How do you distinguish between Japanese, Chinese and Korean by sight? Korean language help

There is a general consensus among linguists that Korean belongs to the Altaic family of languages ​​that originated in North Asia and includes Mongolian, Turkic, Finnish, Hungarian and Tunguso-Manchu (Manchu) languages. Despite the fact that Korean and Japanese have some similar grammatical structures, the historical relationship between the two languages ​​has yet to be established.

For writing Korean uses a mixture of Chinese characters (Hanja) and its own Korean alphabet known as Hangul, or Hangul alone, just as, for example, most Indo-European languages ​​use Arabic characters to write numbers.
Due to the wide variety of sounds, Korean there is no problem that exists in Japanese writing, which, according to some experts, must retain a significant number of Chinese characters in order to highlight a large number of homonymous sounds.

Although Korean and Chinese are not similar in terms of grammatical structure, over 50 percent of the total vocabulary Korean language comes from Chinese borrowings. It is a reflection of the cultural dominance of China for over 2 millennia.

A large number of constituent Chinese characters were borrowed from Japan in the nineteenth or twentieth centuries to translate modern Western scientific, technical, and political vocabulary. From there they came to Korea during the colonial period. Post-1945 United States Influence found its reflection in a number of English words that were borrowed in the Korean language. Unlike Chinese, Korean does not cover dialects that are mutually incomprehensible. However, there are regional differences in both vocabulary and pronunciation.

Division into North Korea and South Korea.

It is not clear to what extent the honorary grammatical forms of the language survived in the north of the country. The North Korean regime is pursuing a policy that attempts to get rid of most foreign borrowings, as well as old terms of Chinese origin. Western borrowings also cease to be used.
Pyongyang views hanja, or Chinese characters, as symbols of "groveling," as a result of which they are systematically removed from all publications. An attempt was also made to create new words of exclusively Korean origin. Parents are encouraged to give their children Korean names rather than Chinese names. However, about 300 Chinese characters are still taught in North Korean schools.

Probably, you have repeatedly come across documents that have come to you: certificates of conformity, invoices or price offers, in a completely incomprehensible language. As a rule, everyone thinks that the text came in Chinese, although it is possible that it is Japanese or even Korean. How to visually distinguish Asian languages ​​and send for translation to the correct translation company specializing in Japanese to Russian translation or Chinese to Russian translation, such as translation for business, without tarnishing your reputation by not knowing the basic differences between recording systems? The answer is in our instruction “How to distinguish between Japanese, Chinese and Korean languages».

How to understand where is what language without being a specialist? It's actually not that hard. Let's figure out how not to get lost in the forest of at first glance, incomprehensible squiggles and squiggles.

To begin with, you can weed out the Korean language, since in Korea, actually, hieroglyphs are not used for writing, there is a letter system called Hangul.

This special alphabet was specially designed by order of the Korean emperor in the middle of the 15th century. Koreans are really lucky, because with the advent of their phonetic alphabet, the need for hieroglyphic notations, which were used before the appearance of Hangul, disappeared, although Korean schoolchildren still learn Chinese characters at their desks, in addition, they also write down their names using Chinese characters. You can visually distinguish Hangul by the presence of a large number of small circles and sticks.

This is how the word Hangul is spelled in Korean: 한글 - small circles and sticks can be found here in abundance.

An example of writing in Korean:

전화로 연락이 안되어서 이렇게 글을 드립니다. 내일 뵙기로 한 약속을 취소해야하는 상황이 생겼습니다. 불편을 끼쳐드려 진심으로 죄송합니다.

Which means something like the following: I could not get through to you, so I am reporting by e-mail that I have to cancel our meeting for tomorrow. I apologize for the inconvenience caused.

Chinese and Japanese use the same characters. This is how the languages ​​differ from Korean, in which, as we already know, they abandoned the system of hieroglyphic writing.

You can compute Japanese by looking at the rounded letters interspersed with square characters. These letters look much simpler than hieroglyphs and, as a rule, consist of two or three lines of an oval shape, although they originally originated from the hieroglyphs that Japanese medieval aristocrats simplified to write their language, based on the consonance of Japanese syllables and Chinese pronunciation of the simplest hieroglyphs. A special alphabet "hiragana" was developed, which is also known as the female alphabet, since at that time only men had access to full education and study of Chinese texts, while women used simplified signs with rounded "feminine" forms. The word "hiragana" looks like this in writing with this alphabet: ひ ら が な. Rounded convex shapes are evident, simple graceful lines differ significantly from the square shapes of hieroglyphic writing. Thus, if you periodically encounter such rounded signs in the text, it means that this is a document in Japanese.

As far as Chinese is concerned, only hieroglyphs are used. Chinese characters are almost identical to Japanese, even the combinations of characters used to denote words are basically the same. The reason for this is the borrowing of hieroglyphs by the Japanese from China.

At the same time, the borrowing of hieroglyphs went in waves and in different eras new layers of hieroglyphs and vocabulary came from China, thus, in Japan, one hieroglyph can have several readings, and not one, maximum two or three in rare cases, as in China. In addition to several Chinese readings from different eras, because the Chinese language also did not stand still, but constantly developed, of the same character, the Japanese use the same characters to write their Japanese words, for example, verbs and adjectives. This is where it becomes necessary to write the signs of the "hiragana" alphabet to denote verb endings.

It is necessary to understand that even with a complete coincidence of Chinese and Japanese characters and their combinations and meanings, they are read differently, therefore the Chinese do not understand Japanese speech, and vice versa, and both sides are forced to use the services of translators and learn a foreign language. Fortunately, as a last resort, there is alternative way for communication between Japanese and Chinese, it is written in hieroglyphs. For example, a Chinese person can write what he would like to say using hieroglyphs, and here it will become clear to the Japanese in general, without grammatical subtleties, what the Chinese meant. Likewise, the Japanese can understand the general meaning of what is written in Chinese newspapers, and vice versa.

You can identify the Chinese hieroglyphic notation by the presence of hieroglyphs, which are small drawings or pictograms with their own specific meaning and reading. Take, for example, the notation: 汉字 是 汉语 的 的 最基本 单元, which means "hieroglyphs are the basic unit of writing in Chinese."

As you can see, there are no small circles like in Korean, no round letters like in Japanese. Only clear square shapes, because every hieroglyph, no matter how large and piled up it is, must fit into a square. From this we conclude that before us is nothing more than a Chinese text.

We will consider Japan and South Korea. North Korea will not suit you. Actually, the first drawback can be traced right away: these are all American possessions, where Russians are very much disliked, because the world will not forget the consequences of the Cold War very soon. Now more specifically.

Japan: advantages

* Very interesting culture. Metal fans will be especially surprised: local bands play power metal.

* Quality is everywhere. Everything that surrounds people in that country - everything is done with high quality. Literally everything. Otherwise, it is simply unprofitable.

