Old Japanese language

Old Japanese
Period	Nara period (710-794)
Formerly spoken in	Japan
Linguistic classification	isolated language (controversial) Altaic (controversial) Japanese languages Old Japanese
Language codes
ISO 639-3	ojp

Old Japanese ( Japanese 上代 日本語 jōdai nihongo or 上古 日本語 jōko nihongo ) is the oldest written form of the Japanese language .

Dating

An exact dating of the period is difficult. The upper limit for this period is 794 when the capital was moved from Heijō-kyō to Heian-kyō . The lower one is unclear. Although wooden tablets ( mokkan ) and relics with fragments of writing were unearthed, the first existing text of greater length is the Kojiki from 712. Without necessarily excluding the small text fragments, this date is discussed as the lower limit for practical reasons. This happens to go well with the Nara period (710–794). It was replaced by classic Japanese .

Writing system

The oldest written sources are the Kojiki (712), Fudoki (720), Nihonshoki (720) and Man'yōshū (after 771).

The earliest texts found in Japan are in Classical Chinese , although they may have been read in Japanese using the Kanbun method. Some of these texts show influences from Japanese grammar, e.g. B. that the verb was placed after the object. In these “hybrid” texts, Chinese characters were occasionally used as phonetic substitutes for the Japanese particles. In the course of time, this phonetic use of the Chinese characters became more and more common, until the Man'yōgana emerged. This system was already used in the non-prosaic passages of Kojiki before it was used in a mature form in the eponymous Man'yōshū .

Phonology

Research into the phonology of Ancient Japanese is based on comparative studies of the pronunciation of Chinese of the time, the Ryūkyū languages, and the reverse analysis of the diachronic change in Japanese pronunciation. Although the majority of the writings of that period represent the language of the Nara court in central Japan, some poems in Man'yōshū are written in other dialects from southern and eastern Japan. Some of these dialect differences can still be found today.

Old Japanese is phonetically different from later language levels. Analysis of the Man'yōgana revealed a special system known as Jōdai Tokushu Kanazukai . This is discussed in more detail in the phoneme section.

The transcriptions of the ancient Japanese words in Kojiki differ from those in Nihonshoki and Man'yōshū , as the former, in contrast to the latter, distinguishes the syllables / mo ₁ / and / mo ₂ /. This is associated with the fact that the historical records of the Kojiki were compiled earlier than those in the Nihonshoki , whereby this earlier distinction, which disappeared shortly afterwards, was preserved.

The modern Japanese syllable [ tsu ] is derived from the affriction of the [ t ] before [ u ] in old Japanese [ tu ] and the modern [ zu ] also from the early modern [ dzu ] and old [ du ]. Certain modern dialects retain this distinction between [ z ] and [ dz ], e.g. B. [ midzu ] for mizu (water) in the Nagoya dialect. A similar process of palatalization resulted in the modern [ tɕi ] from classical and old [ ti ]. However, it cannot be proven when / ti / was palatalized. However, it could have been an affriction in Old Japanese.

Other characteristics that distinguish ancient from modern Japanese are:

• No long vowels or diphthongs .
• Words do not begin with / r / or voiced plosives .
• No ending syllable consonants of any kind.

Some scholars point to a connection between Ancient Japanese and extinct languages ​​of the Korean Peninsula , such as the Gaya language. However, evidence of the relationship between Japanese and languages ​​other than Ryukuyu has not yet been established (see also Japanese Ryūkyū for further details).

Phonemes

Originally it was assumed that Old Japanese had 68 syllables. At the beginning of the 20th century, Shinkichi Hashimoto discovered that in Man'yōshū only a subset of the set of possible characters for certain syllables was used in individual words and a different subset in other words. He attributed this to the fact that in the Heian period, pairs of syllables collapsed into a single syllable.

