AI and Korean studies

There will be a conference at the Korean Studies Institute in Andong later this month examining the future of Korean studies. I’ve always been uncomfortable with “future” studies — how can we know? — but still, it’s a topic worth looking at especially these days when everyone seems to be concerned about an artificial intelligence (AI) future and what it’s going to do to impact us all.

To know the unknowable is the game of looking at the future, but there are some mileposts, some hints we can find. Take the axiom “the past is prologue” and look at some things in the past, and present, that can point to some things in the future.

As far as AI is concerned, my deepest connection with it is in translation — computer-assisted translation. Years ago, a good friend of mine, Kim Han-gon, was a linguist who had been at Seoul National University, but we “scouted him” (using the term the way Koreans use it, not the way it’s used in English) — meaning we recruited him to come teach at Brigham Young University. He was looking at computer translation. He was discouraged thinking it would never happen because of the complicated nature of speech. But he kept working on the project and made significant progress before he left this planet at the age of only 65. In the years since, computers, let’s say AI, has made great progress and translations are remarkably well-nuanced and accurate today.

I couldn’t live without computer translations. I don’t know enough to say whether it is technically AI or not, but it’s surely related. Episodes on my YouTube channel are half in English, half are in Korean, but all of them have subtitles in both languages. I prepare the subtitles myself — I have an editor who does the computer and video editing, but I do the subtitles — and I could not do it without the computer translations. It needs my eye to make corrections, and I make mistakes because of my limited Korean abilities, so the translations are not perfect. But they are close to perfect.

It is interesting to me that the area where the computer translations fail most often is in regard to Korean studies, Korean history and related areas. It just doesn’t seem to know historical terms. For example, the basic concept of “bongwan” — the ancestral place of origin of a clan — a term known to every Korean on the planet, often comes out as “main house.” (Although I just tested it again, and it came out correctly as "ancestral clan seat"). And of course, there are several online translators available, and some are certainly better than others. But generally speaking, on my YouTube channel that is mainly on Korean history and culture, the corrections I need to make are mostly of Korean historical terms.

Having said that, I just now tried several cases, using specific terms of government organization for the Joseon Dynasty, to give you examples of how the translations are inadequate — but, lo and behold, my translator, now, got them all right. Progress! This is a real indication of how quickly things are changing in the online world.

The point is that translation of texts from one language to another is really good these days. And that says that the language barrier, which Korean translates as the “language wall,” has almost disappeared. This means that scholarship in Korean can immediately become part of scholarship in the West, and scholarship on Korea written in the West can immediately become part of the academic world of Korea. This has NOT been the case before.

My own work is a case in point. In my Ph.D. dissertation, I was able to show that the association between Confucianism and male dominance, patrilineal preference, is not that old — only since the late 17th century. Confucianism for 1,000 years before allowed men and women to be on equal footing in regard to inheritance, property control, marriage location and everything except political power. I showed that if Korea wants to have gender equality today, they do not need to reject Confucian values. Male domination and Confucianism are two very different things.

I had hope my work would show up in high school classrooms and textbooks, but it has not happened yet. To promote my own work is the major reason I write this column and why I produce my YouTube episodes.

With AI making it easier to read scholarship from other languages, perhaps that alone will change the educational system.

Another change I can see in the Korean educational landscape is the use of AI to replace the "hagwon" system. The huge amount of money spent on "cram schools" can all evaporate if parents carefully create their own tutoring systems and keep their children at home to study.

As soon as a small group of students gain admission to "SKY" schools — Seoul National University, Korea University and Yonsei University — on the basis of home tutoring on AI programs, parents will see it’s possible and all that money sent to private crams schools —the hagwon — will be redirected to other efforts.

For that matter, in the AI world of the future, Korean studies will broaden and internationalize, the hagwon system will disappear, and perhaps the whole educational system as we know it will be shattered and revised from the bottom up. Maybe. C’mon, it’s the future. Who knows?

Mark Peterson (frogoutsidethewell@gmail.com) is a professor emeritus of Korean studies at Brigham Young University in Utah. The views expressed here are his own.