* The language is very light. At my school I was heavily bombarded by the Russian language: declension / conjugation, gender, case, plural, perfect / imperfect form. This is not the case in Japanese.

* The lowest crime rate in the world.

disadvantages

* The simplicity of the language is offset by a strong contextual dependence. There have been cases when a person saying "yes" meant "no".

* It is easy to find a job only as a doctor and musician. In the case of doctors - the population, the age of which in 3/4 of the cases is more than 60 years. In the case of musicians, copyright, which is especially strictly protected in Japan. In other professions, you need to have either a work experience of 20 years or more (although sometimes 10 is enough), or a recommendation from a serious company, or very non-standard ideas.

* And yes, you cannot hire a foreign worker there for a salary lower than the average in the country.

* Be prepared to die in the workplace. ...

* The Gini coefficient is approximately the same as in Russia.

* Also forget the phrase "Japanese citizenship". You can get it only if at least one of the parents was born there. In other cases, a residence permit is issued. Temporary.

* People are very complex. A very large number of psychological terms in Japanese (). They are also terrible xenophobes: they will make anyone a xenophobe. They are also terrible conformists, this explains the fantastically low crime rate: the slightest offense spoils the whole result of labor, and Asians are very hardworking.

South Korea. An exact copy of the USA, which is not surprising if. But still: the benefits.

* I don't know much about the language, but it doesn't look very complicated.

* Finding a job is not easy, but easier than in Japan.

* You never dreamed of such salaries.

* Very good Gini coefficient and human development coefficient.

* People are very kind and friendly. While studying at the university, I had to communicate, and now - on Twitter.

Disadvantages:

* The culture is very rotten by our standards. Watching TV in that country for residents of the post-Soviet space can end in bloody tears.

* The friendliness of people is due to hypocrisy and commercialism. Undoubtedly, it is so everywhere. But among the South Koreans, this is especially noticeable. Or do you think how to achieve fantastically high salaries in a country where there is only Samsung, Hyundai and Doshirak?

* First place in the world for the average number of suicides ().

In fact, it will take hundreds of printed pages to describe the advantages and disadvantages. But conclusions can be drawn now. If you are just a good specialist - Korea. If you are a true genius of your craft and because of work you do not even see your own bed - Japan.

COURSE WORK

“Nostratic hypothesis. Relationship between Japanese and Korean. "


Introduction

Nostratic hypothesis. Nostratic superfamily and its composition

Lexical and grammatical parallels between Nostratic languages

Japanese language

Korean

Typological evidence of the relationship between Japanese and Korean

The Korean factor in the formation of the Japanese language

Conclusion

Bibliography


Introduction

In its term paper I would like to consider the Nostratic hypothesis, the belonging of the Japanese and Korean languages ​​to the Nostratic language family and, accordingly, the theory of the relationship between the Japanese and Korean languages.

Calculations using the method of linguistic dating, as well as the available cultural and historical foundations, allow us to date the era of the existence of a single pranostratic community from 11-13 millennia BC. BC, which is the most profound of the reconstructions achieved to date.

Apparently, already in ancient times, the Nostratic proto-language was divided into numerous dialects. Many of them completely disappeared millennia ago, while others continued to develop and, as a result, became the starting point for the development of all languages ​​that are now part of the Nostratic superfamily.

Let's try to briefly outline the main stages in the history of Nostratic languages.

The speakers of the Nostratic proto-language belonged, apparently, to the Caucasian race and lived in the territory of Eastern Europe and, possibly, adjacent regions of Asia. Then, as a result of migrations, the ancestors of the modern Altai peoples moved far to the east, reaching the shores of the Pacific Ocean; the Dravids already in historical time moved to the territory of Hindustan (the beginning of this migration dates back to 4-3 millennia BC). The Pra-Kartvelians settled in the Caucasus for a long time, and the speakers of the Semitic-Hamitic languages ​​moved even further south, into the depths of the Arabian Peninsula and North Africa. Proto-Indo-Europeans and Proto-Urals probably remained in the former territories of residence of the bearers of the Nostratic proto-language, with the Indo-European tribes occupying more southern regions.

In the process of resettlement, there was a constant mixing of the speakers of the Nostratic languages ​​with the local population. As a result, those families of languages ​​were formed that are now included in the large superfamily of Nostratic languages.


Nostratic hypothesis. Nostratic superfamily and its composition

The question of kinship of Indo-European languages ​​remained open for a long time. Scientists have noted some lexical and structural parallels between Indo-European languages, on the one hand, and Finno-Ugric, Altai, Semitic-Hamitic, on the other. However, a truly scientific theory was not created until the 1960s, when the so-called Nostratic hypothesis (from the Latin noster "our") appeared. The honor of its development belongs, first of all, to V.M. Illich-Svitych.

According to his hypothesis, Indo-European languages ​​are included in the large superfamily of Nostratic languages. In addition, the following language families are also among the Nostratic:

1. Afrasian (or Semitic-Hamitic). Includes six groups:

a) Semitic (Arabic, Amharic, Syrian, Akkadian, Hebrew and other languages);

b) Egyptian (ancient Egyptian and its continuation Coptic languages);

c) Berber (languages ​​common in the mountains and deserts of North Africa: Kabyle, Shilkh, Reef, etc.);

d) Chadian (Hausa, Angas, Bauchi and some other languages ​​spoken by part of the population of West Africa south of the Sahara);

e) Kushite (the languages ​​of East Africa, the most significant of which are Somali, Beja, Sidamo, Gaul, Iraqi, etc.).

f) Omotic (languages ​​of the South-West of Ethiopia: Gimirra, Ari-banna, Kaffa, Volamo, Ometo, etc.).

2. Kartvelskaya. Includes four languages: Georgian, Chan, Mingrelian and Svan.

3. Dravidian. It includes four large (Tamil, Telugu, Kannada, Malayalam) and a number of small (Kota, Toda, Gondi, Bragui, etc.) languages ​​common in India, partly Pakistan, Nepal and Sri Lanka.

4. Ural. Includes three groups:

a) Finno-Perm (Finnish, Estonian, Sami, Mari, Mordovian, Komi, Udmurt and other languages);

b) Ugric (Hungarian, Khanty and Mansi languages);

c) Samoyed (Nenets, Enets, Selkup and Nganasan languages).

5. Altai. Includes five groups:

a) Turkic (Turkish, Azerbaijani, Turkmen, Crimean Tatar, Tatar, Bashkir, Kazakh, Uzbek, Kyrgyz, Chuvash, Tuvan, Khakass, Yakut and other languages);

b) Mongolian (Mongolian, Buryat, Kalmyk, Dagurian, etc.);

c) Tungus-Manchu (Manchu, Evenk, Even, Nanai, Udege, etc.).

d) Korean (Korean language);

e) Japanese (Japanese, Ryukyus and a number of other languages ​​that are widespread in Japan and are usually classified as dialects of the Japanese language for sociolinguistic reasons).

