We prefer a mixture of wakefulness and sleep, or one of more and less excitement over too much of one or the other. A similar preference on a smaller time scale, along with the assumption that higher-pitch is more exciting than lower-pitch in sounds, is probably responsible for the intricate interplay between these qualities in language and words. Brains may embody a preference for song in a similar way, enjoying the way it stimulates sensory material without pushing it too far in the more or less exciting, disorderly or orderly directions on small scales. This would fit with our use of the word “like” to mean both likable and similar. Songs are more similar to brains than noise.
Basolo, Alexandra L. “Phylogenetic evidence for the role of a preexisting bias in sexual selection.” Proc. R. Soc. Lond. 259.1356 (1995): 307–311.
Basolo, A. L. “Evolutionary change in a receiver bias: a comparison of female preference functions.” Proc. R. Soc. Lond. 265.1411 (1998): 2223–2228
Broughton, W. B. Acoustic behavior of animals. Ed. R. G. Busnel. Elsevier: London, 1963. 824–910.
Cerchio, S., J. K. Jacobsen and T. F. Norris. “Temporal and Geographic Variation in songs of Humpback Whales, Megaptera novaeangliae: Synchronous Change in Hawaiian and Mexican Breeding Assemblages.” Animal Behavior 62 (2001): 313–329.
Chu, K., and P. Harcourt. “Behavioral Correlations with Aberrant Patterns in Humpback Whale Songs.” Behavioral Ecology and Sociobiology 19.5 (1986): 309–312.
Clark, C. W., and P. J. Clapham. Acoustic Monitoring on a Humpback Whale (Megaptera novaeangliae) Feeding Ground Shows Continual Singing into Late Spring. Proceeding of the Royal Society of London 271 (2004): 1051–1057.
Darling, J. D. and M. Berube. “Interactions of Singing Humpback Whales with Other Males.” Marine Mammal Science 17 (2001): 570–584.
Darling, James D., Jo Marie V. Acebes, and Manami Yamaguchi. “Similarity yet a range of differences between humpback whale songs recorded in the Philippines, Japan and Hawaii in 2006.” Aquatic Biology 21.2 (2014): 93–107.
Frankel, A. S., C. W. Clark, L. M. Herman and C. M. Gabriele. Spatial Distribution, Habitat Utilization, and Social Interactions of Humpback Whales, Megaptera novaeangliae, off Hawaii, Determined Using Acoustic and Visual Techniques. Canadian Journal of Zoology 73 (1995): 1134–46.
Green, S. and P. Marler. “The Analysis of Animal Communication.” Handbook of Behavioral Neurobiology, vol. 3: Social and Behavior and Communication. Ed. P. Marler and J. G. Vandenbergh. New York: Plenum Press, 1979. 73–157.
Hartshorne, C. 1973. Born to Sing. Indiana University Press, 1973.
Holy, Timothy E., and Zhongsheng Guo. “Ultrasonic Songs of Male Mice.” PLoS Biology. 3.12 (2005): e386.
Mercado III, E.. 1998. Humpback Whale Bioacoustics: From form to function. Doctoral thesis. Honalulu: University of Hawaii Press, 1998.
Mercado III, E., L. M. Herman, and Adam A. Pack. “Stereotypical Sound Patterns in Humpback Whale Songs: Usage and Function.” Aquatic Mammals 29.1 (2003): 37–52.
Noad, Michael J., et al. “Cultural Revolution in Whale Songs.” Nature 408.6812 (2000): 537–537.
Payne, K. and R. Payne. “Large Scale Changes Over 17 Years in Songs of Humpback Whales in Bermuda.” Zeitschrift fur Tierpsychologie 68 (1985): 89–114.
Payne, Katherine. “The Progressively Changing Songs of Humpback Whales: A Window on the Creative Process in a Wild Animal.” The Origins of Music. Ed. Nils Lennart Wallin, Bjorn Merker and Steven Brown. Cambridge: MIT Press, 2000. 135–150.
Payne, R. S., and S. McVay. “Songs of Humpback Whales.” Science 173.3997 (1971): 585–597.
Payne, K., P. Tyack, and R. Payne. “Progessive Changes in the Songs of Humpback Whales (Megaptera novaeangliae): A detailed Analysis of Two Seasons in Hawaii. Communication and behavior of whales.” Ed. R. Payne. Boulder: Westview Press, 1983. 9–57.
Ryan, Michael J., and A. Stanley Rand. “The Sensory Basis of Sexual Selection for Complex Calls in the Tungara Frog, Physalaemus pustulosus (Sexual Selection for Sensory Exploitation).” Evolution. 44.2 (1990): 303–314.
Sacks, Oliver, and Sacks, Oliver W. Musicophilia: Tales of Music and the Brain. United States, Vintage Books, 2008.
Silber, G. K. “The Relationship of Social Vocalizations to Surface Behavior and Aggression in the Hawaiian Humpback Whale (Megaptera novaeangliae). Canadian Journal of Zoology 64 (1986): 2075–2080.
Wallin, Nils L. Biomusicology: Neurophysiological, Neuropsychological and Evolutionary Perspectives on the Origins and Purposes of Music. New York: Pendragon Press, 1991.
Winn, H. E., and L. K. Winn. “The song of the humpback whale Megaptera novaeangliae in the West Indies.” Marine Biology 47.2 (1978): 97–114.