Linguistics is a world where sounds, communication, and culture intersect, and Stefon Flego, who joined the department in August as a postdoctoral research fellow, understands its terrain. From playing historical instruments to decoding the evolution of speech sounds, he shares his journey during this interview. Learn about his compelling research, his exploration of language evolution, and how he envisions the future of linguistics in the age of artificial intelligence.

How did your passion for linguistics begin, and what drove you to pursue it at the postdoctoral level?

I work in the areas of phonetics and phonology, the subdisciplines of linguistics that are concerned with speech sounds and acoustics. I’ve always been fascinated by sound. I majored in historical performance as an undergrad at the Indiana University Jacobs School of Music, where I played renaissance and baroque string and keyboard instruments. At the time I was really interested in the evolution of tonality and tuning systems in early Western music. I happened to take a course on the history of Germanic languages for my minor and found that studying how speech sounds evolved was just as intriguing. I ended up pursuing linguistics in graduate school rather than music, which is for the best, because I think I make a better researcher than performer.

In layman’s terms, can you describe your current research focus and why it’s important or intriguing to the broader study of language?

Linguistics really came into its own as a modern academic discipline in the 19thcentury when philologists were figuring out how speech sounds across Indo-European languages related to one another historically. Today, linguists like me are still trying to understand basic questions about sound change. In my research, I study factors that could explain why languages’ sound systems change in the way that they do.

For example, words like ‘throw,’ ‘threw,’ and ‘through’ have sounded distinct from one another for most of the history of the English language. But for speakers of many modern English varieties, ‘threw’ and ‘through’ are homophones. It turns out this particular change isn’t really surprising when we consider the role this pair of vowel sounds played in facilitating speech comprehension for the listener. On the other hand, these same communicative factors predict that words like ‘through’ and ‘throw,’ despite variation in their pronunciation across modern dialects, are unlikely to become homophonous for any speaker anytime soon.

To investigate questions relating to language change, I draw on data from a diverse set of languages, including other Germanic languages like Icelandic, Southeast Asian languages like Vietnamese and Hakha Lai (spoken in Myanmar), and West Nilotic languages like Dinka and Nuer (spoken in South Sudan and Ethiopia).

How do you see the future of linguistics evolving, especially with advancements in technology such as artificial intelligence and its intersection with language?

Although linguists have long been describing how languages change over time, understanding why they change has generally been considered a near-impossible question to answer. But thanks to advances in digital information storage and computing power, this is becoming a tractable research area. I’m able to work with large language corpora and use computational modeling techniques to address my research questions, so I think it’s a really exciting time to be in the field! 

I haven’t given a lot of thought to how AI will impact my work. I suppose I’m hoping it will know the research landscape in my field better than me soon and can write up lit review sections for me!

Written by Leslie King