Amuric Loanwords in Old Japanese [slides]

Ver.2.0 Return of the Horse Riders MIYANO, Satoshi Part1 : Introduction ◇Nivkh (Gilyak) – the last remnant of Amuric ・A ”Paleosiberian” language (or languages) ・Today, spoken in the lower Amur river region and the northern part of Sakhalin ・No known genetic relationship with other languages ・Severely endangered Nivkh Amur River Sakhalin Manchuria Hokkaidō Google Map ◇Janhunen’s hypothesis (2010, 2016) ・In the past, Amuric occupied a geographically more central position of the Greater Manchuria Region ・Once belonged to a society with “higher” culture ・Possible connection with the historical 夫餘 (Puyŏ/Fuyu/Fuyo) people who once had a significant cultural presence in ancient Manchuria and beyond 2 Part1 : Introduction ◇Old Japanese (OJ)  A Japonic language spoken in Japan during the 8th century ◇Extensive past researches on OJ loanwords (and cognates)  Major: Koreanic, Chinese, Sanskrit, Ainu, Austronesian, (Ryukyuan)  Other: Altaic, Tai-Kadai, Indo-European, Tamil, Hebrew, etc. ◇This study: Another layer of loanwords in OJ  Amuric : little attention in previous studies …Before discussing Amuric loanwords in OJ, some remarks must be made 3 Part1 : Introduction Vowels ◇Amuric : donor ・Vertical Vowel Harmony in Proto-Nivkh →Vowel height may vary across dialects or depending on affixes VH | VL i | e ə | a u | o Symbol I A U ◇(Pre-)Old Japanese : recipient ・Mid Vowel Raising in Pre-OJ: *e → i, *o → u (Miyake 2003) ・Arisaka’s Law: o*ə+~a alternation e.g. so-ko ‘that-place’, umi-ka ‘sea-place’ ◇Paekche Old Korean: intermediate (at least in some cases) ・Unknown Vowel System Due to these factors, vowel height correspondences can be “lax” 4 Part1 : Introduction Consonants ◇Amuric Development of continuant series from primary stops in Proto-Nivkh (usually in intervocalic positions) Labial Dental Palatal Velar Uvular Fortis stop > Voiceless Continuant p’ > f t’ > ř c’ > s k’ > x q’ > χ Lenis stop > Voiced Continuant p>v t>r c>z k>ɣ q>ʀ Symbol P T C K *Laryngeal contrasts in Nivkh only exist in initial positions (Shiraishi 2006) *PN *c’A-/cA-, *c’U-/cU- → OJ /tV-/, otherwise PN *C → OJ /s/ Continuant consonants in Amuric can correspond to Old Japanese stops 5 Part1 : Introduction In Old Japanese, recent loanwords tend to have… ◇Limited distribution ・Become obsolete/obscure in the later stages of Japanese ・No cognate in the Ryukyuan languages ◇Morphological opacity ・Long (three or more syllables), but unsegmentable ・Unusual variations ◇Distinctive cultural factors ・Pertaining to a foreign concept/culture ・Semantic field: technology, religion, politics… 6 Part1 : Introduction OJ potokey < *pətəkay ‘Buddha’ ◇Morphology: Opaque ◇Culture: Foreign religion Source: Sanskrit buddha ‘enlightened’ → Chinese → Paekche (Old Korean) → Pre-Old Japanese (Pellard 2014) OJ *sitoki > MJ sitogi ‘rice cake’ ◇Distribution: Obscure/No Ryukyuan cognate ◇Morphology: Opaque Source: Koreanic cf. Middle Korean stek < *sVtek ‘rice cake’ OJ iraka ‘roof (tile)’ ◇Distribution: Obsolete/No Ryukyuan cognate ◇Morphology: Opaque ◇Culture: Technological term Source: Unknown “It is