| Classical Japanese | |
| Nativename: | 文語 |
| Also Known As: | Late Old Japanese |
| States: | Japan |
| Era: | Heian period, Shōwa period |
| Familycolor: | Altaic |
| Fam1: | Japonic |
| Fam2: | Old Japanese |
| Fam3: | Early Middle Japanese |
| Script: | Kanji (Kyūjitai), Kana (Man'yōgana, Hiragana, and Katakana; using Historical kana orthography) |
The classical Japanese language (bungo, "literary language"), also called and sometimes simply called "Medieval Japanese", is the literary form of the Japanese language that was the standard until the early Shōwa period (1926–1989). It is based on Early Middle Japanese, the language as spoken during the Heian period (794–1185), but exhibits some later influences. Its use started to decline during the late Meiji period (1868–1912) when novelists started writing their works in the spoken form. Eventually, the spoken style came into widespread use, including in major newspapers, but many official documents were still written in the old style. After the end of World War II, most documents switched to the spoken style, although the classical style continues to be used in traditional genres, such as haiku and waka. Old laws are also left in the classical style unless fully revised.
The terms Japanese: 文語 (bungo, "written language") and Japanese: 口語 (kōgo, "spoken language") are still used for classical and modern Japanese,[1] respectively. Their literal meanings are only historical, as classical Japanese is no longer used, while modern Japanese is the only current written language, despite its name. These terms are often used in descriptions of grammar to distinguish classical and modern inflections. For example, the bungo inflection of the verb 書く (kaku, "to write") is quadrigrade (kaka, kaki, kaku, kake), but its kōgo inflection is quintigrade due to a historical sound change (kaka, kakō < kakau < kakamu, kaki, kaku, kake).
Classical Japanese began to be written during the Heian period, at which point it was very similar to spoken Japanese. It became the written standard for the Japanese language for many centuries, though the spoken language continued to evolve and by the Edo period was substantially different from classical Japanese.[2] This is known as diglossia, a situation in which two forms of a language, in this case a written and spoken form, coexist.[3] During the Meiji period, some intellectuals sought the abolition of classical Japanese, such as the Genbun Itchi movement, which proposed that written Japanese conform to the vernacular spoken language. Futabatei Shimei's 1887 novel The Drifting Cloud was one of the first novels to be written in vernacular Japanese rather than classical. By 1908, novels no longer used classical Japanese, and by the 1920s the same was true of all newspapers.[4] Government documents remained in classical Japanese until 1946.[5] Classical Japanese continues to be taught in Japanese high schools and universities due to its importance in the study of traditional Japanese literature.[2]
Classical Japanese is written in an orthography that differs from modern Japanese in two major ways. These are the usage of old character forms (kyūjitai) and historical kana usage (rekishi-teki kana-zukai).
Old character forms are the forms of Chinese characters (kanji) used in Japan before the post-World War II spelling reforms there. The modern, simplified characters are called .
A few examples follow, with the old characters on the left and the new characters on the right:
The kana spelling of a kanji is not unique; e.g. in modern Japanese, note the difference in the reading of between ("physical body") and ("forms of government"). The above spelling differences are etymological. For example, is just a native Japanese word labeled by a Chinese character with similar meaning, while is totally a new word derived from the combination of original meanings of two Chinese characters (means "politics" and means "body").
In cases like those of the first two, the entire original character has essentially been replaced by a new one, independent of the original's etymology. This type, however, is relatively rare. Another approach is to essentially replace the character with a piece of it, sometimes slightly altered, as in the third and fourth characters. Finally, probably the most common type of simplification is to change one component of the character to reduce the number of strokes and/or make it easier to write, a strategy exemplified by the fifth and sixth examples. Note that, as in the case of the sixth character, the simplification may be very subtle.
In general, old character forms are identical to their traditional Chinese counterparts, but there are some exceptions. For the seventh example character, the traditional and simplified Japanese versions coexisted as different forms of the same traditional character in Modern Chinese, while in Japan, what is now the new character form was at that time considered a variant and rarely used. And in a few cases, like that of the eighth character, the old character form has always been considered a rare variant in Modern Chinese. (However, and are actually the formal forms in Middle Chinese and Old Chinese.)
