Episode 140: Brian Hsu’s Collaborations on Cherokee Word Order, Star Trek, and the Faculty Fellowship
February 6, 2025 | Ruby Wang
Brian Hsu (FFP ’24) joins us this episode to share how he fell in love with language and linguistics from a young age. Additionally, Hsu describes his recent projects about uncovering Cherokee word order and “Starship Generative Enterprise,” a Star Trek-spin on modern linguistics. We also discuss Hsu’s time as a Faculty Fellow in Spring 2024, where he worked on “Principled Probability in Language.”
Listen or subscribe on Apple Podcasts.
Transcript
Kristen Chavez: Welcome to the Institute, a podcast on lives and works by fellows and friends of the Institute for the Arts and Humanities at the University of North Carolina at Chapel Hill. I’m your host, Kristen Chavez.
Brian Hsu is an assistant professor of linguistics at UNC-Chapel Hill and is a specialist in formal linguistics. His recent research includes the syntax of clauses and nominal phrases and exceptional sound patterns in language. In spring 2023, he was a Burress fellow in the IAH’s Faculty Fellowship Program, focusing on his project titled “Principled Probability in Language”. In 2019 he received the Schwab Academic Excellence Award from the Institute. He then went on to receive other awards from the university, including the Junior Faculty Development Award in 2022 and the William C. Friday Class of 1986 Award for Excellence in Teaching in 2023.
Brian, welcome to the podcast.
Brian Hsu: Thanks so much for having me.
KC: Great. Thank you. Well, let’s jump in. For those in our audience who may not be as familiar with linguistics, can you give a brief overview of what that field is like?
BH: Sure. Well, it’s a fairly broad field. I would sum it up as the academic study of language. So generally, linguists ask questions like, “What are languages like?” And, “why are they the way that they are?” And this can be answered in a lot of ways. So, some of it is looking at, you know, the actual languages of the world, the 7,000 or so. What you know, why? Why are there so many? How do they come to be why do they change over time?
Other linguists might ask questions like, how does language interact with society? So how do certain speech forms come to be associated with particular groups of people depending on say, things like ethnicity, race, class and how do people strategically use language to express aspects of their identity? So that’s kind of what like, so, the fields of sociolinguistics do.
So, I work in what we might call formal linguistics. So, it’s really more about looking at the kinds of structures found in language. So, for example, the types of word order patterns that you see or don’t see across languages, the kinds of sound patterns that you see or don’t see. You know, my work, for example, aims to come up with models for how language is represented in the mind. So, like you abstractly, you can think of the kinds of word order patterns or sound patterns that speakers create as being guided by, like templates that exist in the mind. So, part of my research is trying to find out what, what those might look like.
KC: Great. Can you share a little bit about what sparked your interest in linguistics and how you got into this line of work?
BH: Sure, I think starting from, you know, even when I was a child, I think I was just always very interested in language, you know, like when I was, like, a very young child, I was just like, fascinated with, like, foreign accents, basically. And I just thought they were really fascinating. I love trying to do them. When I was in like, middle school, I started learning French, which I then kept learning for quite a while. So, I just always enjoyed learning languages or learning about them.
Interestingly, as a child, I also really liked, I think it was in fifth grade, it got really into this game called Krypto. It’s like, K-R-Y-P-T-O. And it’s actually kind of like a math related game where you’re given like five numbers, like 1, 4, 7, 12, 13, or something. And then you’re given like a target number, which will be like 78 or, you know, 224 and the like goal of the game is to come up with, like, taking the five numbers that you have, like using just arithmetic operations, like adding, subtracting, multiplying, finding some combination of those to get to the target number in like, I think, like something like, as few steps as possible, or as fast as possible. Don’t exactly remember, but I was really into the game. I just thought it was, like, so fun.