for example a strong hypothesis that OJ iraka 'roof, roof tile' is a loanword, but we do not know from where.” (Frellesvig 2010) 7 Part2 : Amuric loanwords in Old Japanese Unless noted, Proto-Nivkh (PN) reconstructions and modern Nivkh dialectical forms are based on Fortescue (2016) OJ yup- ‘tie, bind’ ・This word is also found in the Ryukyuan languages Source: PN *yup- < *i-hup- ‘tie, bind’ ・Obviously related ・The Nivkh word is based on a simple verb stem *hup- and a prefix *i(PN *yup- is a complex form, while OJ yup- is not) →OJ yup- ‘tie, bind’ must be a loanword from Amuric ・Ryukyuan cognate can be explained by several ways ‣Early Amuric loan into (Pre-)Proto-Japonic? ‣Later Japanese loan into Proto-Ryukyuan? OJ napo ‘still’ < Pre-OJ *napə or *napo ◇Distribution: No Ryukyuan cognate Source: PN *napə < *na-pə ‘until now’ ・Obviously related → OJ napo < Pre-OJ *napə, but it violates Arisaka’s law ・PN *napə ‘until now’ is a derivation of Amuric root *na ‘now’ cf. PN *na-v ‘now’ →OJ napo must be an Amuric loanword 8 Part2 : Amuric loanwords in Old Japanese OJ tadamuki ‘forearm’ < *ta-N-tamuki ‘hand-GEN-forearm’ ◇Distribution: Obsolete/No Ryukyuan cognate ◇Morphology: Opaque ・OJ tadamuki ‘forearm’ < *ta-N-tamuki ‘hand-GEN-tamuki’ →*tamuki can not be analyzed as a Japonic word (its meaning was probably just ‘forearm’) Source: PN tamk ‘hand’ < Pre-PN *tamVk ・In Nivkh, syllable initial or final consonant clusters are secondary (Janhunen 2016) OJ sitoto < *sitətə ‘small bird’※1 ※1 a long-tailed passerine bird (hōjiro in NJ) ◇Distribution: Obsolete/No Ryukyuan cognate ◇Morphology: Opaque Source: PN c’rat < Pre-PN *c’Vtat ‘small bird’※2 ※2 a long-tailed sparrow-sized bird , akin to hōjiro acc. Takahashi (1942) ・In Nivkh, continuant consonants developed from primary stops (Janhunen 2016) ・Final segment /o/ in the OJ word is an epenthetic echo vowel cf. MC pak (博), tok (徳) → OJ paka-toko (博徳, personal name) 9 Part2 : Amuric loanwords in Old Japanese OJ asi ~ a- < *asuy ~ a ‘foot, leg’ ◇Distribution: Absent from Proto-Ryukyuan ◇Morphology: Unusual variation ・OJ variation asi ~ a- cannot be explained by Japanese internal mechanism Source: PN *ŋazl ‘foot’, *ŋacɣ ‘leg’ ・ŋ(a)- is a common prefix found in many body-part words in PN ・The OJ variation asi ~ a- originated from different affixation in Amuric PN *ŋazl ‘foot’ < *ŋ(a)-ac-Vl → Pre-OJ *asur > *asuy ‘foot, leg’ PN *ŋacɣ ‘leg’ < *ŋ(a)-ac-(ɣ) → Pre-OJ *a OJ J toko < *təkə ‘everlasting’ ◇Morphology: Unusual variation (see below) Source: PN *cək ‘(a) long time’ ・OJ variation toko ~ soko cf. kuni-no-toko-tati (国常立) ～ kuni-no-soko-tati (国底立) in Nihon-shoki → From a foreign consonant /c/ in Amuric 10 Part2 : Amuric loanwords in Old Japanese OJ patakey < Pre-OJ *patakay ‘mange, scabies (疥)’ ◇Distribution: Obscure/No Ryukyuan cognate. ◇Morphology: Opaque Previous analysis suggested by Martin (1987) OJ patakey ‘mange, scabies’ < PJ *panta ‘skin’ + *kak[a-C]i ‘scratching’ But… ・Proto-Japonic *-nt- (> OJ /d/) > OJ /t/ : Exceptional and unmotivated →OJ patakey cannot be etymologically related to PJ panta ‘skin’. →MJ variation fatake ~ fadake was