Historical kana usage is the system of kana (i.e., phonetic character) writing used in Japan before the post-war reforms. More specifically, it is the version of kana orthography standardized in the Meiji Period (since before that time kana usage was not standardized). It is, broadly speaking, based on the pronunciation of Japanese in the Heian Period, the time-frame in which Early Middle Japanese (on which the grammar of classical Japanese is based) was spoken. There are several differences between historical kana usage—which is also referred to as —and the modern kana orthography, called or . Some of these differences apply primarily to Sino-Japanese readings of Chinese characters, while others apply primarily to native Japanese words, and still others apply equally to both groups of words.
Broadly speaking, the differences are:
Some examples follow (old spellings are on the left, new spellings on the right; kana in parentheses represent the pronunciation of the preceding character):[6] [7] [8] [9]
There are some exceptions to this sound change, although they are rare. They include (haha "mother", expected form hawa), (hoho "cheek", expected form hō), (ahiru "domestic duck", expected form airu), and (afure-ru "overflow", expected form aore-ru or ōre-ru. Sometimes, as in the case of the first two exceptions, the sound change form exists, usually with a slightly different meaning (hawa is a hyper-formal and very respectful term for mother) or is used in different contexts (hō is generally used in isolation, while hoho is generally used in compounds). In other cases, as is true of the second two exceptions, the unchanged form is the only one that exists. In addition to these exceptions, some dialects may preserve these sounds as they were at any stage of the language.
This section uses Nihon-shiki romanization for,, and .
Some examples:
Native Japanese words
Sino-Japanese words
There are no known exceptions (besides the aforementioned ones regarding wo) in standard Japanese, and no dialects preserve the distinction between pronounced as //wi// and pronounced as //i//, pronounced as //we// and pronounced as //e//, and/or pronounced as //wo// and pronounced as //o//, but some of the Ryukyuan languages (which are also descended from Proto-Japonic) do.
This section uses Nihon-shiki romanization for,,, .
Some examples:
Native words
Sino-Japanese words
Loanwords
There are no known exceptions in standard Japanese pronunciation, although there are many dialects (such as the Tosa dialect) that preserve the distinction between historical pronounced as //z// and pronounced as //d// in speech, usually by using pronounced as //ʑi// and pronounced as //zu// for historical pronounced as //z// and pronounced as //d͡ʑi// and pronounced as //d͡zu// for historical pronounced as //d// (see Yotsugana). In writing, the distinction is preserved in single morphemes in cases where a sequence (chidi) or (tsudu) was historically produced by rendaku (such as in chidim-u, "shorten", and tsuduk-u, "continue", pronounced as if chizim-u and tsuzuk-u, respectively), or in compounds where a phonemic pronounced as //ti// or pronounced as //tu// has been voiced to pronounced as //zi// or pronounced as //zu// (such as in mi-dika "one's surroundings" and kana-dukai "kana usage", pronounced as if mi-zika and kana-zukai, respectively). This usage is a holdover from this rule.
In modern Japanese, the small kana,, and (ya, yu, and yo) are used to indicate palatalized consonants (Yōon) when following an I-column (I-dan) kana of the K-, G-, N-, B-, P-, M-, or R-rows (; Ka-, Ga-, Na-, Ba-, Pa-, Ma-, Ra-gyō). For example:
When a small Y-row (Ya-gyō) kana follows an I-column kana of the S-, Z-, T-, D-, or H-rows (; Sa-, Za-, Ta-, Da-, Ha-gyō), the preceding consonant is changed:
These three kana cannot follow A-row (A-gyō) or W-row (Wa-gyō) kana in this way.
In historical kana, all of these examples are written with large kana,, and (ya, yu, and yo). So the previous examples would be written:
This is the only historical kana rule that does not reflect a historical pronunciation. It is also one of only two rules (along with the geminate rule) that create ambiguity for the reader (excluding the exceptions listed above for the H-row rule). For instance, the aforementioned word (kyaku) is not differentiated in historical kana from the word (kiyaku "agreement") when written in historical kana: both are written (kiyaku).