Some scholars reject the existence of the Altaic family as a unity, similar to the Indo-European or Dravidian language families, however, the fact that all the above-mentioned groups and languages ​​are included in the Nostratic superfamily now almost no linguists have doubts.

This was the originally intended composition of the Nostratic superfamily. However, recent studies have shown that the Nostratic superfamily includes at least two other language families:

6. Yukagiro-Chuvan. Includes North Yukaghir, Kolyma-Yukaghir, as well as the extinct Chuvan and Omok languages. Within the Nostratic superfamily, the Yukaghir family is closest to the Uralic languages.

7. Chukotka-Kamchatka. Includes two groups:

a) Chukchi (Chukchi, Koryak, Kerek, Alyutor languages);

b) Kamchatka (Itelmen and a number of extinct languages ​​of Kamchatka).


Lexical and grammatical parallels between Nostratic languages

V.M. Illich-Svitych managed to find more than 350 lexical matches connecting six given language families and testifying to the kinship of their languages. Among the reconstructed Nostratic words are vocabulary denoting body parts, natural phenomena, basic actions and signs, i.e. the lexical layer that is the most stable and least susceptible to borrowing. To confirm this, consider some of the correspondences between the reconstructed forms and the modern lexemes of the six Nostratic language families.

1. Nostr. * bura "snow (sand) storm": family-boor. * bwr "sandstorm, wind" (Arabic barih "hot wind with sand, Somali fora" strong wind with dust "), Indo-European * bher" storm, rage "(Russian storm, Icelandic byrr" fair wind ") ; Ural. * pura "blizzard" and * purka "blizzard" (Fin. purku "blizzard, blizzard); alt. * bura / bora "storm, blizzard" (Tatar buran "blizzard", Evenk borga "blizzard, blizzard").

2. Nostr. * gura "to swallow": sem -.- boor. * g (w) r "to swallow, throat" (Arabic gr "to swallow"), Indo-Hebrew * guerhu "to swallow" (Russian throat, Tocharian kor "throat", Lit. gerti "to drink", Greek barathron "mouth, throat"); Ural. * kurk (k) a (Finn. kurkku "throat, neck"), Dravid. * kur- "pharynx, throat" (Tamil. kural "throat"), maps. * qorq- (?) "throat" (cargo qorq "throat").

Currently, scientists have discovered 1200 lexical convergences between various language families that are among the Nostratic languages. Certain parellels are also observed in the field of morphology:

Nostr. * -ka diminutive suffix of names: kartv. * k / ak / ik (Georgian mamiko "daddy"), Indo-Hebrew * -k- (Old Ind. asvaka "horse", Russian mouse-to-a), Ural. * kka / kka (Finn. vasikka "calf", Nenets. jaxako "river"), Alt. * ka / * ka (Yakut agakam "my father").

In the history of linguistic groups and languages, individual words have undergone quite significant changes at times, explained by the action of the laws of linguistic development. So, Russian. lynx and drilled. nohoy "dog", it would seem, have nothing in common, but the etymological analysis shows that at the level of proto-forms (Indo-Hebrew * luk "-" lynx ", compare German Luchs, Latin lynx - Alt. * loka / luka" lynx, arctic fox, dog ") these words reveal an undeniable kinship.

In one of his articles, V.M. Illich-Svitych wrote: "A real study of this relationship is possible only at the level of the reconstructed proto-languages ​​of the corresponding groups," therefore progress in the reconstruction of the forms of various proto-languages ​​that are part of the Nostratic leads to an increase in general Stratic parallels and clarification of our knowledge about the structure of the Nostratic proto-language.

Composition and classification of the Altai family

The Altai family includes the Turkic, Mongolian and Tungus-Manchurian languages, as well as, possibly, the Eskimo-Aleutian languages. Some also include isolated Ainu and Nivkh languages. Until this is clarified, the Eskimo-Aleutian, Nivkh and Ainu are conventionally combined into a Paleo-Asian (otherwise - Paleo-Siberian) territorial grouping. According to their grammatical structure, all "classical" Altai languages ​​are agglutinative nominative languages. If the Altai community expands due to a number of Paleo-Siberian languages, many of which are ergative, then the grammatical structure of the latter will be considered, obviously, of substrate origin. And the expanded Altai family itself may already be called the Para-Altai macrofamily, which early separated from the para-Nostratic (i.e., Nostratic in the broad sense - with the Afrasian branch), but later the Afrasian (even more divergent). Well, if the Nivkh with the Ainu fall into the paranostratic association, then the Para-Altai community separated even earlier than the Afrasian. The disintegration of the Altai linguistic community (family) took place, presumably, in 6-5 thousand BC. (according to glottochronology - 17 matches in Swadesh's 100-word list). The Altai language family also includes Japanese and Korean languages, which will be discussed below.


Japanese language

Verbs and adjectives have inflectional forms that are formed by replacing endings; these forms indicate the syntactic position (predicate forms, gerunds), time, mood. There can be suffixes between the stem and the endings with the meaning of passive, causative, negation, politeness different types etc. Other categories of words are not subject to inflection: these include substantives (nouns, pronouns and numbers), adverbs, postpositions, conjunctions and interjections. The usual order of words in a sentence is "subject - object - predicate" (SOV), the definition precedes the defined. The grammatical meaning of nouns, subordinate phrases and sentences is determined by the postpositions following them. Thus, a noun followed by a postposition ga is a subject, a noun followed by a postposition o is a direct object. The ka particle at the end of a sentence turns it into an interrogative. In Japanese, there are certain forms and constructions (called courtesies) that indicate relative hierarchy social statuses the speaker, the addressee and the person in question.

There are (apart from the not widely spread Japanese Latin alphabet) two types of writing. The first type is those borrowed from China in the 6-8 centuries. hieroglyphs ("kanji"). Their number reached several tens of thousands, but in modern writing only approx. 3 thousand hieroglyphs. The second type is phonetic writing, the common name for all its types is "kana". Two varieties of kana are now common: hiragana (more rounded) and katakana (more angular); hiragana and katakana independently of each other evolved from hieroglyphs in the 9-10th centuries. Kana is basically a syllabic writing: a syllable from a vowel and a consonant is written in one sign, the second components of long vowels, diphthongs and end-syllable nasals are written with special characters. In modern texts, hieroglyphs usually indicate the roots of full-valued words, and grammatical elements - affixes, postpositions, particles, conjunctions, as well as interjections - are written in hiragana. Katakana is commonly used to record new borrowings, mainly from of English language that do not have a hieroglyphic way of writing. Normal Japanese text is characterized by a combination of hieroglyphs, katakana, and hiragana; special Japanese punctuation marks, Arabic numerals, and sometimes the Latin alphabet are also used. The usual direction of writing, as in China, is from top to bottom from right to left, although some scientific and informational texts are printed horizontally from left to right. The manuscript distinguishes at least three writing styles: square (more angular), regular, and fluent (more simplified).

