Date of Degree


Document Type


Degree Name





Juliette Blevins

Committee Members

Marisa Hoeschele

David C. Lahti

Michael I. Mandel

Subject Categories

Animal Studies | Behavior and Ethology | Comparative and Historical Linguistics | Comparative Psychology | Ornithology | Other Linguistics | Phonetics and Phonology | Typological Linguistics and Linguistic Diversity | Zoology


Bioacoustics, animal communication, evolutionary linguistics, cultural evolution, vocal learning, phonotactics


Learned acoustic communication systems, like birdsong and spoken human language, can be described from two seemingly contradictory perspectives. On one hand, learned acoustic communication systems can be remarkably consistent. Substantive and descriptive generalizations can be made which hold for a majority of populations within a species. On the other hand, learned acoustic communication systems are often highly variable. The degree of variation is often so great that few, if any, substantive generalizations hold for all populations in a species.

Within my dissertation, I explore the interplay of variation and uniformity in three vocal learning species: budgerigars (Melopsittacus undulatus), house finches (Haemorhous mexicanus), and humans (Homo sapiens). Budgerigars are well-known for their versatile mimicry skills, house finch song organization is uniform across populations, and human language has been described as the prime example of variability by some while others see only subtle variations of largely uniform system. For each of these species, I address several questions related to variability and uniformity: What is the typical range of variation? What are the limits of variation? How are those two issues related? And what mechanisms underlie variability and uniformity?

In chapter 3, I investigate a potential domain of uniformity in budgerigar warble: the segment. Segments, units divided by acoustic transitions rather than silence, have been largely ignored in non-human animal communication. I find that budgerigars can achieve a high degree of complexity and variability by combining and arranging these small, more stereotyped units. Furthermore, I find that budgerigar segment organization is not only consistent across independent budgerigar populations but is consistent with patterns found in human language.

In chapter 4, I investigate variability in house finch song. I present data showing that house finches learn sound patterns which are absent in wild house finch populations. These data suggest that cross-population variation in house finch song is narrower than what is permitted by the house finch song learning program.

Finally, in chapter 5, I focus on human language, the most well-described communication system. Here, I research a sound pattern that is absent in the majority of known languages. I find that the rare pattern has independently developed at least six times. In every case, the historical pathway which led to the rare pattern was the same. The historical development in these six linguistic lineages suggests that the overall rarity of the sound pattern is the result of acoustic similarity.

These data illuminate the evolutionary forces that give rise to, and limit, variation. The results of this dissertation have wide-ranging implications, from necessary revisions of linguistic theories, to understanding epigenetic interactions, to the application of evolutionary theory to complex behavior. While these projects within the dissertation are all different, evidence from all three projects support the following claims: (i) cross-population commonality is not evidence for what a species is able to learn; (ii) peripheral mechanisms have a strong influence in limiting cross-population variability; and (iii) high degrees of variation can emerge from uniform traits.