Origin Theories of the Korean Language

Gerardo Martínez
Brigham Young University
Introduction to Comparative Historical Linguistics
LING 450
Professor Cynthia Hallen
February 24, 1998


Korean language is spoken by about 60 million people and is considered to be one of the major languages in the world. The origin of this remarkable language is rather obscure. As I was collecting the information for the present paper, I noticed that most of the sources begin by acknowledging the difficulties the study of Korean represents because of the limited amount of data. Huh Woong begins his article "Development of the Korean Language" saying that "there is no material available to us with which to get a glimpse of the language...as a result, we are unable to clarify to what family of languages the ancient Korean tongue belonged" (1993, p. 1). From the Dictionary of Languages we read from Nam Kil Kim that "For a long time scholars have tried to associate the Korean language to one of the major language families but have not been successful in this venture" (1984 , p. 881). After reading a couple of articles I began to realize that telling the origin of the Korean language wasn’t going to be an easy task, so I understood that there were going to be different theories regarding the origin of Korean. My purpose today is to present those theories which explain from where Korean may have come. I begin by presenting the first attempts to explain the origin of Korean, showing additions by other scholars and then I end by presenting the Altaic Theory which is the most accepted by contemporary linguists.

The Four Hypotheses
1. Koreans Came from the East.

According to Kim Chin-u the East theory holds that Korean people originated from Japan and its neighboring islands. Scholars who support this idea believe that Old Korean is an early version of Japanese . They talk about Old Korean being the dialect of Old Japanese. Kim doesn’t expand further on this hypothesis because "this view is largely colonialistic." As we will see, linguistic evidence shows that the movement of people in those regions in the past was from west to east and not the opposite. The idea that migration was from west to east leads us to take a moment to look what the West theory has to add to the origin of Korean.

2. Koreans Came from the West.

Around the third millennium B.C. people living near the Altaic mountains in Central Asia began to migrate eastward. People moving in that direction got in contact with the Paleoasians (a group of natives settled in the peninsula of what is now Korea). Eventually some other groups migrated to areas such as Japan and Siberia. As Kim Chin-u has said linguistic evidence shows that people actually moved from the west. The evidence shows that the ancestors of Koguryo (the nation that occupied the southern Manchuria and the northern half of the Korean peninsula) brought the Yayoi culture to northern Kyushu(now Japan) some 2,300 years ago (1983, p. 36. See also illustration 1 in the Appendix.)

Lee Kim-moon, professor at the Department of Korean Linguistics in Seoul National University, has gathered lexical correspondences between the Koguryo language and the Old Japanese (see table 1).

Lexical Correspondences between Koguryo and Old Japanese


Old Japanese




















Table 1

Because of this and some archeological findings he says, "it is safe to say that Old Korean was not the Peninsular dialect of Old Japanese. If anything, Old Japanese was the Insular dialect of Korean" (1983, p. 36), meaning that Japanese came as a dialect of Old Korean and not that Korean is a dialect of Japanese.

In addition, Homer B. Hulbert supports differently the West theory. Hulbert’s idea was popular when the Linguistics field was just beginning. According to the British scholar, Korean people and language are related to the Dravidian language of India because both languages share syntactic features such as word order subject-object-verb, postposition instead of preposition, and modifiers in front of the head noun. For example, look at Kui (a Dravidian language spoken in the Orissa province of India) shown in table 2:

Word Oder Comparison


jõr_ dhane gãda ãi manaki, imabaive grãpa m_’e.


river very deep is because any one on it to cross unable.


k’angi nõmu kipkki ttaemune amudo ir_l k_nn_ji mot’ae.


The river very deep that no one can cross it.

Table 2

The syntactic features that Hulbert saw between Korean and Dravidian languages are called typological features in linguistics today. Typological similarities are not enough to establish the genetic relationship among languages. Kim Chin-u explains to us why:

"It is an accepted view that two or more totally unrelated languages may nevertheless share typological similarities. Three quarters of a century ago linguistics was still in its infancy, and one can imagine how striking and suggestive the typological similarities between Korean and Dravidian must have looked to Hulbert, especially when Indo-European languages, about the only well-established language family then, al shared a different typology" (1983, p. 16).