And it actually, I eventually realized it’s very similar to what linguists do. It’s ultimately, like, you know, when you see languages of the world, they are complex in many ways. They like, there are lots of different patterns involved, but we’re always trying to find like the sort of like, more abstract core behind them, or try to, we try to propose a small set of like rules or principles that can explain or generate, like all the variety of things that languages do. And so, I think, you know, when I encountered what linguistics was in college, it really sort of paired together those two things. I just really enjoyed thinking. I enjoyed talking about language and learning about language, but I also enjoyed this kind of, like analytical exercise involved. And they just came together in a very nice way for me, I think.
KC: That’s really cool. Thanks for sharing that. You recently published a paper with Benjamin Frey on the word order and its flexibility in the Cherokee language. Can you talk about that research and that collaboration?
BH: Yes. So yeah. So, Dr. Benjamin Frey, so he was, he’s now a professor at UNC Asheville, but for many years he was a professor at UNC in the American Studies Department, and his research was on, a lot of his work is on language revitalization for Cherokee.
So, he is a member of the Eastern Band of Cherokee Indians. The Cherokee language, which is, you know, indigenous to the like southeastern Appalachian range, including Western North Carolina. You know it is endangered. So, at the moment, there’s, you know, there’s fewer than 2000 speakers of the language. Most of them are elderly in the language. You know, it is being passed on, but I think not, not to the extent that people would like. So, there is a certain threat to the language.
So, Ben’s work is about how do we find ways to reverse this, like, how can we, how can we get more speakers of Cherokee? And so, one thing involves just teaching the language, right? Like a lot of people, there is a huge interest in learning Cherokee. And so, in order for that to happen, there need to be, like educational materials. You just need to be, you know, like textbook materials, for example, that like, explain aspects of Cherokee grammar.
You know, so, I was in some early conversations with Ben when I first came to UNC, were about, like, what is needed. It turns out that for some aspects of Cherokee grammar, like, some are, like, explained very well, and there’s plenty of existing materials on them, but others are, you know, let’s say under documented, there’s just not much description, like, about a pattern, including something actually very basic, which is just, if you have a sentence, you know that has, like a verb and one or two noun phrases, how do you order them?
And it turns out that in Cherokee, it’s quite flexible. Like the order of nouns and verbs is, in general pretty flexible. You can find examples of like, nouns before verbs, or verbs for nouns in a sentence, like “John ate,” you can also find like “ate John,” or the equivalent of that in Cherokee. And it turns out that the suspicion we had was that that it’s not random, like there are things that condition which order is preferred, but like at the moment in you know, dramatical descriptions of Cherokee, it was not like what those principles were, were not made particularly clear. So, the origin of my project with Ben was to try to figure out what those what those are.
KC: That’s really interesting. It makes me think a little bit more about how, since I’m not as familiar with linguistics, making me think about, like, how is that helping inform, I guess, ways of language learning in you know, and how it I don’t know, I guess, as someone who’s just a native English speaker and has only dabbled in Spanish, right? But thinking about, oh, you have to, break your own word patterns, of how the adjectives might come together, and stuff like that that I hadn’t thought of in kind of that sense before.
BH: Well, one thing that is interesting is that, you know, on some level, these kinds of patterns where, like word orders, are flexible, but ultimately kind of predictable, they’re not too uncommon. So, there’s a good English example I like to use where, in some cases, in this one pattern of English, there’s some optionality in what, what people prefer, right? So, for example, let’s look at the choice between the roof of the house and the houses roof, right? They seem to mean the same thing, and to most people, they sound kind of like equally good. So, it’s sort of like, there’s, there’s already optionality, but there are certain things, like, if you change certain properties of those nouns, you’ll get a different preference. So, it’s like, if one of the nouns, like, let’s say the possessor noun is a human, but the other object is not, then people really prefer one order than the other. So, for example, John’s house sounds much better than the house of John’s right?
So actually, in the Cherokee pattern, we see something very similar. So, the main finding of the paper that was just published was that, you know, we can pinpoint the factors that predict whether Cherokee speakers will overall prefer to put a noun before a verb, or to put or to put the noun after it.
One of them is basically how the noun is participating in an event. So, if you have something like the dog, like the dogs jumped, the dog ate, the dog barked. These are all events where the dog is what linguists will call an agent, like an active participant. It’s actively doing something.
