Linguists have long examined language in musical contexts, just as ethnomusicologists have considered the language(s) of the musics they study. Though an increasing number of scholars are now combining (ethno)musicological and linguistic approaches in their work, this is still far from the norm in linguistics, where musical elements are often disregarded in analyses of language in musical context. This dissertation aims to challenge this status quo, by introducing and demonstrating new methodologies for integrating musical data into linguistic analyses in three subfields: structural linguistics, sociocultural linguistics, and language revitalization.

Each new method is illustrated through a representative study. In the first, I introduce a novel method for integrating musical transcription into language documentation using ABC notation in ELAN, which reveals a striking correspondence between lexical tone and musical melody in Tlahuapa Tù'un Sàví, a Mixtec language spoken in Guerrero, Mexico. In the second, I present a multimodal discourse analytic method for sociocultural linguistic research, and use it to show how three different Welsh rock artists enact diverse identities through co-temporal code-switching and musical style-shifting. In the third, I demonstrate how linguists can combine musical and linguistic data to create an UTAUloid — a combination speech and music synthesizer for collaborative vocal songwriting — to aid in musical language revitalization efforts, through an example in Cherokee.

Together, the results of these studies illustrate the rich potential of music in linguistic research across subfields, and show that the combination of musical and linguistic data yields unique analyses not possible by examining language alone. More than optional accompaniment, music is an essential component of a discourse functional approach to language in musical contexts, and the methodologies introduced in this dissertation aim to make including musical data as accessible as possible.

The effects of language contact extend well beyond the borrowing of lexical items, and can include morphosyntactic, phonetic, and phonological changes over time (Thomason & Kaufman 1998). One especially common structural outcome of long-term contact is phonetic transfer (Matras 2009:222). The Welsh spoken in Patagonia, which has been in close contact with Spanish for the past 150 years, offers one potential example of this phenomenon: Jones (1984) observes that younger speakers of Patagonian Welsh may be developing unaspirated voiceless stops /p t k/ as a result of Spanish contact.

This paper measures the voice-onset time (VOT) of the Welsh voiceless stops /p t k/ using contemporary conversational speech data from both Patagonia and Wales, for male and female speakers in three age groups (0-29, 30-59, and 60+), to examine the effects of Spanish contact on Patagonian Welsh. Results indicate that the tendencies seen in Jones (1984) have held true, and in fact have generalized to become a feature of Patagonian Welsh for speakers of all ages: Patagonian speakers produce the Welsh stops /p t k/ with significantly shorter VOT values than speakers from Wales.