FAQ FOR SCI.LANG.JAPAN - GENERAL INFORMATION ON JAPANESE LANGUAGE

WRITTING SYSTEMS

This section is a contribution by Olaf Meeuwissen (olaf@usc.edu). Sources: Modern Japanese by Mieko Shimizu HAN, Dictionary of Japanese Culture by Setsuko KOJIMA and Gene A. Crane, Kanji Education by Jason Molenda and private communication with Ken Lunde (author of JAPAN.INF).

[INLINE]

Jump to Sounds and Writing Order of Japanese Characters

General Information

Traditionally, Japanese is written from top to bottom and right to left. Pages are bound at the right hand side, making books and magazines appear as if they are printed backwards to Westerners. Nowadays the younger generation tends to write the Western way (unless for official purposes), that is left to right, top to bottom. This way is also used in scientific books where text is used along with mathematical formulas which are necessarily (really?) horizontal. Literature is printed the traditional way. Written Japanese employs four character sets: romaji, hiragana, katakana and kanji. Spacing between words with the latter three is up to the writer. In books for beginning readers spaces between words may be abundant, but in more advanced texts they may be few. The equivalent of a period is called "maru", a small circle. A comma is represented by a "ten", a long dot. Its use is not governed by any strict rules (like in most languages). Quotations are indicated by the top left and bottom right parts of a square. Question and exclamation marks are used once in a while. The particles ka and ne fulfill the respective functions, so that there is no real need for either mark.

Hiragana and Katakana

The kana, hiragana and katakana, are strictly phonetic syllabaries. Hiragana, the more flowing form of the two, is said to have been invented by the Buddhist priest K&ucirckai (774-835), although this is controversial. Each hiragana symbol derives from a Chinese character of the same sound, but is devoid of any meaning. The invention of hiragana had a significant impact on Heian literature (794-1185) as it enabled women to write (education was a virtually male-only privilege). The most famous contribution, no doubt, is "Genji monogatari" [The tale of Genji] by MURASAKI Shikibu. Katakana, literally "side script" and the more angular of the two kana, is said to have been invented by KIBI no Makibi (693-755). Each katakana symbol derives from parts of a Chinese symbol in the same way as the hiragana do. Katakana were initially used as a pronunciation aid in Buddhist scriptures, but were mixed with Chinese characters from the ninth century on. Both syllabaries use two diacritical marks to modify the pronunciation of several of the kana. Two small dots at the upper right hand corner produce a voiced version, that is: k->g, s->z, t->d and h->b. A small circle at the same position changes h to p. Finally, the kana for "ya", "yu" and "yo" can be combined with the "-i" kana to form a palatalized sound, for example: pi+ya->pya. Knowing hiragana and/or katakana one can write anything one wishes to write in Japanese. In practice however, hiragana are used for originally Japanese words and katakana is reserved for foreign words. Katakana can also be used for emphasis, very much like italics are being used for emphasis in the roman alphabet. This is relatively common in manga. Moreover, the use of kana is, depending on the level of education, limited to (parts of) words which have a grammatical function, such as particles, conjuctions and inflectional endings. The content of what one wants to express is written in kanji. One other use of the hiragana is as pronunciation aid for uncommon kanji. Found on the right hand side or above the kanji, they are referred to as furigana ("handicap kana") or yomigana ("reading kana").

