Outdated Japanese (上代日本語, Jōdai Nihon-go) is the oldest attested stage of the Japanese language. Attested in paperwork from the Nara period (eighth century), it turned Early Middle Japanese within the succeeding Heian period, however the exact separation of each languages is controversial.

Outdated Japanese was an early member of the Japonic household, however no conclusive hyperlinks to different language households have been proved.

Outdated Japanese was written utilizing Chinese characters through the use of an increasingly-standardized and phonetic type that ultimately advanced into man’yōgana. Usually for a Japonic language and for a step within the evolutionary line of contemporary Japanese, Outdated Japanese was a primarily-agglutinative language with a subject–object–verb phrase order. Nonetheless, Outdated Japanese was marked by a couple of phonemic variations from later varieties, corresponding to a less complicated syllable construction and distinctions between a number of pairs of syllables that might have been pronounced identically since Early Center Japanese. The phonetic realization of the differentiation is unsure.

Sources and relationship

Rubbing of Bussokuseki-kahi poems carved c. 752

Outdated Japanese is often outlined because the language of the Nara period (710–794), when the capital was Heijō-kyō (now Nara).[1][2]
That’s the interval of the earliest linked texts in Japanese, the 112 songs included within the Kojiki (712).
The opposite main literary sources of the interval are the 128 songs included within the Nihon Shoki (720) and the Man’yōshū (c. 759), a compilation of over 4,500 poems.[3][4]
Shorter samples are 25 poems within the Fudoki (720) and the 21 poems of the Bussokuseki-kahi (c. 752).
The latter has the advantage of being an authentic inscription, however for all the opposite texts the oldest surviving manuscripts are the outcomes of centuries of copying, with the attendant danger of scribal errors.[5]
Prose texts are extra restricted however are thought to mirror the syntax of Outdated Japanese extra precisely than verse texts do. Crucial are the 27 Norito (liturgies) recorded within the Engishiki (compiled in 927) and the 62 Senmyō (imperial edicts) recorded within the Shoku Nihongi (797).[4][6]

A restricted variety of Japanese phrases, largely private names and place names, are recorded phonetically in historic Chinese language texts, such because the “Wei Zhi” portion of the Records of the Three Kingdoms (third century AD), however the transcriptions by Chinese language students are unreliable.[7]
The oldest surviving native inscriptions, relationship from the fifth or early sixth centuries, embody these on the Suda Hachiman Shrine Mirror, the Inariyama Sword, and the Eta Funayama Sword.
These inscriptions are written in Classical Chinese however comprise a number of Japanese names that had been transcribed phonetically through the use of Chinese language characters.[8][9]
Such inscriptions turned extra widespread from the Suiko period (592–628).[10]
These fragments are often thought-about a type of Outdated Japanese.[11]

Of the ten,000 paper information saved at Shōsōin, solely two are in Outdated Japanese and date from about 762.[12]
Over 150,000 picket tablets (mokkan) relationship from the late seventh and early eighth century have been unearthed.
The tablets bear quick texts, usually in Outdated Japanese and mirror a extra colloquial model than the polished poems and liturgies of the first corpus.[13]

Writing system

Artifacts inscribed with Chinese language characters dated as early as the first century AD have been present in Japan, however detailed data of the script appeared to not have arrived within the islands till the early fifth century.
Based on the Kojiki and Nihon Shoki, the script was introduced by students from Baekje (southwestern Korea).[14]
The earliest texts present in Japan had been written in Classical Chinese, in all probability by immigrant scribes.

Later “hybrid” texts present the affect of Japanese grammar, such because the word order (for instance, the verb being positioned after the thing).[15]

Chinese language and Koreans had lengthy used Chinese language characters to put in writing non-Chinese language phrases and correct names phonetically by choosing characters for Chinese language phrases that sounded much like every syllable.
Koreans additionally used the characters phonetically to put in writing Korean particles and inflections that had been added to Chinese language texts to assist studying them (Idu script).
In Japan, the follow was developed into man’yōgana, an entire script for the language that used Chinese language characters phonetically, and it was the ancestor of contemporary kana syllabaries.[16]
This method was already in use within the verse components of the Kojiki (712) and the Nihon Shoki (720).[17][18]