The other use of small kana in modern Japanese is in the geminate consonant mark (Sokuon),, which is a small version of (tsu). In native Japanese words, this symbol can be used before kana of the K-, S-, T-, and P-rows. For example,
Voiced geminates are generally prohibited by Japanese phonological rules, but they occur in a few loanwords (although they are sometimes pronounced by native speakers as if they were their voiceless counterparts). For example:
Kana of the N- and M-rows can also be geminate, but they are preceded by (n) to indicate gemination instead.
Gemination can occur in Japanese for a variety of reasons. In native words, it occurs either when a historical long vowel elides, as in the aforementioned (massugu, originally maasugu), or randomly, as in the aforementioned (kitto, originally kito). These examples of the geminate consonant marker, along with those found in loanwords, are written with large (tsu) in historical kana. Therefore,
In these cases, the historical usage is not reflecting any historical pronunciation. However, in Sino-Japanese words, geminate consonants are produced by different, more regular processes, and the historical usage for these words reflects historical pronunciations.
The most common way for geminates to be produced in Sino-Japanese words is by the elision of a vowel from the kana,,, or (ki, ku, chi, or tsu). For example:
In historical kana, where the geminate mark is used in the first, second, and fourth examples, a full-sized version of the original kana is used. However, in the third example, (tsu) is used, even though an pronounced as //i// has been elided. The reason for this is that in Early Middle Japanese, when these sounds were borrowed from Middle Chinese, the Japanese language acquired a final pronounced as //t// in the Sino-Japanese morphemes that currently end in (chi, pronounced as //ti//) or (tsu, pronounced as //tu//). Later on, these acquired two forms, one with pronounced as //i// and one with pronounced as //u// (although in syllables beginning with pronounced as //ni//, one form usually begins with pronounced as //zi//, as is the case with). So the semantic difference between Sino-Japanese syllables ending in pronounced as //ti// or pronounced as //tu// is almost always trivial, and the historical pronunciation was identical, so they were not distinguished in writing. Therefore, the previous examples would be written:
Occasionally, gemination may also result from a loss of a vowel after (fu, originally pronounced as //pu//). These cases are complicated by the H-row rule, and perhaps because of that, are also written with in historical kana. For example,
is written
in historical kana.
While this usage does reflect a historical pronunciation, it, like the Y-row rule, produces ambiguity. Furthermore, since these vowels are elided in some compounds but not others, this usage obscures the difference in a way that is essentially impossible to predict.
While there are a few other processes that can cause geminates in Sino-Japanese words, they all apply to N- and M-row kana, and are not written differently in historical and modern kana.
Starting in Early Middle Japanese, as more and more Chinese characters were borrowed into Japanese, the language acquired consonants fronted with glides. Those fronted with the palatal glide are described in the Y-row rule, but Early Middle Japanese also introduced consonants fronted with labial glides (i.e., CwV). These were far more limited in range than their palatal counterparts, however, affecting only the K- and G- rows. instead of pronounced as //a//, pronounced as //u//, and pronounced as //o// for the vowels of onset, like the palatal glides, the vowels of onset for the labial glides were pronounced as //a//, pronounced as //i//, and pronounced as //e//, and used the kana,, and (wa, wi, and we). Finally, while the palatal glides are written with an I-column kana, the labial glides are written with a U-column (U-dan) kana. However, when historical kana was standardized in the Meiji Period, only the syllables with historical pronounced as //wa// were indicated. Nevertheless, some classical texts may indicate the other differences, and some resources will refer to them, so it is useful to be familiar with them. This rule applies exclusively to Sino-Japanese words. Some examples:
(written kuwa, but pronounced kwa) and (written guwa, but pronounced gwa) (indicated in standard historical kana)
(written kuwi, but pronounced kwi), (written guwi, but pronounced gwi), (written kuwe, but pronounced kwe), and (written guwe but pronounced gwe) (not indicated in standard historical kana)
Labialized consonants sometimes occur in modern loanwords, and they are generally dealt with in one of two ways. Firstly, the labialized consonant may be changed from a sequence pronounced as //CwV// to a sequence pronounced as //CuwV//, both in writing and in speech. For example,
In other cases, they may be indicated with a U-column kana followed by a small A-row kana, indicating a labialized consonant. For example,
However, in these cases, an alternative version with large A-row kana generally exists (as it does in this case), indicating a monophthong pronunciation, and many speakers use the monophthong pronunciation regardless of how it is written.