Kanji

The kanji have been borrowed and adapted from the Chinese characters between the sixth and tenth century. During and after their introduction they underwent many changes, both in Japan and China, so that now both languages have their own set of kanji although many similarities exist. After the Meiji restoration (1868) the government has tried to simplify and standardize the Japanese kanji (I'm not sure about this!). In 1946 the last big government interference resulted in the T&ocircy&ocirc kanji. This set contains 1850 kanji, 881 of which made up the Ky&ocirciku kanji. The Jinmei-yôkanji contained an additional 85 kanji in use for family and geographical names. The Jinmei-yôset can be seen as a set of standard non-standard kanji. The Ky&ocirciku set was to be mastered by the end of elementary school and the T&ocircy&ocirc set by the end of high school. Nowadays, the Jinmei-yôset contains 284 kanji, the Gakushûset, formerly Ky&ocirciku, 1006 and the J&ocircy&ocirc, formerly T&ocircy&ocirc, 1945. The Gakushûkanji are still a subset of the J&ocircy&ocirc kanji and taught in elementary school. The J&ocircy&ocirc kanji are taught in high school. It is not necessary to know all J&ocircy&ocirc and Jinmei-yô kanji to be able to read most Japanese. Several studies indicate that knowing about 1000, for example the Gakushû enables one to read some 90%. Knowing the J&ocircy&ocirc allows you to read about 99% of written Japanese. However, not knowing a kanji does not necessarily mean that one can not guess its meaning. Many of the more complicated kanji are combinations of simpler kanji (see below). The kanji are ideographs representing objects and/or ideas. For each kanji three things must be learned: its root meaning, its writing (stroke order and stroke endings) and, the most difficult, its reading (pronunciation). Generally, every kanji has two types of reading. On-yomi or Chinese reading is based on the Japanese imitation of the ancient Chinese pronunciation of the character. Kun-yomi or Japanese reading represents the original Japanese word corresponding to the ancient Chinese meaning of the kanji. Put differently, on-yomi is a mapping of Japanese sounds on the spoken form of written Chinese whereas kun-yomi is a mapping of spoken Japanese on written Chinese. There are no strict rules to decide whether a kanji is read on-yomi or kunyomi, but as a rule of thumb: a kanji occuring in a compound with other kanji is read on-yomi, a kanji all by itself is read kun-yomi. Note that a kanji may have more than one on-yomi and/or more than one kun-yomi. Fortunately, there are some kanji that have either only one on-yomi or only one kun-yomi. An example should make all this a bit clearer. Take the kanji for moon. By itself, just plain moon, it is read "tsuki" (kun-yomi). This is the only kun-yomi it has. Well, let's be fair, it can also be read "zuki", for example in "mikazuki", "three day moon" (->new moon), but this is merely a change for ease of pronunciation (?). By the way, "mikazuki" is a severe exception of the rule of thumb: all three kanji are read kun-yomi fashion. In "first moon" (->first month-> January) it combines with the kanji for one, "ichi", and is read "gatsu". Thus, "ichigatsu" means January. In "moon day" (->Monday) it combines with the kanji for day, "yô bi", and is read "getsu". Thus, "getsuy&ocircbi" is Monday. So, "moon" has two (accidently similar) on-yomi. In many cases the more complicated kanji, in terms of stroke count, are made up of simpler kanji. Their meaning can usually be inferred from the meaning of the simpler kanji. The simple kanji used in those more complicated ones are referred to as graphemes or radicals. Due to size restrictions, every kanji takes up the same amount of space, these graphemes or radicals can become somewhat distorted with respect to the original. In two-grapheme kanji such as "bright", where "sun" and "moon" are placed alongside (and squeezed), the left hand side is called "hen" and the right hand side "tsukuri". Shodô or calligraphy is practiced by many Japanese. Basically, there are three styles to write the kanji: kaisho (print or block style), gy&ocircsho (cursive style) and s&ocircsho ("grasshand" style). Kaisho resembles the style of kanji in mechanically produced texts. Although primarily referring to calligraphy, kaisho is also used to indicate the style of these texts. Formal correspondence is usually written in kaisho, whereas gy&ocircsho is used in personal correspondence. Artistic writings on screens and monuments and in paintings often use s&ocircsho. Easily recognized by its reminiscence to flowing reeds, it is the hardest to read of the three.

Romanization (Romaji)

Romaji is basically a method of writing the Japanese language in the roman alphabet (the very same as the one you're reading now). Originally developed by Jesuit missionaries in the sixteenth century, there now are various methods of romanization. The most common one is the (modified) Hepburn system (hebon-shiki or hy&ocircjun-shiki). It was established by James C. Hepburn (1815-1911) who also founded Meiji Gakuin University. Its popularity stems from the better compatibility with the Japanese sounds than kunrei-shiki (the official, but virtually obsolete, system) and Nihon-shiki. Nihon-shiki is governed by simple rules and is taught in elementary schools. The use of romaji, or roman alphabet, in everyday Japanese is very limited. Abbreviations, advertisements and street signs are among the few instances you will see them. One area in which romaji are in wide-spread use is in text books for studying Japanese for foreigners. Most intermediate and advanced students of the language will agree that the use of romaji will prove an obstacle in learning how to read "real" Japanese.

PRONUNCIATION

The following is a description of the sound system of Japanese. The first three sections are a brief linguistic summary. The next section focuses on practical pronunciation problems faced by foreigners, especially speakers of English. Check also the Quick Pronunciation Guide.

