By employing Information theory to the study of human language, Richard Futrell (UC Irvine) shows that the framework can be broadened from its usual applications such as analyzing the communication systems found within digital networks. Continue reading to learn more about Professor Futrell’s “Information-theoretic approaches to linguistics” course and his hobbies, which include conlanging here:
1. Can you please tell us about your linguistic background?
I’ve been doing scientific research on the nature of human language starting as an undergraduate at Stanford in 2008. I got my PhD in Cognitive Science from MIT in 2017 and now I’m a faculty member in the UC Irvine Language Science department—a growing and exciting program dedicated to quantitative and experimental approaches to language.
2. When did you first join the LSA?
2007, for the Linguistic Institute at Stanford.
3. Can you tell us about the course you are teaching at the Institute?
The course is “Information-theoretic approaches to linguistics.” Information theory is a powerful mathematical framework for analyzing communication systems, which is usually applied to things like encryption, data compression, and building robust digital networks like the internet. It is the origin of the word “bit” for a unit of information.
My goal is to apply the same framework to human language. Information theory lets us answer questions like: What is the amount of information conveyed per word? How redundant is human language? Is human language an efficient code for communication? Are all languages equally complex? Can we derive the universals of language in terms of efficiency? It also gives us models for human language processing that have deep connections to theoretical neuroscience and statistical complexity theory. Plus, the underlying concepts are simple and intuitive, so anyone can start asking and answering these questions with a little bit of background.
4. What research are you currently working on?
All kinds of stuff. Here’s one project I’m particularly excited about. You can use neural networks to derive optimal grammars in terms of maximizing information transfer and minimizing communication difficulty per bit of information. Turns out the optimized grammars show most of the same word order patterns as natural languages.
5. What is your favorite hobby or pastime?
I’m a conlanger—I make up languages for fun. I also make electronic music.
6. In a parallel universe in which you are not an academic/linguist, what would you be?
Bored! Or more seriously, a data scientist.
7. What are you most looking forward to about Davis?
Last time I went to an LSA Institute I was a student (2011 in Boulder). I’m excited to see things from the other side now. Plus, UC Davis is beautiful and has a great linguistics department. I’m looking forward to hanging out with everyone there.
8. Ice cream or Cake? Cats or Dogs? Quarter system or Semester system?
Ice cream, cats, quarters.
9. What advice would you give to graduate students interested in pursuing a career in linguistics?
Don’t forget there is an academic world outside linguistics. I truly believe the answers we seek about human language will require ideas and techniques that are currently housed in other fields. So don’t just get good at linguistics, also go find the buried treasure in what someone else is doing!
View Richard’s UCI webpage here: http://socsci.uci.edu/~rfutrell/