The mystery of the ups and downs

Just about any of you out there who've ever tried to learn Korean will agree that when it comes to tackling this language, we need all the help we can get.

I remember when I first started studying Korean at, of all places, Ghost Ranch, New Mexico. I had taken some linguistics courses in college, so I regarded myself as being at least somewhat enlightened, linquistically speaking. I knew that it takes all kinds of languages to make a world, but I have to admit that I was more than a little taken aback when it became apparent from what our teachers and textbooks told us that Korean was missing some "essential" parts: no "a" or "the," no real plural in the English sense (but they do have a plural you can stick on just about anything you please, including adverbs!), no independent conjunctions — the list could go on.

But the most surprising thing of all was to learn that Korean words had no accent. Our teachers told us that Korean is pronounced evenly, without strongly stressing any particular syllable or raising its pitch. When I objected that the sentences they had us parroting in class certainly didn't sound flat and "accentless" to me, they said it was just the intonation I was hearing.

In a very Confucianlike manner, I accepted what we were taught without further questioning — that is, until I came to Korea and got assigned to a job in Gyeongsangbuk-do (경상북도). Oh, boy! The language I heard all around me there was definitely not flat and accentless. The voice would go up and down, up and down — and not just in any old place at the whim of the speaker. There were definite places in each word or phrase where the pitch could be raised or where it must not be raised.

I soon learned that the dialect of that southeastern part of the country is very conservative in this regard. It preserves features of the pitch accent system of middle Korean which have — according to conventionally accepted linguistic wisdom — disappeared from standard Korean.

A sound spectrogram

This is an example of a sound spectrogram, which linguists can use to study such things as the stress and pitch changes that occur in recorded utterances. Such spectrograms could be used to determine the validity of the frequently heard assertion that standard Korean has no pitch accent.

When I came to live in Seoul after spending more than three years in the Gyeongsang-do region, I was sure I detected many of the same ups and downs in people's speech here, although they were much more subtle and "smoothed out." I mentioned this to a former Peace Corps Korean language instructor, and she told me it was just my overactive imagination being influenced by all that time down in the southeastern part of the country.

I was dissatisfied with that answer, and over the years a niggling sense that the claim that standard Korean had no accent system was just a cop-out remained in the back of my mind and would occasionally pop up in real-life experience. Sometimes, for example, I might use a word (like cheongsan (청산 淸算, meaning 'liquidation") that I had only read in the paper but had never heard spoken, and a friend would say, "There's something wrong with the way you're saying that, but I can't quite put my finger on it." (What was wrong, it turned out, was that I was failing to raise the pitch of the first syllable.)

If you speak some Korean, you can get a feeling for why I kept looking into this "mystery of the ups and downs" by taking a common word you're thoroughly familiar with and trying to pronounce in different ways. How about isseumnida (있습니다)? Keeping the stress even throughout, try raising the pitch of each syllable in turn. Of the four possibilities, only the one in which the second syllable is high sounds right. Even a simple test like that shows that there must be some rules about pitch accent in Korean, whether it turns out that the accent make a difference in the meaning of words or just sounds funny if you misplace it.

After many years of wishing for a dictionary that would show word accents, I was really delighted in 1998 when the Korean Pronouncing Dictionary was published. It touted itself as being the first dictionary to mark all aspects of pronunciation, including accent. Delight quickly turned to disappointment when I discovered that the accents as marked do not conform with the reality of spoken Korean.

Then in 1999, a man named Son Jong-Seop (손종섭) published a book that provided the key to solving the mystery. Entitled Uri mare gojeo-jangdan (우리 말의 고저 장단, 'Pitch Accent and Vowel Length in Korean,' 1999, Jo'ngshin Segesa), this book provides the key to understanding this neglected but so important aspect of modern Korean. Although the book could be better organized, it covers the topic quite thoroughly, giving abundant examples. Of course, in order to read the book you have to already have a fairly good mastery of Korean, ideally including some Chinese characters.

An earlier version of this article first appeared in the Korea Herald on 13 March 2000.