</>Portfolio

Abstract

Language is a phenomenon thought to be unique to humans, but researchers have long been interested in the potential linguistic capacities of other animals. The most promising case for non-human language rests with cetaceans, particularly humpback whales. This paper investigated the current state of research in cetacean communications in order to draw a conclusion on whether or not whales have true language. After careful consideration of what is known about whale vocal physiology, whale song composition and structure, cultural transmission in whales, it was concluded that humpback whales do very likely have language.

Introduction

Language as we know it is intrinsically human. Across the animal kingdom, we appear to be the only species which has developed what we know as language. Other species of animals certainly have evolved their own forms of complex communication (i.e. the dances of bees, the intricate chemical pheromone system of ants and some other eusocial insects, the unique call of the Carolina chickadee, the sub-sonic long-range rumbling of elephants, etc.) but human language is set apart both in terms of complexity and structure. There is, however, one other group of animals which seems like it may possess a comparable mode of communication: cetaceans. Specifically, the songs of humpback whale and of the blue whale are of interest, though other cetaceans (sperm whales, belugas, orcas, bottlenose dolphins, et al.) are also promising. This paper set out to investigate the extent to which whales may or may not have true language, and concluded that cetaceans, especially humpback whales, possess language-like communication.

Methods

Materials for this paper were found primarily by keyword-searching databases of academic journals for various phrases such as “cetacean language,” “whale language,” “whale culture,” “whale song,” “humpback whale syntax,” “dolphin language,” and so on. Additionally, filters to restrict results to recent and peer-reviewed publications were used. Several print books were also considered, but only one, Whale Song by Margret Grebowicz, was ultimately obtained and used. One somewhat older source, a literature review on ape language writting in 1998, was used for the preceeding discussion on the linguistic dead-end found in other apes. Due to the various practical, financial, and logistical challenges involved in conducting primary research on cetaceans, no primary research was attempted.

Background

Great Apes

There is some familiarity in the popular awareness with attempts made by various researchers to teach apes, particularly chimpanzees, to use sign language. It is worth addressing these experiments and the results they yielded, since there is a widely held misconception that any of these experiments were successful. They were not. It was found that apes were capable of learning a small repertoire of signs, i.e. a limited vocabulary. The underlying structure of language, however, was never reproduced. The apes could not form sentences, questions, or otherwise productively or syntactically combine the lexical units they had been taught (Hixson, 1998).

Defining Language

Human language is comprised of more than just semantic units; it relies on structure. Specifically, there is a tree-like hierarchical structure thought to be inherent to all human language. Because of this structure, it is possible to arange semantic units in novel ways in order to express novel ideas. Human language, therefore, exhibits both productivity and semanticity. There are some examples of natural communication systems which are both semantic and hierarchically structured which still are not language, however. Birdsong and honey bee dances both convey structured information. However, these systems lack a means of productivity. While bird song can be hierarchically structured, there is no underlying syntax to guide the placement of units within the tree; in otherwords, the hierarchy is essentially arbitrary, and carries no meaning. Honey bee dances are equally syntax-less. This means that, unlike in human language, these and other such animal communication modes are not capable of generating infinitely recursing structures (Cannon, 2023).

Interest in Whale Song

In contrast to birdsong, bee dances, and non-human primates, all of which have been thoroughly examined in the literature since the 1940s, cetacean vocalizations are a relatively recent focus of syntactic examination. Today, precious little is known for certain about whale song, including the whales’ reasons for singing at all. For example, because adult humpback males seem to sing far more often than females or juveniles, it has been hypothesized that singing is part of the mate selection process. However, juveniles and females are also known to sing, and it has been observed that song also mediates male-to-male interactions (Cannon). Despite this gap in our understanding, however, much can be said about the structure of whale song through observation and statistical analyses. That humpback whale song exhibits a hierarchical structure has been known since the 1970s.

Findings

Physiology

There is a significant difference between the methods of vocal production between humans and whales. In human speech, air is pushed through folds in the larynx, which vibrate, and the resulting sound can be modulated into all the sounds used for human language via the system of muscles in the throat and the oral and nasal cavities. By contrast, whales do not expel air in order to sing. Instead, they have U-shaped laryngeal folds which cause vibrations when air is pushed through to a laryngeal sac (an organ unique to whale anatomy, making egressive sounds. The air in the laryngeal sac is then recirculated to the lungs, making ingressive sounds (Cannon). Humans are also capable of making ingressive sounds (such as the sound of a gasp) but while we commonly make these sounds, they are not phonemic in any human language. Other types of contrastive phoneme series, however, are extremely common. Whales using alternating egressive and ingressive sounds in their vocalizations may have some impact on syntax, especially if egressive and ingressive sounds are truly contrastive to the whales, and thus can be used to differentiate between semantic elements.