The book version of the Japanese language, even nowadays, differs markedly from the spoken one. Many words of Chinese origin are used in writing, where they are understandable due to hieroglyphic notation, but are avoided in oral speech due to homonymy (formal coincidence of words with different meanings). In vocabulary and grammar in the book variants of the language, words and forms borrowed from the old literary language are common. Thus, the Ieba “if someone says” can appear in the book version in the old form of the Iwaba. Many particles and postpositions, lost in the spoken version of Japanese, can appear in the book: for example, ori and nomi can be used instead of kara and dake in the meaning of "from" and "only".

The study of Japanese in Japan has a long history; in fact, Japan is one of the few non-European countries in which a national linguistic tradition took shape and developed, reaching its highest development in the period from the late 17th to the middle of the 19th century; at the turn of the 19th and 20th centuries. this tradition came into contact with the European one. The first acquaintance of Europeans with the Japanese language occurred in the late 16th and early 17th centuries, when Portuguese missionaries settled in the country; they created the first dictionaries (1595, 1603) and the first grammar of the Japanese language (J. Rodrigish, 1604). This was followed by more than two centuries of almost complete closure of Japan to Europeans; connections were resumed only in the 1860s, when there were numerous Japanese grammars written by scholars from various European countries; by this time in Japan there was already a grammar written according to Dutch samples by S. Tsurumine (1833). In the 20th century. the Japanese language became the object of description within the framework of new linguistic trends emerging in the West; in particular, American linguists B. Blok, R. E. Miller constructed descriptive descriptions of the Japanese language; the most complete description of the grammar of the Japanese language in the West was published by S. Martin. Theoretically important results were obtained by Japanese linguists both in Japan proper (S. Hashimoto, M. Tokieda, S. Hattori, etc.) and in the USA (S. Kuno, S. Kuroda, M. Shibatani, etc.); an interesting page in the history of linguistics and sociology was represented by the Japanese "school of linguistic existence", which took shape at the turn of the 1940-1950s. Comprehension of the peculiarities of the structure of the Japanese language had a noticeable influence on the theoretical constructions of such linguists as C. Fillmore, J. McColi, A.A. Kholodovich, W. Chafe. At present, Japanese studies is a large and developed branch of linguistics, which is facilitated by the modern high status of the Japanese language in the world (derived from Japan's status as an economic superpower).

In Russia, the study of the Japanese language dates back to the 18th century, but the intensive development of domestic Japanese studies, like Western European studies, began with the "opening" of Japan to the outside world in the middle of the 19th century. The first Japanese-Russian dictionary was created in 1857 by I.A. Goshkevich, the first grammar by D.D. Smirnov in 1890. regular teaching of Japanese began; Petersburg and Vladivostok became the main centers of domestic Japanese studies; later Moscow was added to them. An outstanding contribution to domestic and world Japanese studies was made by ED Polivanov, NI Konrad, AA Kholodovich; various aspects of Japanese linguistics are devoted to the works of V.M. Alpatov, I.F. Vardul, I.A. Golovnin, N.A. Syromyatnikov, S.A. Starostin, N.I. Feldman.


Korean

The language of Koreans is spoken on the Korean Peninsula, China, Japan, Russia and the United States. Korean is spoken by about 46 million people (1971, estimate). Modern literary Korean (pyejunmal is the standard language) is based on the Seoul dialect of the central dialect; in the DPRK, the Pyongyang dialect is considered the norm (munhwao is a cultural language).

There are 6 dialects in Korean: northeastern (including the Korean dialects of Northeastern China), northwestern, central, southeastern, southwestern and Fr. Jeju.

There are various hypotheses about the origin of the Korean language (Dravidian, Japanese, Paleo-Asian, Indo-European, Altai). Many attribute it to the Tungus-Manchu group of languages. So, Korean is an isolated language of the Altai family (another isolated language of this family is Japanese). Apparently, the ancestors of Koreans came to the Korean Peninsula from somewhere from Manchuria, Mongolia or Altai about three to four millennia ago.

Features of consonantism: the presence of 3 rows of noisy consonants (weak voiceless - aspirated - reinforced voiceless) [b-n-nx as in Greek, Armenian, Pragermanic], which are neutralized at the end of a syllable, and a "two-faced" phoneme "l" / "r" ; uncharacteristic concatenations of consonants at the beginning of a syllable; variety of alternations at the junction of syllables and words. The vowel system is distinguished by the richness of monophthongs and diphthongs, phonemic duration (with a change in pitch); traces of syngarmonism.

There is a strong tendency towards analyticism. A wealth of case forms and grammatical categories of refinement. Word order: subject - object - predicate (SOV); the dependent word is always in preposition.

The Korean alphabet Hangul ("great writing") was developed by a group of scientists under the leadership of King Sejong in 1443. The official date for the creation of the Korean writing system is 1446, when the document "Hongmin chonim" ("Guidance to the people on correct pronunciation") was published. The Hangul syllable is formed by inscribing all the sounds that make it up into an imaginary square (from top to bottom and from left to right), which makes the resulting symbol similar to the Chinese character. However, these are not hieroglyphs - they are just combinations of characters.

The Korean alphabet consists of 24 letters (14 consonants, 10 vowels). Before Hangul, the Koreans used Chinese hieroglyphic writing, which managed to maintain their positions in office work and among the educated ruling elite until the beginning of the 20th century. Due to the high proportion of Chinese words in the Korean vocabulary, a system of mixed hieroglyphic-alphabetic writing has emerged, in which hieroglyphs are used to convey Chinese borrowings, while letters are used to denote verb endings, unchangeable particles and native Korean words (almost like in Japan, only the Japanese and their own original words can be written in hieroglyphs). This type of writing still prevails in South Korea despite attempts to limit or prohibit the use of Chinese characters.

The usual Korean newspaper text, about 3/4 (80%) consists of Chinese borrowings. Therefore, if all words of Chinese origin are written in hieroglyphs, then hieroglyphs will make up about half of the text (since suffixes and endings are still written in the Korean alphabet).

In the most complete dictionary of hieroglyphs, which was prepared about a thousand years ago, 53 thousand characters were taken into account. It is a known fact that some hieroglyphs did not even make it into this gigantic dictionary, so there are even more hieroglyphs, most likely about 60 or even 70 thousand. However, this does not mean at all that a literate person should know all of them. The overwhelming majority of these hieroglyphs are different versions of the same sign or archaisms. Even the most educated people are rarely able to memorize more than 10 thousand signs, while an ordinary person, even in China, where hieroglyphs are used very widely, 4-5 thousand signs are more than enough for life. In Korea and Japan, even a well-educated person rarely knows more than 3000 characters.