influenced by MJ fada ‘skin’ < PJ panta cf. MJ sira-fatake ~ sira- fadake ~ sira-fada for ‘白癩 (lit. white leprosy)’ ・*kak-i ‘scratching’ > *kay : unwarranted in Pre-OJ OJ patakey has no Japonic etymology → there must be an external source Source: Amur Nivkh vəri ‘pustule’ < PN *wə(ɣ)rk-i < Pre-PN *wA(k)tVk-Vy (my reconstruction) ・A long discussion is required to reconstruct the Pre-PN form of AN vəri ‘pustule’ 11 Part2 : Amuric loanwords in Old Japanese Proto-Nivkh *wə(ɣ)rk-i ‘pustule’ cf. AN varx ‘scab’ in Savel'eva&Taksami (1970) PN *wəri ‘pustule’ in Fortescue (2016) PN *vərɣ-i ~ *vərx ‘scab’ in Nikolaev (2015) 〇AN vəri ‘boil, pustule’ : Obviously a derivation of AN vərk- ‘rot’ 〇/v/ ≠ /w/ distinction: lost in AN, but preserved in SSN (cf. SSN wəřk- ‘smell bad’) 〇Other dialectical forms with -ɣř- rather than -řk- (or -rk-) e.g. ESN vəřk- ~ vəɣřkə- ‘rot’, (also NSN wəɣřt ‘pus’) → metathesis? (i.e. *wəřk-kə- > vəɣřkə-) Uncertain at the current stage → lost consonant? (i.e. *wəɣřk- > vəřk-) Note that there are… ・Frequent metathesis of consonant clusters in Nivkh cf. AN n’ivx ~ ESN n’iɣvŋ ‘person’ ・Frequent loss of /ɣ/ and /ʀ/ in consonant clusters in Nivkh (VɣC, VʀC → VC) ・AN vivus ~ ESN viɣvuř ‘belt’ ・AN oʀri ~ ESN ori ‘back of neck’ cf. PN *wərk- ‘rot’ in Fortescue (2016) 〇PN word for ‘rot’ : *wə(ɣ)rk→ PN word for ‘pustule’ : *wə(ɣ)rk-i > *wəɣr-i > AN vəri (Metathesis and Cluster Reduction) 12 Part2 : Amuric loanwords in Old Japanese PN *wə(ɣ)rk-i ‘pustule’ — further reconstruction is possible 〇Development of diphthong –Vy to monophthong –i in Nivkh ・Internal evidence : dialectical variations in Nivkh AN n’evrqay: ESN n’evřqi ‘eyelash’ AN tavʀi: ESN tavʀay ・External evidence : loanwords in Ainu PN *t’laŋi ‘reindeer’ → Ainu tunakay ‘id’ ‘crab’ 〇Secondary nature of syllable initial or final consonant clusters cf. PN *t’laŋi ‘reindeer’ < Pre-PN *t’Vla-ŋa-y (Janhunen 2016) 〇Cross-linguistically typical consonant lenition in intervocalic or inter-sonorant positions (e.g. -VtV- > -VrV-) 〇Vowel height alternation due to the vertical vowel harmony cf. AN vərk- ‘rot’ ～ varx ‘scab’ ～ vəri ‘pustule’ PN *wə(ɣ)rk-i < Pre-PN *wA(k)tVk-Vy ‘pustule’ →Most likely the source of Pre-OJ *patakay ‘mange, scabies’ 13 Part2 : Amuric loanwords in Old Japanese One problem left —correspondence between PN *w and OJ /p/ ・Amuric and Japonic : both have primary /w/ (Whitman 2012, Janhunen2016) → an Intermediate language which lacks /w/ cf. Certain Germanic loanwords in English: intermediate French language caused the correspondences between Proto-Germanic *w and English /g/ Proto-Germanic has /w/ (Old) French lacks /w/ English has /w/ *wītaną → guide → guide *wardāną → guarde → guard *warjaną → garantie → guarantee ・Koreanic : 〇No primary /w/ 〇Frequent source of OJ loanwords This word is not found in Middle Korean (a descendant of Silla Old Korean) →Now lost Paekche Old Korean (POK) was the intermediate language Summary of this etymology Pre-PN has /w/ *wA(k)tVk-Vy ‘pustule’ POK lacks /w/ → *pA(k)tVkVy Pre-OJ has /w/ → *patakay ‘mange, scabies’ 14 Part2 : Amuric loanwords in Old Japanese Other Examples