Further research finally found 88 syllables:

a	i		u	e		O
ka	ki 1	ki 2	ku	ke 1	ke 2	ko 1	ko 2
ga	gi 1	gi 2	gu	ge 1	ge 2	go 1	go 2
sa	si		see below	se		so 1	so 2
za	zi		to	ze		zo 1	zo 2
ta	ti		do	te		to 1	to 2
there	di		you	de		do 1	do 2
n / A	ni		nu	no		no 1	no 2
Ha	hi 1	hi 2	hu	hey 1	hey 2	ho
ba	to 1	to 2	bu	be 1	be 2	bo
ma	wed 1	wed 2	must	me 1	me 2	mo 1	mo 2
ya			yu	ye		yo 1	yo 2
ra	ri		ru	re		ro 1	ro 2
wa	wi			we		Where

For reasons explained in the Phonetics section , / p / is also written instead of / h / .

The CV ₁ syllables are referred to as Type A syllables ( 甲類 , Kō-rui ) and the CV ₂ syllables as Type B syllables ( 乙類 , Otsu-rui ). In Japanese linguistic sources, CV _甲 and CV _乙 are used instead of CV ₁ and CV ₂ .

Shortly after the Kojiki , the distinction between mo ₁ and mo ₂ quickly disappeared and the number of syllables was reduced to 87.

Many hypotheses have been proposed to explain the duplication of syllables, including:

• an 8-vowel system,
• Palatalization of the previous consonants,
• palatal and labial gliding sounds / y / or / w / and
• a 6-vowel system.

This matter is currently the subject of intense debate, so there is no general consensus.

transcription

When transcribing it should be noted that this does not necessarily support a hypothesis and that the subscript 1 or 2 can refer to the consonant or vowel.

There are several competing transcription systems. In some only the Type B syllables are marked, e.g. B. with a trema over the vowel: ï, ë, ö for i ₂ , e ₂ and o ₂ and i, e and o for i ₁ , e ₁ and o ₁ . The problem here is that a special pronunciation of the vowel is implied indirectly, as well as words in which the 1/2 difference is unclear, as / to / in / toru / or / kaditori / cannot be reliably differentiated.

Other transcription systems

Trema	a	e	ë	i	ï	O	ö	u
Trema, expanded	a	ê	ë	î	ï	O	ö	u
	a	e1	e2	i1	i2	o1	o2	u
Yale Romanization	a	ye	ey	yi	iy	Where	O	u

Phonological rules

The vowel types in a single morpheme had certain phonological restrictions:

• -o ₁ and -o ₂ do not appear together,
• -u and -o ₂ generally do not appear together and
• -a and -o ₂ generally do not appear together.

These rules indicate 2 vowel groups: / -a, -u, -o ₁ / and / o ₂ /. Vowels from one group do not mix with those of the other; -i ₁ and -i ₂ can appear with both groups. Some interpret this phenomenon as an indication of a vowel harmony of Old Japanese as it occurs in Altaic languages .

phonetics

Vowels

A phonetic description of the vowels depends on the hypothesis that is followed.

Consonants

according to Bjarke Frellesvig:

		labial	Alveo-dental		Palatal	Velar
Obstruents	Unvoiced	* p	* t	* s		* k
	Voiced, prenasal	* ᵐb	* ⁿd	* ⁿz		* ᵑg
Nasals		* m	* n
Approximants		* w			* j
Flap			* ɾ

Realization of / h /

/ h / was probably realized phonetically as * [ p ]. This assumption was predicted by the following textual and phonological analyzes. So the modern / h / causes a discrepancy in the pairing of the unvoiced versus the voiced consonants. For / k, g /, / s, z /, / t, d / and finally / h, b / the pair / h, b / does not match , since the unvoiced variant of / b / / p /. Comparisons with the Ryukyu languages ​​show [ p ] where [ h ] is spoken in modern Japanese . Since the two languages ​​separated at some point in the past, this can be taken as evidence that Japanese [ h ] was once spoken like Ryukyu [ p ].

In later language levels such as Classical Japanese , / h / was spoken as [ ɸ ], which is only preserved in modern Japanese with / hu /. For example, Portuguese missionaries who visited Japan in the early 17th century wrote the entire / h / kana series as "fa, fi, fu, fe, fo". Descriptions by Korean visitors in the same century suggest an unvoiced labial fricative . The oldest reference to [ ɸ ] comes from the 9th century. In 842 the monk Ennin describes in the Zaitōki that Sanskrit -p is more labial than Japanese. Hence, the general opinion is that between the 9th and 17th centuries / h / was spoken as [ ɸ ]. Some linguists assume that / h / was already realized as [ ɸ ] in Old Japanese .