Contrary to nationalist propaganda, the introduction of the alphabet is by no means an unconditional blessing, as indicated by the supporters of the widespread use of hieroglyphics, who continue their resistance. They emphasize that hieroglyphics, firstly, is a writing system common to all countries of the Far East - China, Japan, Korea, Taiwan, Singapore, Hong Kong and, historically, Vietnam. Now the strengthening of economic ties between these countries is one of the most important tasks of their foreign policy. Korea's abandonment of hieroglyphics undermines such ties and makes it difficult for Koreans and their neighbors to understand each other. The second argument is that hieroglyphics makes the etymology of words "transparent", makes it easy to understand their origin and, if necessary, simply create new words and expressions from Chinese roots. Compared to new formations from Korean roots or borrowings from Western languages, such neologisms are distinguished by their brevity and ease of use. Thirdly, without hieroglyphs, understanding of special texts is often impossible due to widespread homonymy. Fourth, knowledge of hieroglyphics is a prerequisite for understanding old Korean culture.

Hangul is complemented by hanchcha - Chinese characters. In North Korea, only Hangul is used, in South Korea, both Hangul and Hanjcha are used.


Typological evidence of the relationship between Japanese and Korean

At present, there are many works on the topic of proving the kinship of the Japanese and Korean languages, as well as on the topic of proving the kinship of the Japanese language and various languages ​​of the Altai family. The most significant authors are Starostin, Murayama, Hattori.

It should be noted that the proofs of the relationship between Japanese and Korean languages, for the most part, were based on rather vague phonological and lexical methods.

Actually, in general terms, the method consisted of the following: the reconstruction of the Pro-Japanese phonological system and the reconstruction of the Pro-Korean / Pra-Altai phonological system were done. Then the vocabulary from the so-called "basic vocabulary" of both languages ​​was compared, reconstructed using the reconstruction of phonological systems. After comparing the reconstructed "basic" vocabulary of both languages, it was concluded that the percentage of similar lexemes is sufficient to assert that Proto-Japanese and Proto-Korean are related languages, and therefore modern Japanese is related to modern Korean.

This method cannot be considered successful due to the following considerations:
1) vocabulary and phonology are much less stable levels of the language than the structural-morphological level, therefore the solution to the problem "can these languages ​​X and Y be related languages" should always begin with an answer to the question: "are these languages ​​structurally similar?" , it is possible to proceed to a comparison of vocabulary and phonological systems only after an affirmative answer to this question is received; unfortunately, neither Starostin, nor Murayama, nor Hattori, nor anyone else from the Altaic studies was asked.

2) the reconstruction of the phonological system of the Pro-Japanese / Proto-Korean language is a system that is difficult to verify, it, for the most part, lies outside the field of linguistic science; one cannot guarantee that it was precisely such phonetic laws that operated in Pro-Japanese, since phonetic laws operate within groups and families of languages, and are not universal;

3) the data in question very often did not represent any system; it often seems that the author of this or that work, since he has dealt with both Japanese and Korean for many years, intuitively feels their kinship and tries to convey his intuitive conviction to the readers through a kind of "shamanic ritual";

The significant advantages of the typological method are that there is no need to carry out reconstruction, the reliability of which cannot always be guaranteed, the languages ​​being compared are compared either directly or through some other languages, this makes it possible to quickly and easily assess whether these languages ​​can be genetically related and in the case of a positive answer, move on with confidence, proceed to identifying material similarities:

1. Both Japanese and Korean are 100% agglutinating languages.

2. In both Japanese and Korean prefixing is prohibited, the linear model of the word form is Altai.

3. In both Japanese and Korean verbs do not change by person and number: Comparison of the dictionary form:

meaning

Japanese

Korean

there is a structural similarity in the vocabulary forms: standard verb ending in Japanese (u / ru) standard verb ending in Korean (ta / da).

In both Japanese and Korean, specific verb forms are mostly formed from stems, in Japanese verbs have 5 stems, in Korean - 2. Both in Japanese and Korean, verb stems are formed according to a similar algorithm - changing the ending of the verb word form.

Both Japanese and Korean have 2 basic grammatical tenses: past and present - future.

Let's compare how the past tense is formed in Japanese and in Korean:

japanese korean

dictionary form dictionary form

elapsed time elapsed tense

go go

itta katta

do do

sura hada

Sita Haetta / Hatta

live live

bag of bald

sunda saratta

there is a clear structural similarity in the way the past tense is formed in both languages.

Long view comparison:

In Japanese, the long form is formed from the TE-form and the auxiliary verb IRU ~ IRU to the existential verb used with animate objects.

In Korean, the long form is formed by attaching the ending KO + the verb ITTA - "to be" to the root of the verb.

I now live in Seoul

Jap .: IMA wa BOKU ga SOURU de SUNDE IRU.

corr .: CHIGIM un NE ga SOUR e SALGO ITTA.


The Korean factor in the formation of the Japanese language

Scientists began to study the problem of the relationship between the Korean and Japanese languages ​​more than three centuries ago. True, up to 1945 in Japan, this problem was dealt with quite little due to the fact that the ruling circles instilled the theory of the "divine origin of the Japanese nation", which denied the very possibility of the existence of any relatives. In 1717, Arai, in his book Oriental Tastes, compared the two languages, and at the end of the 18th century Fujihara Shoguhatsu put forward a theory of the relationship between the Japanese and Korean languages. These works, devoted to the study of the relationship of two languages, were often limited to a superficial comparison of words with similar pronunciation from the vocabulary of the Korean and Japanese languages, not paying attention to other patterns. Since the beginning of the 20th century, the study of Korean-Japanese language relations by Japanese scholars has been mainly aimed at proving the origin of the Korean language from Japanese, which reflected the biased research that was consistent with the policy that Japan pursued towards Korea. The most representative scientists engaged in research in the field of Korean-Japanese language relations in the pre-war period were Kanazawa Shonaburo, Shimmura Izuru, Oguru Shimpei. With the active penetration of Japan into the international arena at the end of the 19th century, Western scholars also paid attention to the problem of the origin of the Japanese language (Gutslaff, de Rosny, Edkins, Aston). But unlike Japanese scientists, Western European linguists and historians have put forward several hypotheses about the genetic relationship of the Japanese language. The most popular of these were the "northern" and "southern" theories of the origin of the Japanese language. Supporters of the "northern" theory, trying to prove the relationship of the Japanese language with the Altaic languages, often compared only their lexical similarity, which could not but carry the factor of chance. The first scientifically systematized work in this regard was the research of Aston V., expressed in his work "A Comparative Study of the Japanese and Korean language", published in 1879. His work had a great influence on scholars engaged in scientific research in the field of kinship of the Japanese language. Nevertheless, it was not possible to find a strict proof of the relationship between the Japanese and Korean languages ​​in the first half of the 20th century, although the founder of Altai linguistics, Ramstadt G., conducted research in this area rather deeply, and a professor of Helsinki University in 1920 for the first time concretized the relationship of the Japanese language with the Buyo language ... This led to the fact that many scientists began to lean towards the version of the origin of the Japanese language from the Austronesian languages.

