This dissertation uncovers an alternative theory of the human voice that decenters speech in premodern Chinese literature from the third to the seventeenth centuries. Through an examination of records of reportage, anecdotes, tales, biographies, poetry, and commentary, all centering on three types of unusual voices—whistling (xiao 嘯), a kind of sonic storytelling called kouji 口技 (vocal virtuosity), and bird speech (qinyan 禽言), I show that these voices create surprising connections among the human vocal apparatus, the body, and language and challenge normative accounts of the human voice from the past as well as in the present. Each of these voices resembles a type of speech mediated by a vocal apparatus—for example, poetry (which was often orally composed), regular storytelling, or human conversation—while destabilizing an aspect of the signifying process essential to speech. Whistling animates a poetic voice without the form of poetry or words, kouji puts speech and mere sound on an equal footing, and articulate speech produced by chirping, calling, and talking birds demonstrates a misalignment between the interiority and the exteriority of a speaking voice. I use literary texts from the seventeenth century as a focal point for trans-historical resonances between authors and ideas from different periods in the Chinese history as well as between premodern Chinese theories of the voice and their modern counterparts shaped by Euro-American contexts. Taken as a whole, these unlikely dialogues invite us to reconsider what it means to have a human voice at all.