Hierarchical Structure

The base unit of the humpback whale song is called a “note.” A note is a sound usually 1 to 4 seconds in length, with momentary pause between it and other notes. Several notes together make up a “phrase.” A phrase is usually, though not always, repeated 2 to 6 times, comprising a “theme.” A full “song” is then composed of several themes. The song may take several minutes to complete, and the whale will usually repeat it for several hours or days. This structure of notes comprising phrases comprising themes comprising songs may be compared directly to the human language hierachical structure of phonemes comprising morphemes comprising words comprising sentences.

Statistical Analyses

Humpback whale song contains repeating units (described above) which follow a Zipfian power law distribution, meaning that the frequency of distinct elements follows a power law curve (Arnon, 2025). The most common element is twice as common as the next most common element, which itself is twice as common as the third most common element, etc. Human languages reliably follow this law. Moreover, humpback whale song follows Zipf’s law of abbreviation, another charcateristic of human language (Youngblood, 2025). In a dataset which follows Zipf’s law of abbreviation (such as the lexicon of a language), the most frequent elements are also the shortest.

Culture

Virtually all cetaceans exhibit cultural transmission to some degree; for example, orcas, sperm whales, and bottlenose dolphins all display dialectic variation in their calls between different pods. In the case of sperm whales, this dialectic variation is so pronounced that the term for the social groups of sperm whales in the nomenclature is “vocal clan,” (Grebowicz, 2017). Humpback whales live in less tightly related social groups than those of these other mentioned cetaceans, but cultural transmission in humpbacks is extremely evident. A group of traveling humpbacks generally all sing the same song, and over periods of time, it slowly changes. Sometimes, however, a group of humpbacks will adopt the song of some other group in large part or wholesale, and by this mechanism, cultural transmission can be traced around the globe (Garland, 2017). The hybridization events which occur when whales are adopting a new song also give key insights into acquisition behavior. During hybridization events, the whales either go through a period of splicing complete themes from the revolutionary song into the existing song structure, or they create new, hybrid phrases, containing discrete content from phrases in both songs (Garland). This may hint at the capacity of humpback whale song for productivity, a key component of human language.

Limitations

For now, it is impossible to know what information is carried by the vocalizations of any cetacean whatsoever (Harrison, 2024). In fact, some researchers doubt whether semantic information can be encoded in whale song at all. There is no evidence to suggest that whale song encodes the same type of information that human language does (Youngblood). It has been suggested that the term “song” may have been the most correct analysis all along (Grebowicz). After all, human music possesses many of the same statistical characters that language does, yet is neither semantic nor syntactic. At this time, it cannot definitively be said that this is or is not the case.

Conclusions

Despite the aformentioned limitations, evidence does seem to point in the direction of language. While the usage and purpose of whale song clearly must differ greatly from that of human language, too many indicators suggest that we’re looking at an information-rich communication mode exhibiting both semanticity and productivity to some degree.

References

Arnon, I., Kirby, S., Allen, J. A., Garrigue, C., Carroll, E. L., & Garland, E. C. (2025). Whale song shows language-like statistical structure. Science (American Association for the Advancement of Science), 387(6734), 649–653. https://doi.org/10.1126/science.adq70558

Cannon, Cutler. “A Theoretical Account of Whale Song Syntax: A New Perspective for Understanding Human Language Structure.” Proceedings of the Linguistic Society of America, vol. 8, no. 1, 2023, pp. 5571-, https://doi.org/10.3765/plsa.v8i1.5571.

Garland, E. C., Rendell, L., Lamoni, L., Poole, M. M., & Noad, M. J. (2017). Song hybridization events during revolutionary song change provide insights into cultural transmission in humpback whales. Proceedings of the National Academy of Sciences of the United States of America, 114(30), 7822–7829. https://www.jstor.org/stable/26486118

Grebowicz, Margret. Whale Song. Bloomsbury Academic, an imprint of Bloomsbury Publishing Plc, 2017.

Harrison, Jack. “Music, Language, and Technology in More-Than-Human Sonic Cultures.” Journal of the Royal Musical Association, vol. 149, no. 2, 2024, pp. 627–37, https://doi.org/10.1017/rma.2024.39.

Hixson M. D. (1998). Ape language research: A review and behavioral perspective. The Analysis of verbal behavior, 15, 17–39. https://doi.org/10.1007/BF03392921

Youngblood, Mason. (2025, February 5). Language-like efficiency in whale communication. Science Advances, 11(6), eads6014. https://doi.org/10.1126/sciadv.ads6014