For instance, the primary line of the primary poem within the Kojiki was written with 5 characters:[19][20]

Middle Chinese[b]yaekjuwmawtatu
Outdated Japaneseya-kumo1tatu
‘many clouds rising’

This methodology of writing Japanese syllables through the use of characters for his or her Chinese language sounds (ongana) was supplemented with oblique strategies within the complicated combined script of the Man’yōshū (c. 759).[21][22][23]


In man’yōgana, every Outdated Japanese syllable was represented by a Chinese language character. Though any of a number of characters could possibly be used for a given syllable, a cautious evaluation reveals that 88 syllables had been distinguished within the Kojiki:[24][25]

Syllables within the Kojiki

The system has the identical gaps of yi and wu that had been present in later types of Japanese. Nonetheless, many syllables which have a contemporary i, e or o occurred in two varieties, termed varieties A (, ) and B (, otsu), denoted by subscripts 1 and a couple of respectively within the above desk.[24][26]
The syllables mo1 and mo2 aren’t distinguished within the slightly-later Nihon Shoki and Man’yōshū, decreasing the syllable depend to 87.[27][28]
All of these pairs had merged by the Early Middle Japanese of the Heian period.[29][30]


A number of totally different notations for the sort A/B distinction are discovered within the literature, together with:[31][32]

Widespread notations for the sort A/B distinction
listed notationi1i2e1e2o1o2
Kindaichi, Miller, Ōnoiïeëoö
modified Mathias–Millerîïêëôö
Yale (Martin)yiiyyeeywo
Frellesvig and Whitmaniwiyeewoo


There isn’t any consensus on the pronunciation of the syllables distinguished by man’yōgana.[33]
One problem is that the Middle Chinese pronunciations of the characters used are additionally disputed, and for the reason that reconstruction of their phonetic values is partly primarily based on later Sino-Japanese pronunciations, there’s a hazard of circular reasoning.[34]
Extra proof has been drawn from phonological typology, subsequent developments within the Japanese pronunciation, and the comparative research of the Ryukyuan languages.[35]


Outdated Japanese had open syllables of the shape (C)V topic to extra restrictions:

  • Phrases didn’t start with r or the voiced obstruents b, d, z, and g, apart from a couple of loanwords.[36]
  • A naked vowel didn’t happen aside from word-initially: vowel sequences weren’t permitted.[37]

In 1934, Arisaka Hideyo proposed a set of phonological restrictions permitted in a single morpheme. Arisaka’s Regulation states that -o2 was usually not present in the identical morpheme as -a, -o1 or -u.
Some students have interpreted that as a vestige of earlier vowel harmony, however it is rather totally different from patterns which might be noticed in, for instance, the Turkic languages.[38]


The Chinese language characters chosen to put in writing syllables with the Outdated Japanese vowel a counsel that it was an open unrounded vowel /a/.[39]
The vowel u was a close back rounded vowel /u/, not like the unrounded /ɯ/ of Trendy Normal Japanese.[40]

A number of hypotheses have been superior to elucidate the A/B distinctions made in man’yōgana. The difficulty is hotly debated, and there’s no consensus.[31]
The widely-accepted and conventional view, first superior by Kyōsuke Kindaichi in 1938, is that there have been eight pure vowels, with the sort B vowels being extra central than their kind A counterparts.[41]
Others, starting within the Nineteen Thirties however extra generally for the reason that work of Roland Lange in 1968, have attributed the sort A/B distinction to medial or last glides /j/ and /w/.[42][32]
The diphthong proposals are sometimes linked to hypotheses on pre-Outdated Japanese, however all exhibit an uneven distribution of glides.[32]

Examples of reconstructions of kind A/B distinctions[31]
iwiewewooKikusawa (1935)
iïeëoöKindaichi (1938), Miller (1967)
iïjeəjoəArisaka (1955)
jiijeeoɵHattori (1958)
jiijeewooLange (1968, 1973)
iwijeewooUnger (1977), Frellesvig and Whitman (2008)
iïeɛoɵŌno (1982)
iɨeəjoəMiyake (2003)

The excellence between mo1 and mo2 was seen solely in Kojiki and vanished afterwards.
Distributionally, there might have as soon as been *po1, *po2, *bo1 and *bo2.[28]
If that was true, a distinction was made between Co1 and Co2 for all consonants C aside from w. Some take that to help that Co1 might have represented Cwo.