of Possible Amuric loanwords in OJ(1)  PN *a- ‘over there’ → MJ a- ‘distal demonstrative’  PN *ceqa- ‘firm’ = OJ sika-to ‘certainly, clearly’, OJ siko ‘strong, tough’  SSN *d’ankř ‘vomit’ → OJ tagur- ‘id’ ・OJ tagur- ‘vomit’ has no Ryukyuan cognates. ・The Nivkh word is probably a complex form. cf. SSN d’a-nd ‘vomit’  PN *esqa ‘hate’ → MJ isakap- < *√isak(a)-Ap- ‘quarrel’ <*’hate each other’  AN ərkə ‘very nearly’ (Nedjalkov 2013) < *ara-ka → OJ ataka-mo ‘as if’ cf. PN *ara ‘almost’  PN *ərkə(r) ‘shore, bank’ → MJ ataka, atake *’river mouth (place name)’ cf. PN *ərŋ ‘river mouth’  PN *ha- ‘be such’ → OJ ka- ‘id’ ・The OJ word has no Ryukyuan cognate  PN *i- ‘verbal prefix’ = OJ i- ‘verbal prefix’  PN *kul- ‘cut out’ = MJ kur- ‘hollow out, gouge out’  PN *kut- ‘fall down’ = OJ kutat- ‘come down’ ~ kutas- ‘rot (tr)’ ~ kuti- ‘rot (intr)’ ・Other Japanese verb roots are also possibly related (*√kud(u)- ‘collapse’,*√kud(a)- ‘go down’)  PN *k’edr- ’rub on, grate’ → MJ kedur- ‘shave, plane’  PN *k’əz- ‘dig’ → OJ kozi- ‘id’  PN *k’rə ‘cliff, promontory’ → OJ kura-tani ‘steep valley?’ ・The OJ word is quite obscure and has no Ryukyuan cognate. 15 Part2 : Amuric loanwords in Old Japanese Other Possible Amuric loanwords in Old Japanese (2)  PN *-la- ‘permanent quality’ → OJ -ra (adjective suffix with unknown function) ・The Nivkh suffix -la- attaches to qualitative verbs (Nedjalkov 2013) ・This element is also found in many obscure OJ adjectives (e.g. ira, masura, isara, sume-ra…)  PN *ma ‘span between fingers’ → OJ made ‘untill, so much that’ < *ma-de ‘span between two hands’ ・OJ restrictive particle made ‘untill’ must have a nominal source (Frellesvig 2010)  PN *paχ ‘stone’ → MJ fake ‘cliff (in place names)’  PN *poz- ‘lie down’ = OJ pus- ‘lie prone’  PN *p’ar(p’ar)- < *p’ata-p’ata- ‘flap wings’, PN *parparu- ‘lightning’ → MJ fata(fa)ta-gami ‘thunder (god)’ < *patapata-N-kamuy ‘flapping-GEN-god’ ・PN *p’ar(p’ar)- ‘flap wings’ is associated with ‘lightning’ by the thunderbird (Takahashi 1942)  PN *p’oti ‘string, rope’ = OJ podas- < *√pwod(V)-As- ‘tie up, bind’  PN *qos ‘twig for stringing smelt’ → OJ kusa ‘branch’ in saki-kusa (三枝)  PN *q’alŋ < *qala-ŋ ‘tribe’ → OJ kara ‘blood kin’ ・This word is also found in Tungusic  PN *q’ar ‘spade’ = OJ kara-suki ‘t.o. spade’  PN *t’vi- ‘finish’ = OJ tupi ‘end’  PN *waqi ‘box’ → OJ pakwo ‘box’  PN *zaq ‘a kind of small bird’ = OJ sazaki < sa-N-saki ‘small bird’  Nogliki Nivkh toř ‘clay, earth’ (Tangiku 2008) = OJ tuti ‘earth’ 16 Part3 : Discussion How did these Amuric elements enter into OJ? ◇When ・Before the 8th century (Pre-OJ) ・After the split of Japonic into Japanese/Ryukyuan branches →during the 3-7th century (Hattori 1959, Pellard 2016) ◇Where ・From: Central Manchuria (Amuric homeland according to Janhunen) ・Into: Japan (where Pre-OJ was spoken) Sakhalin Amuric Manchuria ◇Which route Only two possibilities 1. From Sakhalin: Through Sakhalin, Hokkaidō, and Northern Honshū 2. From Korea: Through Manchuria and the Korean Peninsula Hokkaidō Korean Peninsula PreAinu Koreanic Pre-OJ ProtoRyukyuan Japan Google Map 17 Part3 : Discussion ◇Historical