sign in sign up
<<
Home , Early Middle Japanese, Japanese language, Japanese sound symbolism, Kikaijima, List of Japanese writers, Mass comparison

The comparative study of the Japonic Thomas Pellard

To cite this version:

Thomas Pellard. The comparative study of the . Approaches to endangered lan- guages in and Northeast Asia: Description, documentation and revitalization, National Institute for Japanese and Linguistics, Aug 2018, Tachikawa, Japan. ￿hal-01856152￿

HAL Id: hal-01856152 https://hal.archives-ouvertes.fr/hal-01856152 Submitted 9 Aug 2018

HAL is a multi-disciplinary open access ’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. The comparative study of the Japonic languages

NINJAL International Symposium Approaches to endangered languages in Japan and Northeast Thomas Pellard Asia: Description, documentation and revitalization, Tachikawa: 6–8 August 2018 CNRS,CRLAO [email protected]

Why do we compare? (5) The goal of is and has always been to explain languages and their evolution (1) Goals and methods of science Another and not less important reason, which makes a • to advance our understanding of the world critical comparison of the Sanskrit with its European sis- • answering “what?” (observation) is not sufficient, we also ters, worthy to be undertaken, is the light thrown thereby need to ask “how?” and “why?” (explanation) upon each of the languages compared, and the clearer • the hallmark of the scientific method is the artificial ex- view we thence obtain of the most ancient forms of each periment, but such experiments are often not possible or respectively, and probably some conception of the original ethical in many fields, especially those who study the past and primitive signification of a great part of the grammat- and unique events: evolutionary biology, archaeology, but ical common to all. It is chiefly by comparison also epidemiology, ecology, sociology, etc. that we determine as far as our sensible and intellectual • one alternative to the controlled experiment is the “nat- faculties reach, the nature of things. Frederic Schlegel justly ural experiment”, or “”: manipulation expects, that comparative grammar will give us quite new is replaced by comparison with other attested cases explications of the genealogy of languages, in a similar • comparison has a long history and has been a standard way as comparative anatomy has thrown light on natural method of enquiry in anatomy for centuries philosophy. (Bopp 1820: 2) We consider that it is worth while to examine many It is usual to call comparative grammar the grammar animals of different species because what is often more that does not only describe but as far as possible explains condensed or more concealed in one species Nature linguistic forms, because it cannot as a rule limit itself to displays more clearly and openly in another. (Regnier the examination of a single language. (Schleicher 1861: 2) de Graaf 1663, Disputatio medica de natura et usu succi Comparison is the only effective tool available to the pancreatici, cited in Cole 1944: 12) linguist for writing the history of languages. We observe results of changes, not changes themselves. It is thus only (2) In linguistics too, comparison is an invaluable tool with the help of combinations that we follow, and can 1. typological comparison: search for the universal laws, follow, the development of languages. (Meillet 1925: 11) tendencies, and possible variations of language (see the Even if we were, by some miracle, handed a complete presentation by M. Shimoji) grammar of Common Indo-European as spoken somewhere 2. historical comparison: explanation of languages by refer- in, say, 4000 B.C. (the date is meaningless), the work of the ence to their evolutionary changes, as revealed by com- Indo-Europeanist would scarcely be done. In fact, it would parison with related languages1 be barely begun. For his task would be, then as before, to relate the facts vouchsafed him to the facts of attested The historical comparative method in linguistics languages: to construct hypotheses, and to demonstrate precisely how it is possible, within a linguistic tradition or (3) Motivation and method traditions, for a language to pass from one system at one • usually, we cannot observe ling. changes, only their results point in time to another system at a later point. (Watkins • there are usually few historical records, so that we have 1973: 101) to rely on comparison with related languages, which con- stitute as many pieces of the original puzzle (6) Contrary to more or less common misunderstandings • we infer (reconstruct) an earlier state (proto-language) • proving relatedness is not much a goal as a result of from which all languages can be plausibly derived the comparative method; nobody needed to prove that • regularity of and arbitrariness of form- the Indo-European languages were related: it was self- meaning associations allow to rule out coincidences evident since they shared irregularities that could only be explained through comparison (4) Alternatives to the traditional comparative method? • reconstructions are not speculative educated guesses but • mass comparison, lexical diffusion, allofams, typology- models that account for correspondences between related based classifications, etc., have not and will never produce languages and that allow to derive attested forms any reliable nor explain any irregular • a phylogenetic tree is not a goal nor simply pigeonholing, • i.. they do not solve interesting problems it is a model of the historical diversification of languages

1. “Comparative linguistics” has been used in the latter sense for more than (7) The most successful example of the comparative method is 200 years. It would be very inappropriate to turn everything on its head by its application to the Indo-European languages rebranding “” as “comparative linguistics”, as awkwardly • the ‘be’ is irregular in most languages, but it recon- proposed by Haspelmath (2018). See Morpurgo Davies (1998) for a good overview of the early history of the field and Hock (1991) for a good structs as a regular verb alternating between full grade in introduction to the concepts and methodology of . singular forms and zero-grade in forms (*h₁és : *h₁s)