Pre-Outdated Japanese

Most students derive the Outdated Japanese vowel system from an earlier four-vowel system, with the most typical Outdated Japanese vowels a, u, i1 and o2 reflecting earlier *a, *u, *i and *ə respectively.[38]Internal reconstruction means that the opposite, much less widespread, Outdated Japanese vowels had been derived from fusions of these vowels.[43]
For instance, the place identify take2ti is derived from a compound of taka- ‘excessive’ and iti ‘market’.[44][45]
One other piece of proof is that many nouns had totally different varieties, relying on whether or not they had been used independently or inside compounds: sake2 ‘rice wine’, which turned saka- in compounds corresponding to sakaduki ‘saké cup’.
In such instances, the certain type is taken into account fundamental, and the unbiased type could also be defined by postulating a suffix *-i that later fused with the ultimate vowel of the basis.[46][47]
The next reductions are proposed:

  • i2 < *ui: kami2/kamu– ‘god, spirit’,[46][47]mi2/mu– ‘physique’,[47][48]nagi2/nagu– ‘a peaceful’.[48]
  • i2 < *əi: ki2/ko2– ‘tree’,[46][47]yomi2/yomo2– ‘Hades’.[47]
  • e1 < *ia: sake1ri ‘blooming’ < saki1 ‘to bloom’ + ari ‘to be’.[49][50]
  • e1 < *iə: pe1ku (correct identify) < pi1 ‘solar’ + o2ki1 ‘put’.[49]
  • e2 < *ai: me2/ma– ‘eye’,[47]ame2/ama– ‘heaven’, ame2/ama– ‘rain’, kage2/kaga– ‘shade’.[51]
  • o1 < *ua: kazo1pu ‘to depend’ < kazu ‘quantity’ + apu ‘to mix’.[44][49][52]
  • o1 < *uə: sito1ri ‘sort of native weaving’ < situ ‘native weaving’ + ori ‘weaving’.[44][49][53]

There are additionally alternations suggesting e2 < *əi, corresponding to:

  • se2/so2– ‘again’, me2/mo2– ‘bud’[47]

Some authors imagine that they belong to an earlier layer than i2 < *əi, however others reconstruct two central vowels *ə and *ɨ, which merged in every single place besides earlier than *i.[50][54]
Different authors attribute the variation to totally different reflexes in numerous dialects and notice that *əi yields e in Ryukyuan languages.[55]

Some authors additionally postulate *e and *o to account for word-final e1 and o1 respectively.[56]
A couple of alternations, in addition to comparisons with Jap Outdated Japanese and Ryukyuan languages, counsel that *e and *o additionally occurred in non-word-final positions at an earlier stage however had been raised in such positions to i1 and u, respectively, in Central Outdated Japanese.[56][57]
The mid vowels are additionally present in some early mokkan and in some trendy Japanese dialects.[58]


Miyake reconstructed the next stock, along with a zero vowel-initial onset /∅/:[59]

Outdated Japanese consonants
voiced prenasalized*ᵐb*ⁿd*ⁿz*ᵑɡ

The unvoiced obstruents /p, t, s, ok/ had the voiced prenasalized counterparts /ᵐb, ⁿd, ⁿz, ᵑɡ/.[59]
Prenasalization nonetheless occurred within the late seventeenth century (in response to the Korean textbook Ch’ŏphae Sinŏ) and is present in some Trendy Japanese and Ryukyuan dialects, but it surely has disappeared in Trendy Japanese aside from the intervocalic nasal cease allophone [ŋ] of /ɡ/.[60]
The sibilants /s/ and /ⁿz/ might have been palatalized earlier than e and i.[61]

Comparative proof from Ryukyuan languages means that Outdated Japanese p continued an earlier voiceless bilabial stop *p.[62]
There’s basic settlement that word-initial p had turn into a voiceless bilabial fricative [ɸ] by Early Center Japanese, as advised by its transcription as f in later Portuguese works and as ph or hw within the Korean textbook Ch’ŏphae Sinŏ. In Trendy Normal Japanese, it’s romanized as h and has totally different allophones earlier than varied vowels. In medial place, it turned [w] in Early Center Japanese however has disappeared besides earlier than a.[63]
Many students argue that p had already lenited to [ɸ] by Outdated Japanese, however Miyake argues that it was nonetheless a cease.[64]