It is also possible that / h / varied freely at the beginning of the word as a voiceless bilabial obstruent [ p ~ ɸ ] and inside the word as a voiced bilabial obstruent [ b ~ β ].

Prenasalization

The voiced sounds / b, d, z, g / were realized prenasally. In modern Japanese, traces of it have only survived for / g / with its allophone [ ŋ ].

* / yi /

It is unknown whether the phoneme * / yi / existed. The words kai 'oar' and kui 'regret' could indicate that Old Japanese avoided clusters of vowels and consonants, so one explanation for this is that it actually meant * kayi and * kuyi . The Chinese used to write Old Japanese could not distinguish the sounds [i] and * [ji], so that this distinction, if it existed in spoken Japanese, was not reflected in written Japanese.

Syllable structure

Traditionally, the difference between type A and type B syllables is seen in the vowel quality and thus 8 different vowel phonemes with different theories for the pronunciation of these vowels are assumed. The syllable structure is thus CV (consonant vowel).

Other researchers see the difference in the syllables themselves through the use of gliding sounds and thus a syllable structure in the form CGV.

The introductory C could be empty. A mere vowel only occurs at the beginning of a word. Voiced consonants and / r / do not appear at the beginning of a word, except for the two loan words / rikizimahi ₁ / and / rokuro /.

Vowel elisions took place to prevent vowel clusters :

• the introductory vowel is omitted: / ara / + / umi ₁ / → / arumi ₁ /,
• The following vowel is omitted: / hanare / + / iso ₁ / → / hanareso ₁ /,
• two consecutive vowels merge into a new one: i ₁ + a → e ₁ , a + i ₁ → e ₂ , o ₂ + i ₁ → i ₂ or
• / s / is inserted between two vowels: / haru / + / ame ₂ / → / harusame ₂ /. However, it is possible that / ame ₂ / was earlier * / same ₂ /.

grammar

pronoun

Personal and demonstrative pronouns were:

	short	long	Local	Directional
Personal pronouns
1st person	wa, a	goods, are
2nd person 1)	n / A	nare 2)
3rd person	si	-
Interrogative	ta	tare
Reflexive	ono 2	(ono 2 re) 2)
Demonstrative pronouns
Proximal	ko 2	ko 2 re	ko 2 ko 2	ko 2 ti
Non-proximal	so 2	- 2)	so 2 ko 2	-
Distal 3)	ka	kare
Interrogative	idu-	idure	iduku	iduti

The difference in function between short and long pronouns is not known. However, the short pronouns are always followed by a particle. With minor exceptions, the long pronouns are never used with genitive particles or in nominal compounds.

1. In addition to the personal pronouns, there are other forms of address for the 2nd person, some of which, in contrast to the pronouns, can be provided with a plural suffix:
   • imasi , masi , mimasi : cf. with imasu ("to be, to be present")
   • kimi : "she (my master)"
   • namuti : from the personal pronoun na and the honorary suffix muti
   • namutati : "you"; from namuti and the plural suffix tati
   • wake , ore : derogatory
   • i : derogatory, but only in connection with genitive as i-ga
2. nare rarely occurs, ono ₂ re only twice in Man'yōshū and so ₂ re not at all.
3. Frellesvig points out that in Old Japanese kare only appear once and ka only appear at least twice in Man'yōshū. Since they cannot be regarded as productive members, the Old Japanese had only two degrees of distance in contrast to the successor, the Classical Japanese, and the predecessor, the Proto-Japanese, with their three degrees of distance.