1 Pellard The comparative study of the Japonic languages

Sanskrit Gothic pIE (11) Comparison within Japonic ‘be.pres.3sg’ ásti est ist *h₁és-ti • the first historical-comparative study of Ryukyuan and Ja- ‘be.pres.3pl’ sánti sunt sind *h₁s-énti panese is Chamberlain’s (1895) grammar of Shuri Okinawan • mainly of historiographic interest today, but Chamberlain • the name of the Greek god Zeus has an irregular declen- did think of comparative grammar as explanatory sion which can be shown to go back to an earlier regular It may be asked of what nature is the light which pattern where lengthened e-grade in strong cases altern- Luchuan [i.e. Ryukyuan] throws on Japanese. We reply ates with zero-grade in weak cases (*di̯ḗu̯- : *diu̯-), with a that it is such as, in general, sister languages shed upon sound law *di̯ > / in Greek each other. (Chamberlain 1895: 6) Japanese , too, mostly remained at the Greek Sanskrit pIE guessing stage. As for the accidence, grammarians could nom.sg zéus dyáus *di̯ḗu̯-s enumerate forms, but were often unable to explain gen.sg di()ós divás *diu̯-és them; they could imagine theories, but could not build trustworthy theories. With Luchuan to refer to the case (8) It is proven that historical-comparative linguistics does work becomes different, for Luchuan stands to Japanese in • Saussure (1879) made the bold hypothesis that some ie about the same relationship as Italian does to French forms could be better explained if we assumed that some and Spanish. (Chamberlain 1897: 56) sounds had been lost, and it was later discovered (Kuryłow- • still indebted to the breakthroughs achieved by Hattori icz 1927) that Hittite preserved in some cases Saussure’s re- Shirō, reprinted in Hattori (1959) and Hattori (2018) constructed sounds:2 ‘protect’ pIE *peh₂-s- > Hittite paḫš-, • the first attempt at a systematic reconstruction of proto- Latin pāstor, pāsco, Sanskrit pāsati Ryukyuan is Thorpe (1983), which also contains many com- • Bloomfield (1928) compared several parisons with Japanese, though not all are convincing and reconstructed a cluster *çk not distinguished in any • Pellard (2019) presents an overview of the current state of language, which he later discovered to be distinguished as the field of Japanese-Ryukyuan comparative linguistics htk in Swampee Cree • for comparison with Hachijō, see Kaneda (2001), Hirako & Pellard (2013), and references therein Japonic comparative linguistics (12) Methodological problems (9) Japonic (also Japanese-Ryukyuan, Japonesian) • a distrust, if not contempt, for reconstruction and a bias • Japanese, and its endangered dialects towards philology: tendency to enshrine cherrypicked Old • Ryukyuan: ~5 endangered languages spoken in the Ryukyu Japanese forms as proto-forms and to posit ad hoc adjust- Islands, a chain of around 50 inhabited islands stretching ments rather than to resort to comparative reconstruction4 from the Southeast of Kyushu to the Northeast of Taiwan3 None of these scholars has attempted to reconstruct • Hachijō: an endangered language spoken 287 km south of the and of the ancestral Proto- , in the Izu archipelago; its phylogenetic position has Ryukyuan language and to formulate the “sound laws” yet to be determined accounting for the linguistic situation found today in • the degree of diversity and the time depth of the family the islands. [… T]he comparative method […] has never can be said to be not much different from that of the been applied to these data. In Ryukyuan studies formal Romance or Slavic families (Pellard 2013; 2016a) historical statements are generally confined to arrange- Japonic ments of variant forms in hypothetical evolutionary se- ? quences, unsupported by comparative method evidence. (Thorpe 1983: 4–5) Ryukyuan Japanese Hachijō • a prejudice against unwritten non-standard varieties • the prominence of linguistic geography over systematic Southern Northern structural descriptions: though the study of the geograph- ical distribution of linguistic features is a valuable tool of Macro-Yaeyama Miyako Okinawa Amami historical enquiry (see the presentation by A. Kaneda & M. Holda), it is insufficient, and we cannot reconstruct a linguistic system from a map Yaeyama Some contributions of Ryukyuan to Japonic (10) Comparisons have usually focused on Japanese and another historical-comparative linguistics language or (Altaic, Austronesian, etc.) • such hypotheses seldom try to explain idiosyncrasies of (13) Apophony in oj Japanese (irregular , and accent, apophony, etc.) • the same i₂ shows two different alternation patterns • in absence of a convincing (i.e. explanatory) hypothesis of • leads to reconstruct two different relationship with other languages, comparison within the sources for oj i₂ Japonic family is the only alternative (Pellard 2011; 2016a) • this is confirmed by comparison with Ryukyuan

2. The now famous laryngeals, see Keiler 1970 and Lindeman 1987. 3. See Shimoji & Pellard (2010) and Heinrich et al. (2015) for a general introduction. 4. This is reminiscent of the situation in (Buchi 2012).