And this is very different from structures like the vase broke or the vase fell. Those are, you know, in those examples, the subject is what we’ll call a patient. So, it’s just undergoing something. It’s not actively doing anything. And so, in Cherokee, it turns out that this affects what the like, how you prefer to order things. So, if it’s if the subject is an agent, like something like the dog barked, that is actually the more common order in Cherokee. But if it’s something like the vase fell, you’re more likely to find a word order like fell the vase.
KC: Interesting.
BH: That’s one of the patterns we see. There’s an interesting interaction this has with like another factor that predicts the order. So, one thing about Cherokee is that in this language, if something is brand new, so being mentioned for the first time in a conversation, there’s a preference for that to come before the verb. As opposed to something that refers to an entity that’s already known or that we’ve been talking about.
So, for example, in English, we usually signal new information with like ‘a’ or ‘an’ like coming before a noun. And if it’s like a noun that we’ve been talking about, usually like ‘the’ will be used. And so it turns out that in Cherokee, it’s like new information, things tend to come before verbs, so like “a dog jumped.” But if you were to say something like the dog right where the dog is older in the story, you’re a bit likelier to find “jumped the dog.” And this is actually the opposite of English, where English we actually kind of prefer to have new information near the end of the sentence and older stuff at the beginning.
So, this is a difference between languages, but nonetheless, sort of seeing that newness of a noun and its thematic role. So that’s like, whether it’s an agent or a patient, those things both interact in Cherokee to sort of push speakers in particular directions in terms of, like, how they order nouns and verbs.
So that was basically the finding and yeah, and hopefully the idea is that this can go into educational materials like now, instead of just having to tell speakers, people who are trying to learn Cherokee, that the order of nouns is flexible, now you can tell them, well, it is flexible, but there are some orders that are more common than others, and we can tell you when like that might happen.
KC: Okay, that’s really interesting, and this might not be relevant to your line of work, but I am curious if you kind of see any shifts over time with these types of speech patterns or word order. I’m thinking specifically with something like Cherokee and the revitalization, as you mentioned, there’s fewer speakers. I wonder if that has changed over time, or is that anything that you’re able to view or track? Or if it’s maybe not as relevant.
BH: Well certainly all languages do change over time, and even if they weren’t in contact with others. And of course, Cherokee is in situation where almost every speaker of Cherokee also speaks English, and English is like the sort of dominant language in areas where Cherokee is spoken. So there probably is. It would be very reasonable to guess that aspects of Cherokee grammar, for example, perhaps some of these word order preferences, you know, may have been influenced over time through contact with English speakers, or just by the fact that you know, if you have, if you are bilingual or multilingual, like all of those languages are kind of all floating around in the brain, and they can influence each other in real time.
But I think at the moment, in terms of, like, you know, do we know anything about how Cherokee word order principles have changed over time? Not that I know of, but it’s something that we could look into. So, it’s sort of like now that we kind of have a snapshot of what Cherokee clausal word order principles are now, or at least in the recent past. In principle, if we, you know, looked at like older written materials, we might be able to actually find out if things have changed over time. I don’t know what the answer is like. My guess is probably, but I couldn’t tell you specifically.
KC: Yeah. Great. Thanks, I was curious. As I mentioned, you recently had a Faculty Fellowship here at the IAH in spring 2023. What was that experience like?
BH: Well, it was a great experience. For those of you who are not familiar with the Faculty Fellowship: You know, when you have this fellowship, you know, one is released from teaching duties for that semester and weekly, the main responsibility is that all the people who have the fellowship, you know, six to eight people, they meet weekly for like, a few hours in what’s called a seminar. And usually, it involves discussion of one of the fellows’ work. So, like, one person sort of presents their work, or we’ve all read the work of one of the fellows, and then we discuss it. And yeah, it’s very nice.