In 1910, the scientific work of Kanazawa Shonaburo "On the common origin of the Japanese and Korean languages" was published, which was aimed at confirming the theory of kinship between Koreans and Japanese. However, these works only compared lexical similarities and did not pay attention to other patterns in the process of studying the relationship between Korean and Japanese.

But in the post-war period, when all sorts of ideological frameworks were thrown away and scientific circles received great opportunities in the scientific study of the origin of the Japanese language, Korean and Japanese scientists (Ono Susumu, Hattori Shiro, Hashimoto, Lee Ki-moon), who were actively involved in the history of the language and dialects of Korea and Japan, as well as the analysis of ancient folklore texts, in the process of comparative study of Korean-Japanese relations, stable and regular phonetic rules were found, indicating a close relationship between the Korean and Japanese languages.

In 1963, an article by the Korean linguist Lee Ki-moon appeared in the Japanese magazine Chosen Gakuho, which gave a new direction in the search for the origins of the Japanese language. Referring to the material concerning the ancient languages ​​of the Korean Peninsula, Lee Ki-moon and the supporters of his theory (To Su Hi, Murayama S.) found that the closest relative of the Japanese language is the language of the Koguryo state and the tribal union of Kaya (Karak), which were subsequently absorbed by the Silla language. conquered these state formations. After the unification of Korea, the Silla language - the direct ancestor of the modern Korean language - gradually began to assimilate the languages ​​of Goguryeo, Baekje and Kai, as a result of which a standard language for the Korean nation was formed on the territory of the Korean Peninsula. The Korean linguist Lee Geun-soo, relying on the identity of the suffixes Goguryeo and Silla, believes that Goguryeo and Sillach languages ​​had a common linguistic connection, being dialects of one language, one of which later incorporated others. According to Samguk Sagi, 6 Sorabol communities, which later became the basis for the creation of the state of Silla, originated from Joseon, in the ethnic composition of which the Koguryos occupied a significant place. Based on this, it can be assumed that the Goguryeo language was an intermediate link between the Old Japanese language and the Silla language, especially since the Goguryeo language has certain parallels with both the Old Japanese and Sillach languages.

On this moment in the historical annals “Samguk sagi” and “Samguk yusa” only about 80 words of the Goguryeo language and a little more than 10 words of the Kai language, mainly of a topological nature, have been preserved. But, despite the lack of material, the similarities between many words of the Old Japanese, Goguryeo and Kayan languages ​​are quite obvious. As an example, here's a comparison of several words from Goguryeo, Kayak, and Old Japanese:

Ancient japanese Goguryosky Kayaski
Mouth kuti - kuti
Sea us us us
Earth na-vi on on
Valley tani - tan
Water mi mee mee
Mountain take tar tar
Wood ki ky -
Door then - torus
Hare usagi osagam -
Bear godfather coma godfather
Garlic the world mair -

The similarity of the numerals of the Goguryeo and Old Japanese languages, first noted by Shimura Izuru, also speaks of the kinship of languages.

These examples clearly indicate the close kinship ties between the Japanese language and the Goguryeo language, which was spoken by the peoples living in the region of North Korea and South Manchuria. Such a close relationship of the Goguryeo language, apparently, can be explained by the invasion of the Japanese islands by North Korean semi-nomadic tribes, whose language, after the seizure of the islands, began to be used as a means of communication not only among aliens, but also among local residents.

But, despite the obvious facts of a close relationship between the Goguryeo and Old Japanese languages, Japan maintained the closest contacts noted in the annals with Baekje, whose language was related to ancient Korean, but at the same time absorbed Goguryeo elements, since the ruling class that created the state of Baekje, was of goguryeo origin. Judging by the chronicles of the Samguk Sagi, the rulers of Baekche, as well as Goguryeo, traced their ancestry from the Puyo clan. Reflecting their kinship with the puyos, the pekic kings named one of their capitals “Buyo”. In their numerous letters from the Baekje wans to the court of the Chinese emperors, the unity of the origins of Goguryeo and Baekje has been repeatedly emphasized. In the Chinese chronicles "Nanshi" it is reported that the language of the Baekche people is similar to that of Goguryeo.

In general, the Korean linguist To Su Hee divides the language into two categories: 1) the early Baekje language, related to the Goguryeo language, and 2) the late Baekje language, close to the Silla language.

In the early period, when representatives of Goguryeo began to migrate from the northern part of the Korean Peninsula to the region of the Hangang River, a language belonging to the northern language branch of Buyo was spread in Baekje. In the process of the seizure of the lands of the Mahan tribal union by the state of Baekche, the Goguryeo language, used by settlers from the north, was mixed with the language of local tribes, which belonged to the southern language branch of the Khan. Despite this, the Baekje state was characterized by bilingualism, when the aristocratic circles and the local population continued to speak their own languages.

In the later period, as a result of prolonged wars with Goguryeo, Baekje lost its northern territories, where the central headquarters of the Baeki kings was located and the bulk of the people from Goguryeo, who belonged to the northern language family, lived. The loss of territories in the Han River region and the transfer of the capital to the inner regions of Baekje resulted in the fact that differences between the Baekje and Goguryeo languages ​​gradually began to appear, associated with the introduction of the language of the local Khan tribes into the “aristocratic” language of Baekje. It was during this period that the most active relations between Baekje and Japan took place, including at the linguistic level. Some scholars suggest that in the 6-7 centuries, the official language of communication in the first capital of Japan - Nara was precisely the Pekic language. To Su Hee, emphasizing that most of the words from the Baekje language that have survived to this day, refer precisely to the late period, made a comparison with ancient Japanese material, thereby trying to prove the kinship between Baekje and Japan.