As we can see neither the east nor the west theories can stand as strong explanations of the Korean origin. We now move on to see what the south theory has to offer.

3. Koreans Came from the South.

The other version of the Southern theory holds that Korean is related to the Austronesian languages. Linguistic support for this argument includes open syllables, the honorific system, numerals, and several body parts. On the other hand anthropological evidence includes rice cultivation, tattooing, and the myth of an egg as the birth place of royalty.

Susumu Ohno, a Japanese scholar, supports the south belief by showing that the names of body parts are a core vocabulary and as such there is little room to believe that borrowing could have happened. Ohno gives an example using the word for lung in Korean. Table 3 shows the same word compared to other Austronesian languages:

Body Part’s Core Vocabulary



Maori of New Zealand


Mentawai of Mumatra


Visayan Island of the Philippines


New Guinea


Table 3

The evidence is strong and if it is put that way, one can securely believe that Korean is related to the Austronesian Languages. However, the problem is that the linguistic and the anthropological data doesn’t overrule the northern elements.

4. Koreans Came from the North: The Altaic Theory

The North or Altaic theory says that Korean belongs to a language family called Altaic The Altaic family has three important branches: Turkic, Mongolian, and Tungusic or Tungus. The following data shows the genetic affinity among the Altaic languages:


pA. proto-Altaic
Ma. Manchu
MK. Middle Korean
Ev. Evenki
Mmo. Middle Mongolian
Go. Goldi
J. Japanese
Mgu. Monguor
Yak. Yakut
K. Korean
Chu. Chuvash
Mo. Mongolian
OK. Old Korean
MT. Middle Turkish
OT. Old Turkish
OJ. Old Japanese
Tu. Tungusic

Ma. fon ‘season’, Mo. on ‘year’, MMo. hon ‘year’, Kgu. fän ‘spring’, K. pom ‘spring.’
Ma. fulgije ‘to blow’, Lamut hu-, Mo. ülige-, Mmo. hülie-, K. pul- ‘to blow’, pA. püli-
Ma. firu- ‘to pray’, Ev. hiruge-, Mo. irüge ‘prayer’, Mmo. hirüer-, K. pil- ‘to pray’
Ma. fusu- ‘to sprinkle’, Ulcha pisuri- , Mo. ösür-, Mgu. fudzuru-, K. put- ‘to pour’, Go. pis-
Go. Nalum ‘raw’, Mo. nilay un, K. nal
Ev. alas ‘thing’, Mo. ala ‘Underside of things’, OT. al ‘frontside’, MT. alïn ‘underside’, K. arae, ‘below, under’, pA. al
Mo. cilayun ‘stone’, hu. cul, OT. tas, Yak. tas, K. tol ‘stone’, pA. tïlza
Ma. muke ‘water’, Go. mue, Ev. mu, Mo. mören ‘river’, K. mul ‘water’

In the next section we will go through the development of this theory and then we will be able to see where the Korean language fits.

A. Johann von Strahlenberg and The First Hypothesis, 1730.

The significance of Strahlenberg is simple: he was the first scholar to notice certain similarities between Turkic, Mongolian, and Manchu-Tungus. Strahlenberg named these and other languages "Tatar Languages,"thus setting the ground for further research.

B. Alexander Castrén and the Altaic Family, 1862.

Working with personal suffixes in the same languages Strahlenberg had studied, Alexander Castrén applied linguistic criteria to determine if these languages truly belong to the same family, the Altaic family. Castrén included in the Altaic family Finno-Ugric, Samoyed, Turkic, Mongolian, and Manchu-Tungus. The work by Castrén is valuable because later on scholars recognized the Altaic family as the Ural-Altaic family, which extends from Scandinavia, Hungary , and the Balkans in the west, to the easternmost reaches of the Amur and the island of Sakhalin, and from the Arctic Ocean to central Asia. To summarize, Castrén was the first one to organized the Altaic language family. Currently the linguistic community accepts the Uralic part as scientifically correct. On the other hand, scholars are still debating regarding the Altaic aspect which includes Korean.

C. Gustaf John Ramstedt: The Founder of the Altaic Theory, 1957.

I have briefly talked about the early attempts to build a theory to explain language relationships that have lead us to the Altaic group. It is here that we ought to recognize Gustaf John Ramstedt considered by Yi Sang-Ok as "the most outstanding Altaicist in the history o f Altaic comparative studies and the real founder of the Altaic Theory"(1983, p. 44).

