In this bonus with Dr. Charles Kemp, we look at why the Chinese writing system has actually become visually more complex over thousands of years. Dr. Kemp explains how his team used "perimetric complexity" to measure characters and why the need for distinctiveness in a growing vocabulary overrides the natural drive for simplicity.
In this bonus with Dr. Charles Kemp, we look at why the Chinese writing system has actually become visually more complex over thousands of years. Dr. Kemp explains how his team used "perimetric complexity" to measure characters and why the need for distinctiveness in a growing vocabulary overrides the natural drive for simplicity.
đź”— Share your familect recording in Speakpipe or by leaving a voicemail at 833-214-GIRL (833-214-4475)
🔗 Watch my LinkedIn Learning writing courses.
🔗 Subscribe to the newsletter.
đź”—Â Take our advertising survey.Â
🔗 Get the edited transcript.
đź”—Â Get Grammar Girl books.Â
| HOST: Mignon Fogarty
| Grammar Girl is part of the Quick and Dirty Tips podcast network.
| Theme music by Catherine Rannus.
| Grammar Girl Social Media: YouTube. TikTok. Facebook. Threads. Instagram. LinkedIn. Mastodon. Bluesky.
[Computer-generated transcript]
Mignon Fogarty:  This ad-free podcast is part of your Grammarpalooza subscription.
Greetings, Grammarpaloozians! Thank you so much for being here. Today, I am with Professor Charles Kemp from the University of Melbourne. In the main episode, we talked about fascinating research regarding words for snow. If you haven’t listened to that already, check it out. But now, we’re going to discuss some of his earlier research on interesting scripts. Dr. Kemp, welcome back to the Grammar Girl podcast.Â
Dr. Charles Kemp: Hello. I'm glad to be here.
Mignon Fogarty: Thank you. So, I was reading an article about your research, and I just thought it was so fascinating. We talked to Danny Bate a while ago about alphabets—how the English alphabet formed. And we talked about how it went from these complicated hieroglyphics to our relatively simple letters that we have today in the alphabet. And I thought that makes a lot of sense, that things would get simpler over time. That's just kind of how things work. And then your research found, you know, some languages where that doesn't happen at all. So, tell us more about that.
Dr. Charles Kemp: Well, we were studying the Chinese script and how it’s evolved over thousands of years. And we went into this thinking that it would be a project about simplification. So we were going to study, for example, how it is that the frequency of a character or a word affects the rate at which it simplifies.
But what we found is that if you compare Oracle Bone characters, which are the earliest existing forms of characters, to the modern characters that are used today, the modern characters are actually visually more complex. And so we found evidence not for simplification, but for complexification over those thousands of years.
Mignon Fogarty: What a surprise! And was this all just by, you know, visually inspecting old and new symbols, or were there ways you quantified it?
Dr. Charles Kemp: We did quantify it. There's a measure called perimetric complexity, which was already in the literature as a way to measure the visual complexity of a symbol or a character. And so I think that's probably the most standard way to measure visual complexity, and that's what we used.
Mignon Fogarty: How does that work?
Dr. Charles Kemp: It's based on computing the—it's kind of like you draw a box around the character and you measure the length of the entire perimeter, and then you measure the length of the components inside the character. So if you imagine something like a circle—just a big black circle—that would have about minimal perimetric complexity. But if you imagine a circle and now you've taken out a lot of bits in the middle of the circle so there are all these contours inside the outer perimeter, that's going to have high visual complexity.
Mignon Fogarty: Hmm. So interesting. And so why do you think that Chinese has gotten more complex?
Dr. Charles Kemp: Well, I think there are two big things to say about this. So first, the number of characters has increased over time. And when that happens, as more characters enter the written language, it's important for the characters to be distinctive—to be distinguished from one another. And so that creates a pressure which makes things more complex because there are only a relatively small number of very simple forms that can be used. So I think that's one factor.
The other factor is Oracle Bone characters—these are characters that have survived because they were etched on bones and used in rites having to do with divination and so on. Those are not the first-ever Chinese characters. Those are just the first characters that we have access to now. And at that time, Oracle Bone characters—the language was mature; it had been used for hundreds of years already. So I wouldn't be surprised if there was this initial process of a simplification and it's just that we've missed it. It isn't in the historical record because those very earliest characters haven't survived.
Mignon Fogarty: I bet when you're carving things on bones, too, there's more incentive to keep it simple than if you're using ink.
Dr. Charles Kemp: Yes, that's absolutely right. So in retrospect, it makes sense that those Oracle Bone characters would have been simpler than the later ones. But, you know, some of the later ones actually are done in clay as well, and so there, it's a lot easier to inscribe something in clay than to etch it on a hard bone.
Mignon Fogarty: Yeah. I think it was Keith Houston who—I was talking to him about the printing press, and, you know, it was invented in China, but because of the enormous number of symbols, it didn't really take off. Because you had to have so many—I forget what they're called, but the squares that, you know, you put into the printing press. And when Chinese has so many of them, the printing press didn't necessarily make it that much faster. It was only when it went into the Latin alphabet that, you know, you had the limited number of—I think they're called punches—that that's when it really caught on.
So there are different languages, and they have these different things that you don't ever think about. And when I was—I think there was an illustration that went with your article, and it had one of these incredibly complex Chinese characters, and I was just squinting and looking at it because I could hardly distinguish all the fine lines inside and where they were going. And I was wondering how people even read it. And I thought, I feel like this would need to be simplified, for me at least, without a magnifying glass to read it. I mean, is this an issue in Chinese?