There are no known exceptions to this rule, but some dialects (such as the Kagoshima dialect) preserve the distinction.
Modern Japanese has the moraic nasal (n), which can represent a variety of sounds depending on what sounds come before and after it. Syllable final nasals are believed by many scholars to have existed in Proto-Japonic, but all agree that they were lost by the time of Old Japanese. They first re-appeared in Early Middle Japanese, with the introduction of Middle Chinese loanwords ending in -n and -m. Therefore, the majority of occurrences of (n) in modern Japanese occur in Sino-Japanese vocabulary. Originally, syllabic n and m were phonemically and phonologically distinct, although the distinction was never written down, and was lost by Early Modern Japanese. For example,
However, some native Japanese words also have (n). This happens exceedingly rarely, and usually results from sound elision. An exhaustive list of every example out of all regular-use characters with the syllabic nasal in their native Japanese readings numbers only 13 characters (0.61% of the regular-use set) giving rise to 14 readings. They are
From the elision of a vowel following /m/ or /n/
From the elision of a full mora
From the preservation of an Old Japanese pre-nasalized consonant in a modern Japanese word
From abbreviation of another pronunciation on this list
From multiple processes
From some semantic (rather than phonetic) process
Of course, there are also some words with this sound that either lack Chinese characters or were coined in the modern or Early Modern Japanese eras, when (n) had been fully incorporated into the language. For example,
Regardless of how it came to be, the Japanese orthography lacked the character (n) or any equivalent. Therefore, until the spelling reforms of 1900, (mu) was generally used to represent the syllabic nasal. Sometimes, this convention may be preserved by modern writers, but standard historical kana distinguishes (mu) from (n).
There is one exception. In classical Japanese, there is an auxiliary verb (jodōshi) (mu) which indicated the volitional. It, too, underwent vowel elision, and came to be pronounced as pronounced as //m// and then pronounced as //n//. However, the conventions of standard historical kana call for this auxiliary verb (and any word derived from it) to be written with (mu) even though they are pronounced as (n).
Since the auxiliary verb (mu) is non-existent in modern Japanese, there are no dialects that preserve the distinction expressed in this rule. However, some may preserve the distinction between final pronounced as //m// and pronounced as //n//.
Two other significant differences involve the way that kana are used in general, rather than which kana are used. The first is that Chinese characters in classical texts are often fully marked with ruby text (Furigana), especially in old laws and other very important documents. Ruby text is still widely used in modern Japanese, but only for characters with non-standard or ambiguous pronunciations, or sometimes in materials designed for children or foreigners. The second difference is that, especially in legal documents, Katakana were often used in the way that Hiragana are used in modern Japanese, to write out adjective and verb inflections, suffixes, and particles (Okurigana), and for the aforementioned ruby text.
Finally, kana iteration marks were far more common in classical Japanese, and sometimes used in ways that are considered completely obsolete in modern Japanese.
For an example of a major document written in the classical style, see as an example the original text of the 1890 Meiji Constitution, which is written in classical Japanese using historical kana, old character forms, kana iteration marks, and Katakana in place of Hiragana (although it lacks universal ruby text).
Classical Japanese has the following verb classes and stem forms:
Inflectional form = +
| Inflectional Class | Inflectional form | Translation | |||||||
|---|---|---|---|---|---|---|---|---|---|
| stem | Irrealis | Infinitive | Conclusive | Attributive | Realis | Imperative | |||
| Quadrigrade | (-i) | (-a) | (-i) | (-u) | (-e) | 'hear' | |||
| Upper Monograde | – | (-i) | (m-iru) | (-ire) | (-i[yo]) | 'see' | |||
| (-wi) | (-wiru) | (-wire) | (-wiyo) | 'use' | |||||
| Lower Monograde | – | (-e) | (-eru) | (-ere) | (-e[yo]) | 'kick' | |||
| Upper Bigrade | (-i) | (-u) | (-uru) | (-ure) | (-iyo) | 'pass' | |||
| Lower Bigrade | (-e) | (-u) | (-uru) | (-ure) | (-e[yo]) | 'receive' | |||
| K-irregular | – | (-o) | (-i) | (-u) | (-uru) | (-ure) | (-o) | 'come' | |
| S-irregular | – | (-e) | (-i) | (-u) | (-uru) | (-ure) | (-e[yo]) | 'do' | |
| (-se) | (-si) | (-su) | (-suru) | (-sure) | (-seyo) | 'set the date' | |||
| N-irregular | (-a) | (-i) | (-u) | (-uru) | (-ure) | (-e) | 'die' | ||
| R-irregular | (-a) | (-i) | (-u) | (-e) | 'be, exist' | ||||
The (yo) at the end of the imperative forms is optional in classical Japanese, although exceedingly common.