Evidence ・Historically, the Nivkh language has been spoken in the lower Amur River and northern Sakhalin ・The Okhotsk people from Sakhalin, identified as Mishihase (粛慎) in Nihon-Shoki, traveled as far south as Sado island in the 6-7th century (Hudson 2017) →Possibly Pre-Proto-Nivkh speakers? ◇Linguistic Evidence ・Prehistoric language contacts between Nivkh and Ainu →Probably took place in Hokkaidō (Vovin 2016) ・One possible Amuric loanword into OJ via an Ainu intermediate. PN *yup- ‘tie, bind’ < *i-hup→ Proto-Ainu yup- ‘id’ → OJ yup- = Proto-Ryukyuan yup- ‘id’ But note that ・No linguistic indication that the OJ word was borrowed from Ainu 18 Part3 : Discussion ◇Linguistic Evidence (1) Amuric loanwords into OJ via POK: PN *w→OJ /p/ ・PN *wə(ɣ)rk-i ‘pustule’ → POK *pA(k)tVkVy → Pre-OJ *patakay ‘mange, scabies’ ・PN *waqi ‘box’ → OJ pakwo ‘box’ ・PN *walɣ- ‘open one’s heart’ < *w-alɣ→OJ paruk-/parak- ‘open’, MJ faruk- ‘(mind) becomes clear’ One more possible Amuric loanword via POK OJ kusa ‘branch’ in saki-kusa ‘three-branch (三枝)’ ・OJ saki-kusa ‘three-branch’ must be a loanword from Koreanic (POK) cf. MK seyh < *seki ‘three’ ・OJ kusa ‘branch’ has no viable Koreanic source. (MK kaci ‘branch’ is unlikely related) ・Possible Amuric source: PN *qos ‘twig for stringing smelt’ > AN/ESN qos ‘id’ →has a straightforward phonological match with the OJ word →Further study is required for clarifying the original meaning ・ If this etymology is correct, the Amuric word must be borrowed into POK first and from there re-borrowed into (Pre-)OJ as part of a compound 19 Part3 : Discussion ◇Linguistic Evidence (2) : Shared vocabulary between Amuric and MK 1. Probable Amuric loanwords into MK • • • • • • • • • • • PN *esqa ‘hate (it)’ → MK ich- ‘id’ PN *hil-ɣ ‘tongue → MK hye ‘id’ < Pre-MK *hyel PN *k’uti ‘hole’ → MK kwut ‘cavity’ PN *k’ici-‘raise’ → MK chi- ‘rise, raise’ PN *liɣr < *liɣ-r ‘wolf’ → MK ilhi ~ ilhuy ~ ilhoy ‘id’ PN *n’e- ‘put on head or shoulders’ → MK ni-, nyey-, i- ‘put on head, place above’ PN *pasq ‘one of a pair, half’ → MK pcak ‘one of a pair, pair’ PN *q’acŋ ‘kind, sort’ → MK kaci ‘id’ PN *waqi ‘box’ → MK pakwoni, pakwulley ‘basket’ PN *war ‘trouser’ → MK pati ‘id’ PN *wela(ɣ)- ‘naked’ → MK palkapes- ‘take off all clothes’ < *palka ‘naked’ 2. Uncertain borrowing direction • • • • • • • • PN *a- ‘kinship prefix’ = MK a- ‘id’ PN *k’rə ‘cliff, promontory’ = MK kwol ‘valley’ PN *paʀ ‘stone’ = MK pahwoy ‘boulder’ PN *q’ar ‘spade’ = MK kal- ‘plow, cultivate’ PN *tam(a)- ‘silent’ = MK tamul- ‘shut the mouth’ PN *tata- ‘whole’ = MK ta ‘all’ PN *zaq ‘a kind of small bird’ = MK say < *saCi ‘bird’ Nogliki Nivkh c’osq ‘soot’ (Tangiku 2008) = MK swusk ‘charcoal’ 20 Part3 : Discussion From Korea? ◇Linguistic Evidence (3) : Amuric presence in ancient Paekche POK makari ‘true, right (正)’ ・Attested in Paekche royal titles from the Nihon-shoki (日本書紀) ・ makari-orikuku (正夫人) ・ makari-yomo (世子) ’true queen (consort)’ ‘heir (=true child)’ ・Also found in a lordly title of Koguryŏ from the Old Book of Tang (舊唐書) ・*makri-key (莫離支) ‘true king/ruler’ (Beckwith 2004). ◇Distribution: No Middle Korean cognate ◇Morphology: Unanalyzable in Koreanic ◇Culture: Foreign political title →Likely a recent loanword Source: PN *məkər- ‘straight’ ・Related to PN *maɣ(-tur) ‘true, right’ (also AN məks ‘true, right’, ESN məklkař ‘honest’) → Amuric root *mAk- ‘true, right, straight’ ・Since PN *məkər- is a complex form, this word must be Amuric in origin ・Likely borrowed into POK sometime after the split of Koreanic into Silla/Paekche variants and before the fall of Paekche (660). 