Verbs

Old Japanese differentiated between 8 verb conjugations: four-stage ( 四段 yondan ), upper one-stage ( 上 一段 kami ichidan ), upper two-stage ( 上 二段 kami nidan ), lower two-stage ( 下 二段 shimo nidan ), K-irregular ( カ 変 ka- hen ), S-irregular ( サ 変 sa-hen ), N-irregular ( ナ 変 na-hen ) and R-irregular ( ラ 変 ra-hen ). Lower one-level verbs ( 下 一段 shimo ichidan ) are not yet available.

conjugation

Verb class	Mizenkei 未然 形 unrealis form	Ren'yōkei 連用 形 conjunctive form	Shūshikei 終止 形 final form	Rentaikei 連 体形 attributive form	Izenkei 已然 形 Realisform	Meireikei 命令 形 imperative form
four-stage	-a	-i 1	-u	-u	-e 2	-e 1
upper single stage	-	-	-ru	-ru	-re	- (yo 2 )
upper two-stage	-i 2	-i 2	-u	-uru	-ure	-i 2 (yo 2 )
lower two-stage	-e 2	-e 2	-u	-uru	-ure	-e 2 (yo 2 )
K-irregular	-o 2	-i 1	-u	-uru	-ure	-o 2
S-irregular	-e	-i	-u	-uru	-ure	-e (yo 2 )
N-irregular	-a	-i	-u	-uru	-ure	-e
R-irregular	-a	-i	-i	-u	-e	-e

Mizenkei ( Irrealis ) and Izenkei (Realis) are named as contrast pairs after the function of the corresponding verb form with the particle ba : with Mizenkei + ba a hypothesis or conditional is expressed and with Izenkei + ba a real condition.

Thematic and athematic stems

Verbs with a stem ending in a consonant are called "athematic". These follow a four-step, upper two-step, S-, R-, K- or N-irregular conjugation.

Verbs with a stem ending in a vowel are called "thematic". These follow a one-step conjugation above.

Irregular verbs

There are some verbs with irregular conjugations.

• K-irregular: k- "come"
• S-irregular: s- "do"
• N-irregular: sin- "die", in- "go, die"
• R-irregular: ar- "to be", wor- "to be"

The conjugation classes are named after the last stem consonant.

Adjectives

There were 2 types of adjectives: regular adjectives and adjectival nouns .

The regular adjectives are divided into two types: those with the Renyōkei ending in -ku , and those with the -siku ending. There are two types of inflection:

Adjective class	Mizenkei	Renyōkei	Shushikei	Rentaikei	Izenkei	Meireikei
-ku	-ke 1	-ku	-si	-ki 1	-ke 1 or -ke 1 re
	-kara	-kari	-si	-karu	-kare	-kare
-siku	-sike 1	-siku	-si	-siki 1	-sike 1 or -sike 1 re
	-sikara	-sikari	-si	-sikaru	-sikare	-sikare

The -kar- and -sikar- forms are derived from the verb ar- ("to be"). The Renyōkei inflection ( -ku or -siku ) gets ar- as a suffix. The inflection follows the R-irregular conjugation of this. Since the Old Japanese avoids vowel groups , the resulting -ua- changes to -a- .

The adjectival nouns have only one inflection:

tribe	Mizenkei	Renyōkei	Shushikei	Rentaikei	Izenkei	Meireikei
Adjective noun	-nara	-nari	-nari	-naru	-nare	-nare

Particles ( joshi )

Old Japanese knew the following particles:

Kaku joshi ( 格 助詞 )

Kaku joshi are particles that indicate the case of nouns.

• Nominative : The nominative particle i was already out of date and fell out of use according to the old Japanese.
• Accusative : wo (changes to wo-ba after the topic particle ha )
• Genitive : Mainly ga and no _{2 were} used for the genitive . ga is mainly used possessively and no _{2 is used as an} attribute, but both are also used appositionally, with ga rather emphasizing the previous part and no ₂ the following part. This means that with personal pronouns only ga (e.g. wa-ga , “my”; na-ga , “your”; si-ga , “to be”, ta-ga , “whose”) and with demonstrative pronouns only no ₂ ( ko ₂ -no ₂ , "this"; so ₂ -no ₂ , "that") is used. In subordinate clauses, both are also used to mark the subject. The genitive particle tu is beginning to become uncommon in Old Japanese, as it mainly occurs only after location.
  Genitive particles that are already out of date are ro ₁ , ro ₂ , na - the latter may also be a dual particle that has fallen out of use - and da , which in turn is an outdated variant of na . They are available e.g. B. also in the following expressions: kamu-ro ₁ -ki ("spiritual being [genitive] male"), kamu-ro ₁ -mi ("spiritual being [genitive] female"), o ₁ -ro ₂ -ti ("large [Genitive] wild, mighty being "), ma-na-ko ₁ (" eye [genitive]? Child; eyeball "), ke-da-mono (" hair [genitive] being ").
• Dative : ni
• Allative : The noun he ₁ ("side, direction") was first used as an allative particle.
• Ablative : ablative particles were equally ranked yo ₁ ri , yo ₁ , yuri and yu . The noun kara was first used as an ablative particle.
• Comitive : to ₂