Pre-Outdated Japanese

Inner reconstruction means that the Outdated Japanese voiced obstruents, which all the time occurred in medial place, arose from the weakening of earlier nasal syllables earlier than unvoiced obstruents:[65][66]

  • b /ᵐb/ < *-mVp-, *-nVp-: e.g. abi1ki1 ‘trawling’ < ami1 ‘web’ + pi1ki1 ‘pull’.
  • d /ⁿd/ < *-mVt-, *-nVt-: e.g. yamadi ‘mountain path’ < yama ‘mountain’ + mi1ti ‘path’.
  • z /ⁿz/ < *-mVs-, *-nVs-: e.g. the title murazi < mura ‘village’ + nusi ‘grasp’.
  • g /ᵑɡ/ < *-mVk-, *-nVk-.

In some instances, there isn’t a proof for a previous vowel, which leads some students to posit last nasals on the earlier stage.[36]

Some linguists counsel that Outdated Japanese w and y derive, respectively, from *b and *d sooner or later earlier than the oldest inscriptions within the sixth century.[67]Southern Ryukyuan varieties corresponding to Miyako, Yaeyama and Yonaguni have /b/ comparable to Outdated Japanese w, however solely Yonaguni (on the far finish of the chain) has /d/ the place Outdated Japanese has y:[68]

  • ba ‘I’ and bata ‘abdomen’ comparable to Outdated Japanese wa and wata
  • Yonaguni da ‘home’, du ‘sizzling water’ and dama ‘mountain’ comparable to Outdated Japanese ya, yu and yama

Nonetheless, many linguists, particularly in Japan, argue that the Southern Ryukyuan voiced stops are native improvements,[69] adducing a wide range of causes.[70]

Some supporters of *b and *d additionally add *z and *g, which each disappeared in Outdated Japanese, for causes of symmetry.[71]
Nonetheless, there may be little or no Japonic proof for them.[36][72]


Vowel elision or fusion occurred to stop vowel clusters.
When a monosyllabic morpheme was adopted by a polysyllabic morpheme that started with a vowel, the second vowel was dropped:[73][74]

  • wa + ga + ipe1wagape1

In different environments, the primary vowel was dropped:[73][74]

  • ake + uaku
  • to2ko2 + ipato2ki1pa

Elsewhere, the vowels seem to have fused:[75]

  • ko2 + iki2


Though trendy Japanese dialects have pitch accent techniques, they had been often not proven in man’yōgana. Nonetheless, in a single a part of the Nihon Shoki, the Chinese language characters appeared to have been chosen to signify a pitch sample much like that recorded within the Ruiju Myōgishō, a dictionary that was compiled within the late eleventh century. In that part, a low pitch syllable was represented by a personality with the Middle Chinese stage tone, and a excessive pitch was represented by a personality with one of many different three Middle Chinese tones.
(An analogous division was used within the tone patterns of Chinese language poetry, which had been emulated by Japanese poets within the late Asuka period.)
Thus, it seems that the Outdated Japanese accent system was much like that of Early Middle Japanese.[76]


As in later types of Japanese, Outdated Japanese phrase order was predominantly subject–object–verb, with adjectives and adverbs previous the nouns and verbs they modify and auxiliary verbs and particles persistently appended to the primary verb.[77]


Many Outdated Japanese pronouns had each a brief type and an extended type with connected -re of unsure etymology.
If the pronoun occurred in isolation, the longer type was used.

