Rok Sim (심록)
University of South Carolina
Click to view photo
Rok Sim wearing a gat, a traditional Korean hat
The photo above shows me wearing a gat (갓), a traditional Korean hat. Long before I became a linguist, I dreamed of becoming a daegeum (대금) player—a performer of the Korean large bamboo transverse flute. That early connection to Korean language, sound, and cultural tradition continues to shape how I think about language today.

Hello, I’m Rok Sim (심록), a linguist specializing in syntax and semantics.

My research investigates microvariation in English and Korean, with particular attention to how small differences in grammatical form reflect broader patterns at the syntax–semantics interface. I use corpus-based and experimental methods to study these phenomena, combining evidence from naturally occurring language data with speakers’ judgments and interpretations.

My recent publications examine variation in English passive-like need-constructions, including need washing (English Language & Linguistics) and need washed (American Speech). These studies examine how grammatical variants that appear similar on the surface are nevertheless constrained by lexical properties, constructional patterns, and subtle differences in meaning.

I apply this same focus on structure and meaning to historical syntax and semantics in my dissertation, The Historical Development of Indefinites in Korean: A Corpus-Based Analysis (PDF). The project examines the historical development of Korean indefinite expressions, focusing especially on the amwu and nwukwu paradigms. In Modern Korean, these paradigms overlap in particle-marked forms such as amwu-to / nwukwu-to, amwu-lato / nwukwu-lato, and amwu-na / nwukwu-na. These forms are roughly comparable to different uses of English anyone, including negative-polarity uses such as I didn’t see anyone and free-choice uses such as Anyone can participate.

The puzzle is that these Korean forms are neither fully interchangeable nor clearly separate. Although amwu- and nwukwu-forms can occur with the same particles, they show subtle differences in meaning, distribution, and grammatical behavior. To untangle this system, I use corpus data from 1700 to 1990 to trace how the two paradigms developed over time. I find that they began to coexist more systematically from the 1880s onward and underwent a major shift between the 1880s and the 1930s in how their functions were divided across forms with the same particles. This project shows how grammatical change emerges through the interaction of syntax, semantics, and sociohistorical context.

Building on my foundation in corpus-based and experimental methods, I am currently extending my research to large language models (LLMs). By comparing LLM performance with corpus patterns and experimental results from human speakers, I investigate how well these models capture subtle differences in grammatical form, meaning, and acceptability.

You can find my CV, Publications, In Progress, Presentations, and Teaching through the menu above. You can also visit About Me to learn more about my background and interests.