Kakari joshi ( 係 助詞 )

• Subject : ha was used to indicate a normal or contrasting subject and mo was used for emphasis.
• Affirmation: namo ( namu ), so ( zo ), koso
• W-questions : ka
• Yes / no questions: ya

Fuku joshi ( 副 助詞 )

Fuku joshi form adverbial phrases.

• bakari : "about, approximately" (from hakar- , "measure; plan")
• dani : "at least"
• made (ni) : "to"
• nomi ₂ : “only, only, only; completely, completely, completely "
• saw : "also, moreover, even, at least" (cf. the modern sae )
• si : "also, even"
• simo : "even"
• sura : "even, at least"

Setsuzoku joshi ( 接 続 助詞 )

Setsuzoku joshi follow finite verb forms to form subordinate clauses.

• gane : "so that"
• gani : "as if"
• mono where : "although"
• mono kara (ni) , mono yuwe (ni) : "as, while"
• ni : "as, because"
• to :
  • Concessive clause (“even if, though”), often followed by mo → tomo
  • "So, for the purpose"
• where : "there, because"
• yuwe (ni) : because (after nouns also "because")

Shūjoshi ( 終 助詞 )

Shūjoshi are sentence end particles.

• gane : Opt
• kamo : interrogative; exclamation
• moga : Desiderative
• miyu : "it seems". Miyu comes from the final form of miye- , the passive of the verb mi- ("see").
• na : prohibition (“not”); exclamation
• so , koso , ka , ya : see Kakari joshi

Kantō joshi ( 間 投 助詞 )

Kantō joshi form interjections .

• ro : Only rarely occurs in the central but often in the eastern dialect.
• we where, ya, yo

The suffixes for the Meireikei probably derive from the interjections yo (central dialect) and ro (eastern dialect).

auxiliary verbs

The inflexible are also considered particles.

Relation to Mizenkei

tribe	Mizenkei	Renyōkei	Shushikei	Rentaikei	Izenkei	Meireikei	function
ba	not flexible						Conditionalis
H-	-a	-i	-u	-u	-e	-e	Repetition
ray-	-e						Potentials
r- / y-	-e	-e	-u	-uru	-ure	-e (yo)	1) Passive, 2) Potentialis, 3) Spontaneity
s-	-a	-i	-u	-u	-e	-e	Courtesy form
s-	-e	-e	-u	-uru	-ure	-e (yo)	1
sim-	-e	-e	-u	-uru	-ure	-e (yo)	Causative 2
to	not flexible						negation

1. Formed transitive and causative verbs and also served as a light politeness form.
2. Later use as a form of courtesy not yet available.

Relation to Renyōkei

tribe	Mizenkei	Renyōkei	Shushikei	Rentaikei	Izenkei	Meireikei	function
ke 1 r-	-a		-i	-u	-e		modal past
-	se		ki	si	sika		direct past
masiz-			-i	-iki			negative assumption 1
tar-	-a	-i	-i	-u	-e	-e	Perfect (derived from -te + ar- ("to be"))
te	not flexible						Te form (adverbial participle)
tutu	not flexible						simultaneous action ("during")

1. Develops into later maz- .

Relation to Shūshikei

tribe	Mizenkei	Renyōkei	Shushikei	Rentaikei	Izenkei	Meireikei	function
mer-		-i	-i	-u	-e		strong assumption
nar-			-i	-u	-e		Hearsay
ras-			-i	-iki	-ikere		conjecture

Connection with Rentaikei

tribe	Mizenkei	Renyōkei	Shushikei	Rentaikei	Izenkei	Meireikei	function
nar-	-a	-i	-i	-u	-e	-e	copula

Relation to Izenkei

• ba: marks a condition or a reason (see also Mizenkei-ba; not inflected)
• do: concessive (not inflected)

Relation to Meireikei

tribe	Mizenkei	Renyōkei	Shushikei	Rentaikei	Izenkei	Meireikei	function
r-	-a	-i	-i	-u	-e	-e	Perfect

Dialects

The Man'yōshū contains poems written in an Eastern dialect.