Baekje language Old Japanese
1. kas, kat "side" side kata
2. koro "horse" kuruma "horse-drawn carriage"
(It is quite possible that this word penetrated into the Japanese language under the influence of the invasion of the "conquering horsemen" of goguryeo origin, who introduced the terminology of nomadic peoples into the vocabulary of the inhabitants of the Japanese islands).
3. coma, godfather "bear" godfather "bear"
(In ancient times, animals of the bear family did not inhabit Japan, therefore the appearance of the word "bear" is the result of a direct borrowing from the language of settlers from the Korean Peninsula. who arrived in Japan, as, according to legend, they were the descendants of the son of Heaven and the bear. It is no coincidence that the Japanese subsequently called the state of Goguryo nothing but “Koma”.
4. kuti "name of the bird" kuti "falcon"
(This word is not found in the Korean annals, but in the 11th Nihon shoki scroll it is reported that a strange bird was presented to Emperor Nintok, which none of the courtiers knew. It is found in Baekje, is easy to tame and serves people well. Since the Baekche people called this bird "kuchi", the name was borrowed from the Baeke language meaning "falcon").
5. kyry "treasures" chicken "treasury"
6. ki, kui "castle, fortress" ki, kui "fortress"
(as previously noted, after the death of Baekje, refugees from this state who settled in Japan, fearing an invasion of the Tansko-Sill army, began to build fortifications of the Korean type. ...
7. koto "musical instrument" koto "old musical instrument"
(Despite the existence in the "Kojiki" of the legend about the Japanese origin of this musical instrument, experts believe that the word "koto" was borrowed from Baekje).
8. makumo "musical instrument" makumo "wind musical instrument"
(This instrument was used specifically for performing Korean-style music).
9. us "sea" us "wave"
10. tal "mountain" take "mountain"
11. there, tum "round" there "surround"
12. ra (na) "earth" on (but) "field"
(The word was clearly borrowed from the Korean language, since it is often found in place names of ancient Korea. For example, the component na (ra) is included in the names of such states as Silla or Sorabol (Japanese Shiragi), Imna (Mimana), Thamna and etc.).
13. miles "water" midu "water"
14. mal, muri "group" mura "group", "flock"
15. puri "village", "village" mura "village" (stand. pure)
16. syoma "island" sim "island"
17. suri "top" litter "sky"
18. chira "little stone" tiri "dust"
19. Koran "dark red" kura "dark", kuro "black"
20. supi "red"

sabi "rust"

sofa "earth red"

The above lexical similarities speak, first of all, not about a simple borrowing of words from the languages ​​of Goguryeo, Baekje and Kai, but about the close relationship of these languages ​​and the methodical and phased penetration of the Korean language into the Japanese archipelago, confirming the historical relationship between state formations on the Korean Peninsula. and Yamato.

This relationship completely denies the assumption of some Japanese scientists (Egami N., Hattori S., Ishida E.) that, despite the presence of a common language ancestor, Japanese and Korean languages ​​separated several thousand years ago, as a result of which the Japanese language was formed already in the Yayoi era. On this occasion, Arutyunov S.A. takes a neutral position, believing that during the Yayoi period there was no significant difference between the languages ​​of the newcomer population of Northern Kyushu and representatives of the central regions of Japan, and therefore there were no special shifts in this respect during the period of conquest.

Moreover, most linguists are inclined to believe that in the Yayoi era, the Japanese language is just beginning to form. Japanese linguist Ono Susumu believes that the Japanese language was introduced to Japan in general only after 300 AD.

The fact that the process of the formation of the Japanese language under the influence of the Korean tribes stretched out for a long period is evidenced by the fact that, despite certain differences, the representatives of the Ryukyu Islands are close to the Japanese linguistically, anthropologically and ethnographically. According to glottochronology, the separation of the Japanese and Ryukyu languages ​​took place 1,500 years ago, i.e. around the 4th century AD Based on this, it can be assumed that at this time a large group of people appeared in Northern Kyushu, who arrived from the Korean Peninsula and displaced some of the people living there from the area. As a result, the ancestors of the Ryukyus people left to the south, and the aliens, who can be called the ancestors of the Japanese, settled not only in Kyushu, but also on the island of Honshu, Shikoku and other adjacent islands. This is evidenced by the fact that the Kyushu dialect is intermediate between the Ryukyu and Western Japanese dialects. In addition, the largest number of neologisms that came from the mainland are characteristic of the Kyushu dialects, while the largest number of archaisms is found in the Japanese dialect of the Tohoku region and the southern dialects of Ryukyu.

Thus, based on the fact that for many centuries there have been very active contacts between the inhabitants of the Japanese islands and alien tribes, and also, taking into account the parallels between the languages ​​of Goguryeo, Baekje and Karak, which are mentioned in ancient Chinese sources, Japanese the language can be correlated as related to a group of ancient Korean languages, which can be called a puyo group. As for the Ryukyu language, it can be attributed to one of the branches of the Proto-Korean language, which was spoken by the inhabitants of the Korean Peninsula during the Neolithic period.

Of course, Japan's transition to a policy of self-isolation led to the fact that the languages ​​of those tribes that did not manage to assimilate with the settlers of the Yayoi and Kofun period, who represented tribes related to Koreans, intervened in the further process of the evolution of the Japanese language. This led to the active borrowing of lexical elements from the Australian language, the phonetic transformation of the Japanese language, which led to the simplification of speech. But be that as it may, the tribes, which for a long time were in close linguistic contacts with their neighbors, gradually and for a long time retain the lexical, phonetic and grammatical features of the foreign language system. Until now, the composition of the modern Japanese language reveals a clear affinity to the ancient Korean terms of hunting and primitive agriculture, as well as vocabulary associated with parts of the human body, natural phenomena, flowers, animals, clothes, etc.

M. Swadesh found that in 1000 years about 19% of the basic vocabulary are replaced in any language. Based on Swadesh's statement and given that 60-70% of words in Korean and Japanese are words of Chinese origin, it can be assumed that the Korean and Old Japanese languages ​​are close to each other and the prevailing influence of Koreans on the formation of the modern Japanese nation, although the participation of aborigines in certain areas is not excluded. Proving the kinship of the Korean and Japanese languages, it is known that the Pro-Japanese language, within the hundred-word list, has 18 coinciding words with the Pra-Türkic language, 17 coinciding with the Pra-Türkic language, 15 with the Pra-Tungus-Manchu language, and with Austronesian - only 6. At the same time, Pro-Japanese and Middle Korean languages ​​within 100 words have 25 matching words. Applying a similar lexicostatistical procedure, Starostin, on the one hand, dispelled the theory of the Austronesian origin of the Japanese language, and on the other, noted that the Japanese language is approximately equidistant from Turkic, Mongolian, and Tungus-Manchu, but is clearly closer to Korean, which gives him reason to assert, that in the Altai language family there is a special “peninsular subgroup in the composition of the Korean and Japanese languages. Based on these data, Starostin believes that this peninsular subgroup separated from the general Altai family in 3-4 thousand BC.

According to Arutyunov S.A., in the Korean language, such processes as the divergence of the adjective and the verb, the erasure of harmony, the simplification of the vowel composition, proceed slowly, while in Japanese they were accelerated and are now either completed or close to completion. Apparently, there must have been very serious reasons for this, which began to operate after the territorial and cultural separation of the Japanese language from its related Korean. Ono Susumu, who was engaged in a lexical comparison of the Korean and Japanese languages, reflected in his works the kinship of about 1000 words. He cited the account and meaning of the roots of the Korean language, which made it possible to establish a relationship between those roots of the Old Japanese language that might seem a little alike. For example, the ancient Japanese word mu ("eye"), akin to the Korean nun, gradually transformed into me. True, some radical researchers of the Japanese language, based on the fact that in the ancient Korean language, as well as in modern Japanese, there was an open syllable law, they believe that the Japanese still speak the ancient Korean language, while the Korean language there was a modernization. As evidence, the fact is given that in ancient times the word “morning” was pronounced by Koreans as asa. In Japan, this word is still used unchanged, while in Korean it was transformed into achim. Thus, Korean scholars argue that modern Japanese is a frozen form of the “island dialect” of the Korean language. An example is the similarity of Japanese to the Jeju dialect. For example, an invitation to enter in Standard Korean sounds like oso osipsio, whereas on Jeju Island the phrase is pronounced like irushimashi (compare with Japanese, irashiyimase).