While the many conjugation classes may seem overwhelming, most of them contain few verbs. The quadrigrade and lower bigrade classes are the primary, containing about 75% and 20% of the verbs in the language, respectively. The upper bigrade class is small (about 56 non-compound verbs), but sizable enough to make an exhaustive list difficult. The other 6 classes all together contain between 22 and 28 verbs, depending on whether basic compound verbs are included or not. An exhaustive list of these follows, with verbs in the conclusive form, as is the most common standard. Chinese character pronunciations are indicated by hiragana in parentheses following the given character. The first spelling listed for a given verb is the most common, and those that follow are alternative spellings. Some of these spellings are generally used for slightly different connotations of the same verb, while others are simple alternatives. In later reference, only the first spelling (in pre-World War II orthography) will be used, and the transcription will be based on the historical spelling. A blank cell in one (or both) of the "modern" columns indicates that the modern spelling and/or transcription is the same as the pre-World War II version.
| Japanese (Pre-World War II orthography) | Japanese (Modern orthography) | Romanization (Pre-World War II orthography) | Romanization (Modern orthography) | Translation | |
|---|---|---|---|---|---|
| (Kami ichidan katsuyō dōshi "Upper monograde conjugation class verbs") | |||||
| Ki-ru | To wear | ||||
| Ni-ru | To resemble | ||||
| Ni-ru | To boil | ||||
| Hi-ru | To sneeze | ||||
| Hi-ru | To dry | ||||
| Hi-ru | To winnow | ||||
| Mi-ru | To go around | ||||
| Mi-ru | To see | ||||
| Kagami-ru | To learn from | ||||
| Kaherimi-ru | Kaerimi-ru | To reflect upon | |||
| Kokoromi-ru | To try | ||||
| I-ru | To shoot (an arrow) | ||||
| I-ru | To douse (with water) | ||||
| I-ru | To cast (metal) | ||||
| Wi-ru | I-ru | To sit | |||
| Wi-ru | I-ru | To carry (constantly) | |||
| Hikiwi-ru | Hikii-ru | To lead (an army) | |||
| Mochiwi-ru | Mochii-ru | To use | |||
| (Shimo ichidan katsuyō dōshi "Lower monograde conjugation class verbs") | |||||
| Ke-ru | To kick | ||||
| (Ka-gyō henkaku dōshi "K-irregular verbs") | |||||
| K-u | To come | ||||
| (Sa-gyō henkaku katsuyō dōshi "S-irregular conjugation class verbs") | |||||
| S-u | To do | ||||
| Ohas-u | Owas-u | To be/go/come (honorific form) | |||
| (Na-gyō henkaku katsuyō dōshi "N-irregular conjugation class verbs") | |||||
| In-u | To go away | ||||
| Shin-u | To die | ||||
| (Ra-gyō henkaku katsuyō dōshi "R-irregular conjugation class verbs") | |||||
| Ar-i | To exist | ||||
| Imasukar-i | To exist (honorific form) | ||||
| Haber-i | To serve (humble form) | ||||
| Wor-i | Or-i | To be | |||
Note that these translations are glosses, and may not reflect certain nuances or rare alternative meanings.