With genitive particles or in nominal compounds, the quick type was used, however in different conditions, both type was potential.[78]

Personal pronouns had been distinguished by taking the genitive marker ga, in distinction to the marker no2 used with demonstratives and nouns.[79]

  • The primary-person pronouns had been a(re) and wa(re), had been used for the singular and plural respectively, although with some overlap. The wa- varieties had been additionally used reflexively, which means that wa was initially an indefinite pronoun and steadily changed a.[79]
  • The second+particular person pronoun was na(re).[80]
  • The third-person pronoun si was a lot much less generally used than the non-proximal demonstrative so2 from which it was derived.[81]
  • There have been additionally an interrogative pronoun ta(re) and a reflexive pronoun o2no2.[80]

Demonstratives usually distinguished proximal (to the speaker) and non-proximal varieties marked with ko2– and so2– respectively.
Many varieties had corresponding interrogative varieties i(du)-.[82]

Outdated Japanese demonstratives[83]

In Early Center Japanese, the non-proximal so- varieties had been reinterpreted as hearer-based (medial), and the speaker-based varieties had been divided into proximal ko- varieties and distal ka-/a- varieties, yielding the three-way distinction that’s nonetheless present in Trendy Japanese.[84]


Outdated Japanese had a richer system of verbal suffixes than later types of Japanese.[85]
Outdated Japanese verbs used inflection for modal and conjunctional functions.[86]
Different classes, corresponding to voice, tense, aspect and mood, had been expressed through the use of elective suffixed auxiliaries, which had been additionally inflected.[87]

Inflected varieties

As in later types of Japanese, Outdated Japanese verbs had numerous inflected varieties.
In conventional Japanese grammar, they’re represented by six varieties (katsuyōkei, 活用形) from which all of the others could also be derived in a similar way to the principal parts used for Latin and different languages:[88]

Mizenkei (irrealis)
This kind by no means happens in isolation however solely as a stem to which a number of particles and auxiliaries are connected.[89] Unger calls it a “pseudostem” as a result of the purported inflection was initially an preliminary *a of the suffixes connected to that stem.[90]
Renyoukei (adverbial, infinitive)
This kind was used because the infinitive.[91] It additionally served as a stem for auxiliaries expressing tense and facet.[92]
Shushikei (conclusive, predicative)
This kind was used as the primary verb concluding a declarative sentence.[86] It was used additionally earlier than modal extensions, last particles, and a few conjunctional particles.[93] The conclusive type merged with the attributive type by about 1600, however the distinction is preserved within the Ryukyuan languages and the Hachijōjima dialects.[94]
Rentaikei (attributive, adnominal)
This kind was used because the verb in a nominalized clause or a clause modifying a noun.[95] It was additionally used earlier than most conjunctional particles.[96]
Izenkei (realis, exclamatory, subjunctive)
This kind was used as the primary verb in an exclamatory sentence or because the verb in an adverbial clause.[97] It additionally served as a stem for the particles -ba (provisional) and -do (concessive).[98]
Meireikei (crucial)
This kind expressed the imperative mood.[97]

This method has been criticized as a result of the six varieties aren’t equal, with one being solely a combinatory stem, three solely phrase varieties, and two being each. It additionally fails to seize some inflected varieties.
Nonetheless, 5 of the varieties are fundamental inflected verb varieties, and the system additionally describes nearly all prolonged varieties persistently.[99]

Conjugation courses

Japanese verbs are categorized into eight conjugation courses, every being characterised by totally different patterns of inflected varieties.
Three of the courses are grouped as consonant bases:[100]

Yodan (quadrigrade)
This class of standard consonant-base verbs contains roughly 75% of verbs.[100] The category is so named as a result of the inflections in later types of Japanese span 4 rows of a kana desk, comparable to 4 vowels. Nonetheless, in Outdated Japanese, 5 totally different vowels had been concerned.[101] The bases are nearly the entire type (C)VC-, with the ultimate consonant being p, t, ok, b, g, m, s or r.[102]
Na-hen (n-irregular)
The three n-base verbs type a category of their very own: sin- ‘die’, -in- ‘depart’ and the auxiliary -(i)n- expressing completion of an motion. They’re usually described as a “hybrid” conjugation as a result of the adnominal and exclamatory varieties adopted an analogous sample to vowel-base verbs.[103]
Ra-hen (r-irregular)
The irregular r-base verbs had been ar- ‘be, exist’ and different verbs that included it, in addition to wor- ‘sit’, which turned the existential verb or- in later types of Japanese.[104]
Conjugation of consonant-base verbs[105]
Verb classIrrealisInfinitiveConclusiveAdnominalExclamatoryCrucialGloss
r-irregularara-ariariaruareare‘be, exist’

The distinctions between i1 and i2 and between e1 and e2 had been eradicated after s, z, t, d, n, y, r and w.