21 Part3 : Discussion ◇Which route Sakhalin 1. ”From Sakhalin” 〇Not implausible 〇Scarce linguistic evidence 2. ”From Korea” 〇Bulky linguistic evidence ・Amuric loanwords via Koreanic in OJ ・Amuric loanwords in MK ・Direct evidence for Amuric presence in ancient Paekche Amuric Hokkaidō Manchuria Korean Peninsula PreAinu Koreanic Pre-OJ ProtoRyukyuan Japan Google Map Most (if not all) of the Amuric elements in Old Japanese traveled through Manchuria and the Korean Peninsula 22  Another layer of loanwords in Old Japanese: Amuric  These Amuric elements originated from Manchuria and traveled to Japan (mostly) via the Korean Peninsula during the 3-7th century (or earlier in Korea)  At least some of the Amuric loanwords in Old Japanese were borrowed through Paekche Old Korean  More investigation of Amuric elements in Manchuria, Korea, and Japan.  Linguistic reconstruction of Amuric based on these external information  Reconstruction of the ethnolinguistic (proto-)history of Northeast Asia 23 The End 24 Notes ・Unless noted, Proto-Nivkh reconstructions and modern Nivkh dialectical forms are based on Fortescue (2016) with some transcriptional modification ・Pre-Proto-Nivkh reconstructions are my own and they are largely based on Janhunen (2016) Abbreviations OJ: Old Japanese MJ: Middle Japanese NJ: Modern Japanese POK: Paekche Old Korean MK: Middle Korean PN: Proto-Nivkh AN: Amur Nivkh ESN: East Sakhalin Nivkh NSN: North Sakhalin Nivkh SSN: South Sakhalin Nivkh WSN: West Sakhalin Nivkh GEN: genitive tr: transitive intr: intransitive Symbols = : Shared vocabulary (with uncertain borrowing direction) → : Borrowing >: Change(s) within a language √: verb root C : consonant V : vowel A : Central Vowel (a~ə) I : Front Vowel (e~i) U : Back vowel (o~u) P: Labial T: Dental C: Palatal K: Velar or Uvular 25 Refferences (1) • • • • • • • • • • • • • • • • • Arisaka, Hideyo (有坂英世), 1934.古代日本語における音節結合の法則” 國語と國文學 22(1). Beckwith, Christopher I, 2004. Koguryo: The Language of Japan’s Continental Relatives. Leiden & Boston: Brill. Egami, Namio (江上波夫), 1967. 「騎馬民族国家」 中公新書. Fortescue, Michael, 2011. "The relationship of Nivkh to Chukotko-Kamchatkan revisited." Lingua 121.8 : 1359-1376. Fortescue, Michael, 2016. “Comparative Nivkh Dictionary.” Lincom. Frellesvig, Bjarke, 2010. A history of the Japanese language. Cambridge University Press. Gruzdeva, Ekaterina, 2004. "Numeral classifiers in Nivkh." STUF-Language Typology and Universals 57.2-3: 300-329. Hattori, Shiro (服部四郎), 1959. 