The Sounds of Japanese

The following are the meaningful sound elements (phonemes) of Japanese:

Note to linguists: The above phonemes should be enough to account for all native Japanese and assimilated Sino-Japanese words. However, some recent borrowings from other languages may require additional phonemes (the Japanese words for 'tsar', 'violin', etc.).

Phonemic Rules

The phonemes listed above may change when they are actually pronounced. The following are some of the most important phonemic rules.

Accent

The accent system of Japanese is based on pitch, not stress. Each syllable is pronounced with about the same force, but accented syllables are pronounced at a higher pitch. See also Accent (problems for learners of Japanese) below. Some generalizations: All high-pitch syllables in a word are contiguous. If the first syllable is high, then all the remaining syllables are low. If the first syllable is low, then the second syllable is high. One-syllable words may have either a low rising pitch or a high falling pitch. The following examples, given in standard romanization, are taken from the introduction to the dictionary Daijirin (Sanseido, 1988). These are all of the single-word accent patterns up to six syllables. Capital letters indicate the high pitch. The mark ^ at the end of a word indicates that the high pitch may carry over into the following word (especially if the following word is a particle such as wa or ga ).

Problems for Learners of Japanese (English speakers)

SHORT AND LONG SOUNDS

All of the vowels and most of the consonants of Japanese may be pronounced either short or long. In normal speech, long sounds are held for about twice the length of time as short sounds. The difference between long and short sounds is very significant, and Japanese have no trouble at all remembering which words have short sounds and which words have long sounds. Here are some minimal pairs, i.e., words distinguished only by the length of the sound:

CONSONANT + /Y/ CLUSTERS

When the semivowel /y/ follows a consonant, learners of Japanese often insert a vowel between the consonant and /y/. However, no vowel belongs there and the presence of a vowel can lead to misunderstandings. It may take some practice to learn how to say these clusters smoothly. Examples:

/R/

In many dialects of English, /r/ is pronounced before vowels with the lips slightly rounded and the tongue not touching the roof of the mouth, while /l/ is pronounced before vowels with no lip rounding and with the tip of the tongue touching lightly against the roof of the mouth. The Japanese /r/ sound varies more or less freely between these two sounds, i.e., sometimes the tongue touches the roof of the mouth and sometimes it doesn't. (That's why new students of Japanese may hear the sound sometimes as /r/ and sometimes as /l/.) However, unlike English /r/, there is no lip rounding with Japanese /r/. The lips should stay relaxed. (When Japanese comedians imitate foreigners speaking Japanese, they generally include a lot of heavily rounded /r/ sounds and /r/-colored vowels, because those are the features that sound most distinctively non-Japanese to them.)

/HU/ = /FU/

When the phoneme /h/ comes before /u/, it is pronounced as a bilabial fricative. What this means is that the lips are brought close enough together to create a light hissing sound. Unlike English /f/, however, the lower lip does not touch the upper teeth.

VOWELS

Most English vowels are diphthongs, i.e., the tongue and/or lips move as the vowel is being pronounced. In American English, for example, the vowel sound in the word 'cake' starts as the mid front vowel /e/ and glides to near the high front vowel /i/. The vowels in the word 'two' and 'boat' have more lip rounding at the end than the beginning. In Japanese, however, vowel quality remains nearly constant from the beginning to the end of a vowel. Heavily diphthongized vowels are a clear indicator of a foreign accent in Japanese. As described in the Sounds section, Japanese vowels often become voiceless in certain contexts. This means that, to a foreign ear, the vowel seems to be missing: doko desu ka sounds like doko des ka arimasusounds like arimas chikakusounds like chkaku shita sounds like shta This does not happen all the time, though. If a speaker is emphasizing a particular word or speaking slowly, then these vowels might remain voiced. Sentence-final vowels seem to be voiced more often in women's speech than in men's.

ACCENT

As described in Accent (Pronunciation), Japanese accent is based on pitches, not stress. An English speaker learning Japanese is apt to put too much emphasis on certain syllables. To Japanese ears, this sounds bouncy and sing-songy. Many native Japanese teachers do not emphasize accent when they teach Japanese pronunciation, probably because there is a lot of regional variation. In particular, many words have opposite accent patterns in Kanto and Kansai. For the foreign student, however, it is better to try to acquire some kind of native accent (even if it isn't standard NHK style) rather than to retain a strong foreign accent.

Click here to go to the index for the FAQ.

Rafael Santos (santos@mickey.ai.kyutech.ac.jp) Last Update: April 17, 1995