In addition, the translations are for the classical meaning of the verb, which may differ from the modern meaning of the verb if it has survived into modern Japanese either slightly (e.g., ki-ru, which meant "to wear [in general]" in classical Japanese, but means "to wear [from the waist up]" in modern Japanese), or significantly (e.g., wi-ru, which meant "to sit" in classical Japanese, but primarily means "to be" (for animate objects) in modern Japanese). Some may have the same meaning, but a different pronunciation (e.g., kagami-ru "to learn from", which is generally pronounced and written kangami-ru in modern Japanese). Also, even for those verbs which have survived with the same meaning and form, many are archaic and rarely used in modern Japanese (e.g., hi-ru "to sneeze", with the same modern meaning and form, but almost never used). On the other hand, some have kept the same meaning, form, and prominence into the modern language (e.g., mi-ru "to see", one of the oldest surviving verbs in the language and also one of the most common, both in classical and modern texts).
(imasukar-i "to exist", honorific form) has three pronunciation variants, each of which can use either Chinese character: (imasugar-i), (imasokar-i), and (imasogar-i).
Finally, the "modern" transcriptions are purely orthographic. For example, the modern version conclusive form of the classical verb (k-u "to come") is (k-uru), but the modern form is given in the table as (k-u), which is the way that a modern Japanese writer would write the classical Japanese word, rather than the way they would write the modern Japanese word.
Classical Japanese has the following classes of adjectives and stem forms:
| Class ofInflection | subclass | stem | Irrealis | Adverbial | Conclusive | Attributive | Realis | Imperative | Translation | |
|---|---|---|---|---|---|---|---|---|---|---|
| rowspan="2" | -ku | main | (/) | (-ku) | (-si) | (-ki) | (-kere) | 'be high' | ||
| -kari | (-kara) | (-kari) | (-karu) | (-kare) | ||||||
| rowspan="2" | -siku | main | (/) | (-siku) | (-si) | (-siki) | (-sikere) | 'be beautiful' | ||
| -kari | (-sikara) | (-sikari) | (-sikaru) | (-sikare) |
The existence of irrealis form is still a controversy. Some scholars assume that the ancient construction called (Ku-gohō "Ku-grammar") uses the irrealis form to form nouns from verbs and adjectives; e.g., (yasu-shi "peaceful") → (yasu-ke) + (-ku) → (yasukeku "peace of mind"). Meanwhile, others assumed the construction (-kuba) / (-shikuba) appears to be an irrealis form (-ku) / (-shiku) + particle (-ba) (since that particle usually attaches to the irrealis form). However, the scholars agreeing with "Ku-grammar theory" argue that it's actually (-ku) / (-shiku) + particle (ha; modern pronunciation wa) with a sequential voicing sound change from (ha) to (ba).
The compound forms are derived from continuitive form (-ku) / (-shiku) + (ar-i) → (-kuar-i) / (-shikuar-i), which then became (-kar-i) / (-shikar-i) by regular sound change rules from Old Japanese. The forms then follow the R-irregular conjugation type like (ar-i), but lack the conclusive form.
Similarly, the basic conjugations have no imperative form. When it is used, therefore, the (-kar-e) / (-shikar-e) forms are used. It is however, relatively rare, even in classical Japanese.
There are the following classes for adjectival verbs:
| Class ofinflection | stem | Irrealis | Adverbial | Conclusive | Attributive | Realis | Imperative | Translation |
|---|---|---|---|---|---|---|---|---|
| Nari | (-nara) | (-nari) | (-nari) | (-naru) | (-nare) | 'be static' | ||
| (-ni) | ||||||||
| Tari | (-tara) | (-tari) | (-tari) | (-taru) | (-tare) | ''be quiet, soft" | ||
| (-to) | ||||||||
Adjectival verbs are essentially nouns(or stems of the adjectives) combined with an auxiliary verb, either (-nar-i) or (-tar-i).
Most tari-adjectival nouns are derived from Sino-Japanese vocabulary. For example,「」 is derived from 「Chinese: 悄然」, a Chinese word meaning “quietly, softly”.
The auxiliary verbs are derived from directional particles (ni) + (-ar-i) and (to) + (-ar-i), respectively, yielding (niar-i) and (toar-i), respectively, which then lead to (nar-i) and (tar-i), respectively, by regular sound change rules. They therefore follow the R-irregular conjugation like (ar-i).
As with adjectives, the imperative form is rare, but is used.
The particle is omitted more often than in the spoken style.