So, there’s, you know, in a typical year, or at least in mine as well, you know, you have faculty from all sorts of departments that work on all sorts of things that are, you know, not related to what I do. You know, I work on language. But there’s also people who work on, you know, farming practices in the French Caribbean, or access to medical care in Uganda. Those are actual projects. You know, they’re not things that I necessarily encounter in my day to day life, but it’s just very, you know, interesting to learn about. It’s just nice to see what my colleagues do, and to sort of talk through some of the professional challenges that we might, we might share, even though we’re in like, different fields. So it’s just a very nice way of building community among the faculty at UNC, and yeah, and I definitely feel I felt refreshed coming back into the classroom afterwards.
KC: So as part of your fellowship, you explored the project “Principled Probability in Language,” about these different modes of thought within linguistics, in regards to the variability with word order, as you were kind of talking about before. Can you walk us through some of those concepts and the work that you did?
BH: Yeah, absolutely. So, this, so this is actually a project that kind of grew out of the work on Cherokee, right, so it’s sort of like… Cherokee was a great example of a language where you have a case. You have a pattern that’s probabilistic, so there’s kind of… Speakers have choices in what they can do, but there is, ultimately, there are principles behind it. There are like what there are factors that will predict what speakers might prefer to do. And so, this project that I was been working on, the fellowship project, was thinking through how should we, what are the implications of this for linguistic theory?
And so broadly speaking, there’s sort of two big camps among like, linguists about how, like, how relevant these patterns are, what they tell us. Again, you know, not everyone necessarily falls cleanly into one of these extremes. But on one end, you do have people who think that you know, for linguists, who want to think about what grammars look like in the minds of speakers. On one view, grammars in the mind should only represent rules that apply 100% of the time. So, on this view, if you do have a probabilistic or variable pattern like the one in Cherokee, then all the linguist needs to say is, well, the order of nouns and verbs is flexible, and anything that predicts what that order is something that I don’t need to worry about. Or whatever it is, it is not related to language. Right?
On the other extreme, you have views where you know languages show probabilistic effects. But what this, what they claim that this suggests, is that speakers, you know when they represent language, they know the specific like pieces they might use, like particular words or particular chunks of words, but they don’t represent rules about how these things are used. The speaker is just sensitive to how frequent these items are and how frequently they go together with other words. So in this side on this view, grammars don’t really contain like, templates or rules about what structures are used. It’s really just they arise from these kind of, like probability distributions over items and just speakers memorize, like, what kinds of combinations of things are more frequent than others. So, these are apparently very different views, but the sort of goal was, to my project was to push for a theory and, like, formalize the theory that integrates like aspects of both, right?
So, the ideas, so my view, for example, is that languages do have like, rules, like, so speakers are sensitive to they, perhaps do they do directly represent things like, is this a noun? Is this a verb? Is it old information? Is it new information? So, they, speakers, are aware of these things when they’re producing language. The probabilistic aspect of it comes up, so you know, in my proposal, in the idea is that when speakers are forming a sentence, they have they apply what linguists like to call constraints.
So, they’re sort of like principles that one wants to follow. So, these principles might be things like, if the noun is new information, put it early in the sentence. Or if the noun is referring to an agent, so someone actively doing participating in an event, you want to put that at the beginning. But sometimes these constraints may may conflict with each other. So, if you have something that is an agent who is actively doing something, but also old information. So something that would normally come after the verb, there’s no way to satisfy all of the constraints.
So, the idea is that perhaps in some language patterns, what happens is speakers assign sort of like weights to the constraints. So, they’re kind of like degrees of importance, and these ultimately get mapped to just like probabilities. So, things that are relatively more important, you’re going to satisfy those more often than things that are less important, but at the end of the day, that still leaves room for, like, optional patterns where, you know speakers, there is no one way to say something kind of like the house’s roof versus the roof of the house example.
KC: That’s really interesting. Thank you. Thanks for walking me through that. That was helpful. So in preparing for our interview that we’re doing right now, one thing that I learned about you is that you and fellow linguistics faculty member Caitlin Smith created “Starship Generative Enterprise,” which blends both linguistics and Star Trek. Can you talk about the inspiration behind that and how it came to be?