「日本語の系統」東京：岩波書店． Janhunen, Juha, 2010. "Reconstructing the language map of prehistorical Northeast Asia." Studia Orientalia Electronica 108: 281-304. Janhunen, Juha, 2016. "Reconstructio externa linguae ghiliacorum." Studia Orientalia Electronica 117: 3-27. Halm, Robert, and Jay Slater, 2018. "Application of the comparative method to Nivkh: Other regular sound correspondences." Hudson, Mark J, 2017. Okhotsk and Sushen: History and Diversity in Iron Age Maritime Hunter-Gatherers of Northern Japan’. Oxford: Oxbow. Kōno, Rokurō (河野六郎), 1987, "百済語の二重言語性." 中吉先生喜寿記念・朝鮮の古文化論叢』 国書 刊行会. Miyake, Marc Hideo, 2003. "Philological evidence for* e and* o in Pre-Old Japanese." Diachronica 20.1: 83137. Nedjalkov, Vladimir P., and Galina A. Otaina, 2013. A syntax of the Nivkh language: The Amur dialect. Vol. 139. John Benjamins Publishing. Nikolaev, Sergei L, 2015b. "Toward the reconstruction of Proto-Algonquian-Wakashan. Part 2: AlgonquianWakashan sound correspondences." Journal of Language Relationship 13(4): 289–328. Martin, Samuel Elmo, 1987. The Japanese language through time. New Haven: Yale University Press. 26 Refferences (2) • • • • • • • • • • • • • • • Matsuo, Toshiro (松尾俊郎), 1952.「崖を意味する地名」("A Study of Place Names Meaning “Escarpment".”) 新地理 1.2: 1-10. Omodaka, Hisataka ,et al (澤瀉久孝), 1967. 「時代別国語大辞典上代編」三省堂. Pellard, Thomas, 2014. "The awakened lord: the name of the Buddha in East Asia." Journal of the American Oriental Society 134.4: 689-698. Pellard, Thomas. 2016.「日琉祖語の分岐年代」田窪行則・平子達也(編)『琉球諸語と古代日本語：日琉祖 語の再建に向けて』99–124.東京:くろしお出版. Russell, Kerri L, 2006. A Reconstruction and Morphophonemic Analysis of Proto-Japonic Verbal Morphology. Diss. University of Hawai ‘i. Savel'eva, V., & Taksami, C. (1970). Russko-nivkhskij slovar'. Moskva: Sovetskaja entsiklopedija. 小学館国語辞典編集部, 2006. 「日本国語大辞典 精選版」小学館. Shiraishi, Hidetoshi, 2006. Topics in Nivkh phonology. Diss. Rijksuniv. Shiraishi, Hidetoshi, 2013. "Phonologically conditioned allomorphy in Nivkh." アジア・アフリカの言語と言 語学 7 (2013): 67-79. Shiraishi, Hidetoshi (白石英才), and Tangiku, Itsuji (丹菊逸治), (2013) "ニヴフ語アムール方言の基礎語 彙." 北方言語研究 3 : 201-212. Takahashi, Moritaka (高橋盛孝), (1942). 「樺太ギリヤク語」朝日新聞社. Tangiku, Itsuji (丹菊創路), 2008. Basic vocabulary of the Sakhalin dialect of Nivkh language (Nogliki dialect) / ニヴフ語サハリン方言基礎語彙集 (ノグリキ周辺地域). Tokyo: Research Institute for Languages and Cultures of Asia and Africa, Tokyo University of Foreign Studies (TUFS) / 東京外国語大学アジア・アフリカ言 語文化研究所. Vovin, Alexander, 2010. Korea-Japonica: a re-evaluation of a common genetic origin. University of Hawaii Press. Vovin, Alexander, 2016. "On the Linguistic Prehistory of Hokkaido." On the Languages of Hokkaidō and Sakhalin. Ed. by J. Janhunen (в печати). Whitman, John, 2012. "The relationship between Japanese and Korean." The languages of Japan and Korea: 24-38. 27 Appendix: Sound correspondences Disclaimer: These are not “strict and regular” sound correspondences employed in comparative linguistics for dealing with genetic relationship ◇Vowels ◇Non-obstruents Language PN VH i ə u Front Vowel (I) Mid Vowel (A) Back Vowel (U) VL e a o OJ MK i, ye? o, a u, wo i, (y)e? u, o, a wu, wo ・Vowel height correspondences can be “lax” due to many uncertain factors Others OJ m n n(i)? ∅ w, (p) y n-?, -rk MK m n n(y) > ∅ ? p y l h ・PN *w corresponds to OJ /p/ in loanwords