Dr. Charles Kemp: Well, I think it is an issue, and it's been discussed. So one thing that comes to mind, there's some evidence that rates of myopia are higher in China and certain East Asian countries, and it's thought that, you know, young children spending a lot of time staring at these characters early on might contribute to it. There's also the question of learnability. Like, having to learn and remember these many characters is very, very challenging.
And so there have been discussions through time about whether or not the Chinese script—whether the Chinese government would decide to replace characters with an alternative system like Pinyin. So I think that's no longer a viable option, but it was very seriously discussed about a century ago, I believe.
Mignon Fogarty: Hmm. Are there any other sort of downstream findings from your research that the language is becoming more complicated?
Dr. Charles Kemp: I suppose it's related to the more general theoretical idea about efficiency. And because the idea that characters should simplify and become easier to use is kind of consistent with the idea that languages want to be efficient and language users want language to be efficient.
So I guess one thing it suggests is that we were only looking at a measure of visual complexity, and that's not the only important aspect of complexity. So another aspect of complexity that applies at the level of an entire system is: Are the elements—are the characters in the system—made up of reusable parts?
And by that measure, it's clear that Chinese has become simpler over time. Now characters are built up from a standard repertoire of strokes, and before that, just wasn't the case. So some other researchers, I think after our work, did a similar analysis using a measure of complexity at the system level and found that, as expected, Chinese has simplified a large amount on that dimension. So it's a kind of—one intriguing question for me is thinking about all of the different aspects of complexity and how they work together to shape the historical evolution of a writing system.
Mignon Fogarty: Wow. That's fascinating. That's really interesting research. Let's finish up by getting your book recommendations—one of my favorite parts of the show. Your three books that you would recommend to our listeners. I'm looking for books that, you know, maybe you read 20 years ago and they stuck with you, or a recent favorite—something that would make a great gift is always nice.
Dr. Charles Kemp: Yeah, so the first one I did read about 30 years ago when I was an undergraduate. It is "Hackers: Heroes of the Computer Revolution" by Steven Levy. And this was important for me because at the time I was studying mathematics, and this is the book that convinced me that I wanted to learn computer science and learn about computers.
Partly because the book talked about all the amazing things that people were doing with computers. It talks about the early hacking culture at MIT and the early days of the personal computer revolution—Apple Computer and things like that. So not only the technology was really appealing to me, but the culture that it talked about—a culture where people are judged not on the basis of credentials or academic hierarchy, but what they're able to do, and a culture where there's this philosophy that information should be free and open to all. And I just wanted to be part of a community like that, which is why I got into computer science.
Mignon Fogarty: That sounds wonderful. So actually, you were a computer scientist before you got interested in language?
Dr. Charles Kemp: That's right. I guess I'm among a group of people who are trying to apply computer science and mathematics to understand aspects about language. So my work draws on both of those things.
Mignon Fogarty: So how long ago did you start working on language then?
Dr. Charles Kemp: I did my PhD in psychology and was working on categorization at the time—so how people organize the world into categories. And a lot of time psychologists, including me at the time, study this using very simple toy artificial stimuli like little line drawings of insects or things like that.
And there's good reasons why psychologists do that—that allows them to control all the things they talk about. But it always seemed artificial to me, and so I moved into language because language allows you to study real-world categories—the real labels that people have and use in the world. And so I've been working on that for more than 15 years now and still find it fascinating.
Mignon Fogarty: Oh, that's great. I love combining disciplines like that. It always brings new perspectives. Okay, so what is your second book?
Dr. Charles Kemp: My second recommendation is actually related to the psychology background that I was talking about. It is "The Undoing Project" by Michael Lewis, which talks about the partnership, friendship, collaboration between Amos Tversky and Danny Kahneman.
So Kahneman, of course, is now winner of the Nobel Prize in Economics and maybe the more famous of the two, but Tversky was the person that I really admired. He was one of the best people in the business of trying to use mathematics to describe how people think and capture the laws of thought. And the book is just a really engaging discussion of some of his work and the collaboration that he had with Kahneman.
Mignon Fogarty: Was it like a lifelong collaboration?
Dr. Charles Kemp: Well, one of the interesting things about the book is it talks about how the collaboration fell apart at the end. So they did this amazing groundbreaking work over a number of years, but then went their separate ways and no longer published with each other. And it talks a bit about, kind of, the breakdown in their relationship and why that happened. So on a human level, it's interesting as well as the scientific parts of the story.
Mignon Fogarty: Fascinating. And what is your final book?
Dr. Charles Kemp: My final book—I know these weren't necessarily meant to be related to my work, but the final book is. It's a book called "Words of Wonder: What Endangered Languages Have to Tell Us" by Nick Evans, who's a linguist.
And it talks about vocabulary across a number of different endangered languages and is just full of interesting examples about different words in different languages and what that tells us about culture and how those words kind of capture rich knowledge that's been developed over many years.
And so I—a lot of times I think linguists talk more about grammar and syntax and vocabulary, and, you know, I realize we're on the Grammar Girl Podcast now, but for me, vocabulary has always been the part of language that's most fascinating and intriguing, and this book is about the best introduction to vocabulary across languages that I know of.
Mignon Fogarty: That sounds wonderful. Professor Charles Kemp from the School of Psychological Sciences at the University of Melbourne, where can people find you?
Dr. Charles Kemp: I can be found at charleskemp.com, which has links to my papers and some of the interactive apps we've developed to explore things like the simplification of Chinese characters over time.
Mignon Fogarty: Oh, wonderful. Well, thank you so much for being here.
Dr. Charles Kemp: It's a pleasure.