Gustaf J. Ramstedt (1873-1950) was a Finnish linguist and diplomat. Kim Chin-u commenting about Ramstedt’s remarkable work, said that Ramstedt’s " trips took him as far south as Afghanistan and as far east as the Baikal Lake in Siberia." Ramstedt was assigned to be counsel to Japan and China in 1919 at the end of World War I. He lived in Japan for eleven years until 1930. It was during this period that he came in contact with the Korean language through some Korean people who lived in Japan. Ramstedt’s two-volume Einführung in die altaische sprachwissenschaf (1952, p.157) shows his work on comparative phonology and morphology of Turkic, Mongolian, Manchu-Tungus, and Korean.

While Ramstedt’s work only mentions the four languages as a group who share significant elements, he didn’t establish any intermediate stages between a common Altaic language and the four-language group. It was the work of Nicholas Pope and John C. Street that established these intermediate stages therefore giving more sense to the Altaic theory.

Ramstedt’s Scheme

D.Nicholas Pope (1960)) and John C. Street (1962): The Refined Relationship.

As we move into the last part of our analysis, it is important to keep in mind Ramstedt’s Scheme. Pope established a Mongolian-Manchu-Tungus unity because of the similarities among these languages. He also pointed out that Turkic has more in common with Mongolian-Manchu-Tungus than with Korean. This last observation serves as the basis to declare that Proto-Korean emerged when the Mongolian-Manchu-Tungus-Turkic unity still existed (see illustration 2 in the Appendix). Based on Pope’s assumptions we can better understand Street’s hypothesis.

John C. Street’s hypothesis is best titled "Proto-North-Asiatic," for reasons we will see later. Street basically postulated that Korean, Japanese, and Ainu languages branched off not from the Common Altaic unity established by Pope, but from a stage still older, the Proto-North Asiatic language. In "Pope’s and Street’s Schemes" the text in parenthesis shows where Pope’s and Street’s hypotheses differ. What is even more important is that both Pope and Street suggest that Korean acquired certain characteristics before the other languages did --again strongly suggesting that Korean origin is independent ( branching off as a new language) from the other three languages (See illustration 2 in the Appendix for this important detail.)


As we have already seen, the origin of Korean is not clear. Scholars have earnestly tried to build a theory that explains the rather obscure origin of Korean, but as of today none have succeeded. Studies on the matter go from opinions, folk etymology, to more scientific based theories. One thing we can clearly see is that theories claiming that Korean people and language came from the south and the east are overthrown by linguistic and archeological evidence. Moreover, we have seen that claiming that Korean language is related to the Dravidian languages ( the West theory) is wrong because it’s only based on typological similarities. We conclude therefore that the most accepted theory is the Altaic theory.


Chin-u, Kim. (1983). The Making of the Korean Language. In Korean National Commission for UNESCO, The Korean Language (13-42). Korea: The Si-ya-yong-o-sa Publishers, Inc.

Ohno, Susumu. (1970). The Origen of the Japanese Language. Kodusai Bunka Shinkokai, Tokyo.

Sang-ok, Yi. (1983). The Theory of Altaic Languages and Korean. In Korean National Commission for UNESCO, The Korean Language (43-54). Korea: The Si-ya-yong-o-sa Publishers, Inc.

Woong, Huh. (1983). The Development of the Korean Language. In Korean National Commission for UNESCO, The Korean Language (1-12). Korea: The Si-ya-yong-o-sa Publishers, Inc.

Korean Overseas Information Services. (1990). A Handbook of Korea. (p.45-55 ). Seoul, Korea.

The Korean Language: An Overview. http://www.iworld.net/Korea/Culture/f33.html. 1/9/98.

Language. http://www.littlekorea.com/language.html. 1/9/98.

Maps. http://www.koreanwar.org/html/maps.html. 1/9/98.

Instructor | Textbook & Materials | Course Objectives | Major Learning Activities | Course Requirements & Grading Scheme | Resources | Language Reports | Home

1998-1999 © Dr. Cynthia L. Hallen
Department of Linguistics
Brigham Young University
Last Updated: Monday, September 6, 1999