There have been 5 vowel-base conjugation courses:

Shimo nidan (decrease bigrade or e-bigrade)
The most important common vowel-base class led to e2 and included roughly 20% of verbs.[102]
Kami nidan (higher bigrade or i-bigrade)
This class of bases led to i2 and included about 30 verbs.[102]
Kami ichidan (higher monograde or i-monograde)
This class accommodates about 10 verbs of the shape (C)ii-. Some monosyllabic i-bigrade verbs had already shifted to this class by Outdated Japanese, and the remainder adopted in Early Center Japanese.[106]
Ka-hen (ok-irregular)
This class consists of the only verb ko2– ‘come’.[107]
Sa-hen (s-irregular)
This class consists of the only verb se- ‘do’.[107]

Early Center Japanese additionally had a Shimo ichidan (decrease monograde or e-monograde) class, consisting of a single verb kwe- ‘kick’, which mirrored the Outdated Japanese decrease bigrade verb kuwe-.[108][109][110][111]

Conjugation of vowel-base verbs[105]
Verb classIrrealisInfinitiveConclusiveAdnominalExclamatoryCrucialGloss
i-bigradeoki2oki2okuokuruokureoki2‘come up’

The bigrade verbs appear to belong to a later layer than the consonant-base verbs.[112]
Many e-bigrade verbs are transitive or intransitive counterparts of consonant-stem verbs.[113]
In distinction, i-bigrade verbs are usually intransitive.[114]
Some bigrade bases additionally seem to mirror pre-Outdated-Japanese adjectives with vowel stems mixed with an inchoative *-i suffix:[90][115][116]

  • *-a-i > -e2, e.g. ake2– ‘redden, lighten’ vs aka ‘purple’.
  • *-u-i > -i2, e.g. sabi2– ‘get desolate, fade’ vs sabu- ‘lonely’.
  • *-ə-i > -i2, e.g. opi2– ‘get large, develop’ vs opo- ‘large’.


Outdated Japanese adjectives had been initially nominals and, not like in later durations, could possibly be used to change nouns that adopted.[117][118]
They is also conjugated as stative verbs and had been divided into two courses:[119]

Conjugation of stative verbs[120][121]

The second class had stems ending in -si, which differed solely within the conclusive type, whose suffix -si was dropped by haplology.[122]
Adjectives of this class tended to specific extra subjective qualities.[123]
Lots of them had been fashioned from a verbal stem by the addition of a suffix -si, of unsure origin.[124]

A extra expressive conjugation emerged in direction of the tip of Outdated Japanese by including the verb ar- ‘be’ to the infinitive, with the sequence -ua- decreasing to -a-:[119]

Revolutionary conjugation of stative verbs[125]

Many adjectival nouns of Early Center Japanese had been primarily based on Outdated Japanese adjectives that had been fashioned with suffixes -ka, -raka or -yaka.[126][127]


Though most Outdated Japanese writing represents the language of the Nara courtroom in central Japan, some sources come from jap Japan:[128][129][130]

  • 230 azuma uta ‘jap songs’ in quantity 14 of the Man’yōshū,
  • 93 sakimori uta ‘borderguard songs’ in quantity 20 of the Man’yōshū, and
  • 9 songs within the Hitachi fudoki (recorded 714–718i, however the oldest extant manuscripts date from the late seventeenth century and present important corruption[131]).

They report Jap Outdated Japanese dialects,[131] with the next options:

  • There isn’t any kind A/B distinction on entrance vowels i and e, however o1 and o2 are distinguished.[132]
  • Pre-Outdated Japanese *ia yielded a within the east, the place Central Outdated Japanese has e1.[132]
  • The adnominal type of consonant-base verbs led to -o1, however Central Outdated Japanese had -u as within the conclusive type.[133] An analogous distinction is preserved in Ryukyuan languages, suggesting that Central Outdated Japanese had innovated by merging these endings.[134]
  • The crucial type of vowel-base verbs connected -ro2, as a substitute of the -yo2 utilized in Central Outdated Japanese.[135]
  • There was a bunch of distinctive adverse auxiliaries however don’t appear to be the supply of the totally different negatives within the trendy jap and western Japanese dialects.[135]


Works cited

Additional studying


Please enter your comment!
Please enter your name here