via a Koreanic intermediate ◇Obstruents Language PN Symbol Labial Alveolar Palatal Velar Uvular Nasals PN m n n’ ŋ w y l h P T C K Fortis Stop ⇔ Voiceless Continuant p’ ⇔ f t’ ⇔ ř c’ ⇔ s k’ ⇔ x q’ ⇔ χ Lenis Stop ⇔ Voiced Continuant p⇔v t⇔r c⇔Z k⇔ɣ q⇔ʀ ・Correspondences between PN stops and OJ daku-on are exceptional ・PN *cA-/c’A-, *cU-/c’U- → OJ /tV/, otherwise PN *C → OJ /s/ OJ MK p, (b) t, (d), r s, (z), t- p~β t~l (r) s~z, c k, (g) k~ɣ, h 28 Appendix: Common vocabulary between Amuric and PJ  PN *a(ra)qm ‘hail’ = OJ ara-re ‘id’  PN *cal- ‘drip’ → OJ tar- < PJ tar- ‘drip, hang down’  PN *hal ‘skin, body’ → OJ karada < PJ *kara-Nta ‘body’  PN *k’uti ‘hole’ = OJ kuti ~ kutu- < PJ *kutuy ‘mouth’  PN *ma ‘span between fingers’ = PJ *ma ‘span, interval’  PN *mam- ‘squeeze, crush’ = PJ *mom- ‘rub, knead, crumple’  PN *nav < *na-v ‘now’, *nana ‘recently’ = PR *nama < *na-ma ‘now’ ・PR *nama ‘now’ has no equivalent in Japanese (which has i-ma ‘now’)  PN *qar(ŋ) ‘crow’ = OJ karasu < PJ *kara-su ‘id’  PN *qaw- ‘dry’ = OJ kawak- < *√kaw(a)-Ak- ‘id’ cf. PR kawarak- ‘dry’  PN *q’oʀla ‘mind, soul’ = PJ *kokoro ‘heart, feeling’  PN *t’ləɣi ‘lynx’ = OJ twora ‘tiger’  PN *yup < *i-hup- ‘tie, bind (tr)’ → PJ *yup- ‘id’  Nogliki Nivkh c’osq ‘soot’ (Tangiku 2008) = PJ susu ‘soot’ cf. MK swusk ‘charcoal’  Nogliki Nivkh osq ‘hare’ = OJ usagi < PJ *osanki ‘hare’  Nogliki Nivkh c’xǝm- ‘grip, grasp, squeeze’ → PJ *tukam- ‘grip, grasp’ ・Earlier Amuric influence in Japan? ・Language contacts in Manchuria before the Yayoi migration? ・Later Japanese loanwords into Proto-Ryukyuan? 29 Appendix: Evidences for OJ asi < Pre-OJ *asuy ◇Evidence 1： shrine names Modern Japanese shrine names ・足羽 shrine (NJ Asuwa < OJ *Asupa) in Fukui prefecture ・足次山 shrine (NJ Asuwa-yama < OJ *Asupa-yama) in Okayama prefecture →reading asu for the character 足 ‘foot’ →allomorph asu- for OJ asi ‘foor, leg’. ※Note that one of the main deities of these shrines is named as 阿須波神 (OJ asupa), whose man’yōgana spelling indicates archaic nature of these shrine names. ◇Evidence 2：personal names of Emperor Ninken Personal names (諱) of the 24th Emperor Ninken (仁賢) in Nihon-Shoki ・大脚 (OJ opo-si) ※These are based on traditional reading (no OJ phonetic attestation) Pre-OJ *əpə-suy ・大為 (OJ opo-su) ※ OJ isi ‘stone’ < *iswi cf. OJ iswo-no-kami (石上) ・大石 (OJ opo-s(w)i) ・The character 脚 ‘leg’ for the historical syllable *swi < *suy →suggesting OJ asi ‘foot, leg’ < *aswi < *asuy ・It is even possible that the first spelling variant 大脚 is etymological i.e. OJ opo-si < opo-(a)si < *əpə-asuy ‘big-foot’ 30 Appendix: Development of Pre-PN *wA(k)tVk- ‘rot’ (tentative) PN *wəɣrk‘rot’ AN vərk-c, vəɣrk-c ‘rot’ ESN vərk-t, vəřk-χar-d ‘rot’ SSN wəřk- ‘smell bad’ PN *wəɣrk-əs‘rot’ NSN vūřɣəs-c ‘rot’ (irregular) ESN vəɣřkə-d ‘rot’ WSN vəřɣəs-c ‘rot’ PN *wəɣrk-nt ‘pus’ NSN wəɣřt ‘pus’ Nogliki Nivkh vəɣřkd ‘pus’ Pre-PN *wAktVk-V ‘scab’ PN *warɣ ‘scab’ AN varx ‘scab’ Pre-PN *wAktVk-V-y ‘pustule’ PN *wəɣrk-i ‘pustule’ AN vəri ‘pustule’ NSN wəri ‘pustule’ POK *pAktVkVy ‘pustule’ Pre-OJ *patakay → OJ patakey ‘mange, scabies’ Pre-PN *wAktVk‘rot’ ___ : attested forms 31