To determine the genetic relationship of two languages, conditions are necessary under which there would be a system of stable phonetic correspondences between modern languages, a large number of coinciding basic lexical units and grammatical morphemes.

To confirm the genetic relationship between Korean and Japanese, here are a few examples showing the lexical similarity of these languages ​​at the present stage:

Korean Japanese
1. pada cotton wool "sea"
(The famous Japanese lexicologists Shiratori Kurakito, Kanazawa Shozaburo, Ueda Monnei, Murakami Naojiro and others. It was believed that the Japanese word vata with the meaning "sea" is of Korean origin. This version is supported by the fact that the word bata still exists in the Ryukyu dialects “Bay.” This once again confirms the fact that during the Yayoi period, a significant group of people who spoke the proto-Korean language arrived on the island of Kyushu from the Korean Peninsula. Ryukyu. Later, in the process of transformation and development of the language, when the ancient people ceased to associate the sea only with the strait, through which there was a regular communication between the Korean Peninsula and the Japanese archipelago, the inhabitants of Japan began to denote the sea with the word umi, while the word vata began to carry the meaning " to cross the sea ”,“ to come from across the sea ”, as a reminder that the ancestors of the Japanese arrived and to the Japanese islands from across the sea. (For example, in "Kojiki" the state of Baekche is called vata-no miyake, which is translated by Japanese translators as "overseas granary." Tsushima watari (Strait of Tsushima), vata no hara (sea space), watanaka (in the sea), Watatsumi (sea deity).) At the same time, representatives of Ryukyu still retained the language they spoke at the turn of the century ...
2. horani Torah "Tiger"
(There were no felines in ancient Japan, and therefore the Japanese learned about them only from eyewitness accounts, that is, immigrants to the Japanese islands from the mainland, at the same time borrowing words).
3. stalemate that "field"
(The agricultural term ta "field" in Japanese is very common, especially in place names and names. For example, out of 5736 place names in Niigata Prefecture, 926 (1/5 of all names) include the word ta. Meanwhile, this term is of Korean origin , reflecting the contribution of settlers from the Korean Peninsula in the development of land and agriculture in Japan. It is in the modern territory of Niigata Prefecture that the ancient "Fudoki" reflect the intensive settlement of newcomers from Korea and the development of land. , later transformed into the ancient Japanese pata (now pronounced -hatake) and shortened to one syllable - ta.It is also assumed that the Japanese word ta comes from the Korean ttan "earth".
4. chock currents "time"
5. chot titi "milk"
6. cat koto "thing"
7. catfish sim "Island"
8. mom mi "body"
9. mit moto "bottom"
10. at ue "top"
11. lump godfather "bear"
12. sasym sissy "deer"
13. turim tsuru "crane"
14. tal (so) tory "hen"
15. pam hebi "snake"
16. Paul hara "steppe"
17. piccal hikari "Colour"
18. panyl hari "needle"
19. khal hazel "knife"
20. sat sati "arrow"
21. namul nama "vegetables"
22. cat kasa "straw hat"
23. days susu "coal"
24. mayl mura "village"
25. nal naru "be born"
26. nal on "raw"

In these examples, it should be noted that Korean root words are mostly monosyllabic with vowels at the end, while Japanese root words are most often two-syllable. Apparently, the differences arose due to the fact that the words of the ancient Japanese language, gradually moving away from their common lexical similarities with the Korean language and obeying the open syllable law that became mandatory in the Japanese language, were weighed down by a new vowel, most often -a. This trend can be traced on the example of the Korean word pol ("field"), where the correlation between the Japanese and Korean languages ​​is clearly visible. In the process of transformation of the Japanese language –п, which also existed in the Old Japanese language, was replaced by the modern –х, and the vowel –а was added to the root according to the open syllable rule. Thus, the Korean word pol was transformed into chorus. But in the process of reducing the number of vowels in Japanese and the disappearance of the open vowel -o:, it was replaced by the convenient sound -a. this led to the formation of the modern Japanese word hara.


CONCLUSION

Having cited just a few examples of lexical similarities and based on serious research by Japanese, Korean, and also Western scholars, we can say that in the Korean and Japanese languages ​​there have been and still exist a number of general trends in the development of the language, indicating that the Japanese language has with the Korean language. more related ties than with other languages ​​of East Asia. This gives grounds to speak of the separation of the Korean and Japanese languages ​​into a special subgroup of the Altai language family, which began to form in 3-4 thousand BC. Such a close relationship between the Korean and Japanese languages, most likely, was associated with serious migration processes between the Korean peninsula and the Japanese archipelago in 1000 BC - 1000 AD.


Bibliography:

1. Kieda M. Japanese grammar, vols. 1-2. M., 1958-1959

2. Feldman N.I. Japanese language. M., 1960

3. Sexton M.T., Shapoval V.V. Genealogical classification of languages. M., 1986

4.www.forum.orientalica.com

5. Illich-Svitych V. M. Problems of Indo-European Linguistics, M., 1964

6. Alpatov V.M. Japan: language and society. M., 1988.

7. Starostin S.A. Altai problem and the origin of the Japanese language. M., 1991

8.www.rauk.ru/modules

9. Alpatov V.M. Japanese language.

10. Encyclopedia Languages ​​of the World. Mongolian languages. Tungus-Manchu languages. Japanese language. Korean. M., 1997

11. Baskakov NA Altai family of languages ​​and its study. - M., 1981

12. Kotvich V. Research on the Altai languages. - M., 1962

13.www.wikipedia.org/Korean language

14.www.wikipedia.org/ Japanese


Illich-Svitych V.M.Problems of Indo-European Linguistics, M., 1964

Sexton M.T., Shapoval V.V. Genealogical classification of languages, M., 1986

Baskakov N.A.Altaian family of languages ​​and its study. - M., 1981

Alpatov V.M. Japan: language and society. M., 1988.

Kieda M. Japanese Grammar, vols. 1-2. M., 1958-1959

Www.wikipedia.org/ Japanese

Encyclopedia Languages ​​of the World. Mongolian languages. Tungus-Manchu languages. Japanese language. Korean. M., 1997

Www.wikipedia.org/Korean language

Www.forum.orientalica.com

Www.rauk.ru/modules

S.A. Starostin Altai problem and the origin of the Japanese language. M., 1991

Encyclopedia Languages ​​of the World. Mongolian languages. Tungus-Manchu languages. Japanese language. Korean. M., 1997