Proto-Japanese

4 vowel system

The following shortenings took place:

• * ia> / e ₁ /
• * ai ₁ > / e ₂ /
• * ui ₁ > / i ₂ /
• * o ₂ i ₁ > / i ₂ /
• * au> / o ₁ /
• * ua> / o ₁ /

With this, the proto-vowel system can be reconstructed as / * a, * i, * u, * o ₂ /.

/ h / <* [ p ]

In Proto-Japanese, / h / was very likely spoken * [ p ].

Co ₁ as Cwo

According to distribution, there may once have been * ho ₁ , * ho ₂ and * bo ₁ , bo ₂ . The distinction between / mo ₁ / and / mo ₂ / occurs only in Kojiki and then disappeared. If this is true, then Co ₁ and Co _{2 were distinguished} for all combinations except / where /. This is seen by some as a supporting argument that Co ₁ stands for Cwo.

Sources and individual references

swell

• Hisataka Omodaka: Jidaibetsu Kokugo Daijiten: Jōdaihen. Sanseidō ( 澤瀉 久 孝: 時代 別 国語 大 辞典: 上代 編. 三省 堂 ), 1967, ISBN 4-385-13237-2 .
• Akiho Yamaguchi, Hideo Suzuki, Ryūzō Sakanashi, Masayuki Tsukimoto: Nihongo no Rekishi. Tōkyō Daigaku Shuppankai ( 山口 明 穂, 鈴木 英 夫, 坂 梨 隆 三, 月 本 幸: 日本語 の 歴 史. 東京 大学 出版 会 ), 1997, ISBN 4-13-082004-4 .
• Susumu Ōno: Nihongo no Keisei. Iwanami Shoten ( 大野 晋: 日本語 の 形成. 岩 波 書店 ), 2000, ISBN 4-00-001758-6 .
• Samuel E. Martin: The Japanese Language Through Time. Yale University, 1987, ISBN 0-300-03729-5 .
• Marc Hideo Miyake: Old Japanese. A Phonetic Reconstruction. Routledge Shorton, London, New York 2000, ISBN 0-415-30575-6 .
• Masayoshi Shibatani: The Languages ​​of Japan. Cambridge University Press, 2000, ISBN 0-521-36918-5 .
• John R. Bentley: A Descriptive Grammar of Early Old Japanese Prose. Brill, 2001, ISBN 90-04-12308-3

Individual evidence

1. ↑ Haruhiko Kindaichi: The Japanese Language . Tuttle Publishing, 2011, ISBN 978-1-4629-0266-8 ( limited preview in Google Book Search).
2. ^ ^{A b} Bjarke Frellesvig: The Phonology of OJ. (PDF) In: Japanese Historical Linguistics. Retrieved December 16, 2017 . 
3. ^ John R. Bentley: Old Japanese . In: Nicolas Tranter (Ed.): The Languages ​​of Japan and Korea . Routledge, 2012, ISBN 978-0-415-46287-7 , 7.2.4 Morphophonology, pp. 193 ( limited preview in Google Book Search [accessed October 13, 2018]).
4. ↑ Bjarke Frellesvig: Old Japanese Pronouns. (PDF; 132 kB) In: Japanese Historical Linguistics. Retrieved May 25, 2008 . 
5. ^ Bjarke Frellesvig: Old Japanese Particles. (PDF; 169 kB) In: Japanese Historical Linguistics. Retrieved May 25, 2008 . 
6. ↑ Bruno Lewin : Outline of the Japanese grammar: On the basis of the classical written language . Harrassowitz, 1990, ISBN 3-447-02981-1 , pp. 75–76 ( limited preview in Google Book search).

Web links

• Documents on Old Japanese by Bjarke Frellesvig (English)