BH: Sure! So yeah. So, this is, well, during the pandemic, when I didn’t have, you know, I had a lot of free time, me and Caitlin or Dr Smith, watched a lot of Star Trek: The Next Generation, which I had never seen before. If you’ve watched Star Trek, I think of any series or movies, there’s what people call the techno babble dialogs where characters are talking about things happening on the starship that are not, you know, they’re not real scientific terms, but they sound pseudo-scientific. So, things like, you know, sort of like the secondary warp coil injector to the dilithium warp core has ruptured. I may need to depolarize. You know, things like that.
KC: Yeah, yeah!
BH: And so, what I thought would be funny was to create, and if you’ve watched enough that you find out they kind of have this little format to them, which you can represent in like three panel meme forms. So that is kind of what this, this work does, or that’s what Starship Generative Enterprise is. They’re like little three panel memes that have dialogs like these from Star Trek, except they’re all using terms from linguistics. So, all of them involve, like, actual terms taken from actual linguistics papers. I should say that they’re not being used as gibberish in these and like all of them, they’re actually describing, like proposals or ways to explain patterns. But, you know, I picked them because to outsiders, presumably some of these terms will look, you know, about as foreign as the as what you actually hear on Star Trek. So that was, that was, I thought that was funny. So that’s what it is.
KC: Yeah, I enjoyed scrolling through it. It’s funny! I, you know, to be frank, I did not understand a lot of it, but I also wouldn’t understand a lot in Star Trek itself, so with their techno babble. But yeah, thank you. I just thought it was fun.
BH: Glad you thought so.
KC: Yeah! And so, for our last question, we’ll end with something that we ask all of our guests, can you share a book or creative piece of some sort that has changed your life or has made an impact on you?
BH: Yes, yeah, so I would probably, so I will pick a book and it actually has nothing to do with linguistics. It’s The Death and Life of Great American Cities by Jane Jacobs, which I read when I was in late high school, you know, if you’re if, for those who are unfamiliar with the book, Jane Jacobs was, you know, she was not a professional urban planner, but she basically had gathered a lot of observations about things that she thought made neighborhoods or cities to be successful or thriving versus unsuccessful, and she’d like written about these in a book, this book in particular, and a lot of the sort of like ideas, ultimately, like in the fields of urban planning, were, like, quite revolutionary.
But at least for me, while I was reading it, what I got about it, it’s something that taught me to sort of appreciate and experience cities in a totally new way. So it’s something like having read the book now, you know, you know, if I was walking down the street, I might think to myself, Oh yes, I can see how having mix of buildings of different ages and sizes is conducive to a diversity of businesses or things like, I might notice things like, oh, This neighborhood has really long blocks, we’re making it relatively not as walkable, not as many businesses here, whereas an area that has short blocks might be like more vibrant and have like more like more business activity. So, these are sort of things about cities that, you know, I never noticed before, but like, having read the book, was like, now you kind of see them. They become obvious. You sort of see the world in a new way. And that was really cool.
Also, I think it made me really excited to explore a lot of like the actual neighborhoods that get discussed in the book, like Boston’s North End, or Greenwich Village in New York, or Rittenhouse Square in Philadelphia, which are all still, like, you know, many decades later, still, like, very nice neighborhoods. And, yeah, like, when I was in college, I was like, you know, had to visit. I went to these places, right? Specifically, having read the book. And, yeah, I really enjoyed it.
KC: Awesome. Thank you so much for sharing that. And thanks for taking some time out of your day to be to join us on the podcast.
BH: Yeah. Thank you very much. Thanks so much for having me.
KC: This has been The Institute, a podcast by UNC’s Institute for the Arts and Humanities. This episode was hosted by Kristen Chavez and edited by Ruby Wang. Listen to other and upcoming episodes of The Institute by subscribing on Apple Podcasts, Spotify or wherever you listen to your favorite podcasts. Visit our website, iah.unc.edu, to find past episodes and transcripts. There, you can also learn more about our programs and opportunities for UNC-Chapel Hill faculty, find upcoming events, and read stories that feature our arts and humanities fellows. Thank you for joining us.
Categories: IAH Podcast