2 Pellard The comparative study of the Japonic languages

‘moon’ ‘tree’ ‘mouth’ ‘to fall’ (17) oj tuki₂ : tuku- ki₂ : ko₂- kuti : kutu- oti : otos- • no reflex of oj mesial so₂ in Ryukyuan pR *tuki *ke *kuti *{,o}te • pR mesial *{u,o}- = Hachijō distal u- Yamatohama tsˀɨkˀi kʰɨː kˀutɕˀi ʔutʰɨ- • *{o,ə} in pJ (cf. oj ono₂ ‘REFL’?) Yonamine ɕìtɕǐː kʰǐː kùtɕíː ɸùtˀìɾúɴ̀ adnominal Hirara tsɿkɿ kiː futsɿ utiɿ Yamatohama ʔuɾi ʔuɴ Ishigaki tsɿ̀kɿ́ kíː ɸútsɿ̀ útíɾúɴ Yonamine ʔùɾíː ʔúnù Yonaguni tˀìː kʰìː tˀíː ùtìɾùɴ Irabu uɾi unu Ishigaki úɾì únù (14) Mid Yonaguni úː ùnù • mid vowels are rare in oj and often secondary Hachijō uɾe uno • Ryukyuan requires to reconstruct *e and *o ‘ditch’ ‘water’ ‘daytime’ ‘garlic’ (18) Numeral ‘one’ pJ *minsə ≠ *mentu *piru ≠ * • hapax legomenon in Makura no sōshi (1001) oj mi₁zo₂ = mi₁du pi₁ru = pi₁ru • fite-tu kuruma ‘one carriage’ (186) instead of the expected pR *mizo ≠ *medu *piru ≠ *peru fito-tu kuruma Yamatohama midzo ≠ mɨdzɨ çiɾu ≠ ɸɨɾu • comparative evidence shows that the alternation fito- : fite- Yonamine dʑúː ≠ mìdʑíː pˀìɾúː ≠ pʰìɾúː is original and Japanese fito-tu < pi₁to₂-tu must thus be Hirara mdzu ≠ midzɿ pɿː ≠ piɿ due to analogical levelling Ishigaki ńdʑú ≠ mídzɿ̀ pɿ̂ːɾɿ̀ ≠ píɴ̀ pJ *pitə- *pite-tu Yonaguni ndú ≠ míɴ tsˀuː ≠ çìɾú oj pi₁to₂- pi₁to₂-tu Yamatohama tɕˀu- tˀɨː-tsɨ ‘horse’ ‘sea’ ‘mortar’ ‘medicine’ Yonamine tɕu- tˀiː-tɕi pJ *uma ≠ *omi *{u,o}su ≠ *kusori Hirara pɿtu- pɿti-tsɿ oj uma = umi₁ usu = kusuri Ishigaki pɿtu- pitiː-dzɿ pR *uma ≠ *omi *{u,o}su ≠ *kusori Hachijō to- te-tsu Yamatohama mˀaː ≠ ʔumi ʔusɨ ≠ kˀusuɾi Shuri m̀ˀmà ≠ ʔùmì ʔùːsì ≠ kùsùì The relevance of comparison for the description of Hirara mma ≠ im usɿ ≠ fusuɿ Ishigaki ḿmá ≠ íɴ úsɿ́ ≠ ɸùɕíɾɿ́ Yonaguni mmà ≠ ùnnáɡâ ùtɕî ≠ tsˀùɾî (19) It is true that • synchrony and diachrony should be kept apart (15) Accent and tone (see also the presentation by . Igarashi) • speakers do not have knowledge of a language’s history • until recently, the Ryukyuan accent and tone systems were • historical and comparative considerations are always sec- presented as simply deriving from Japanese ondary to synchronic description • however, the Ryukyuan data requires to reconstruct more distinctions than in Japanese (Hattori 1959 [1958]; Hattori (20) However 2018 [1979]), but the reconstruction of the phonetic content • science is not only about observing and describing of those distinctions is problematic (Pellard 2016b) • languages are no more fixed species than biological ones, ‘rain’ ‘shadow’ and “[n]othing in biology makes sense except in the light emj àmê LF kàgê LF of evolution” (Dobzhansky 1973) Kametsu ʔàḿ LH kʰáɡ̀ HL • it is possible, as did ’s Academy, to define as Yoron ʔàmî LF hágì HL “featherless bipeds (with broad flat nails)”, but e.g. “naked Yonamine ʔàmǐː LRː háɡì HL ape” is phylogenetically more correct, and thus far more Shuri ʔàmì LL kàːɡì LːL enlightening Yonaguni àmì LL kʰàŋî LF • explanation often requires reference to history: why is a cat called a cat? why foot : feet or was : were? (16) Ambivalent pronoun in • many fieldworkers refer to prominent figures such as Boas, • the pronoun na can refer to both 1st and 2nd person in Bloomfield, or Sapir as their models, but all of them had OJ(Whitman 1999) a keen interest in historical comparison a. 奈 何 名 能良佐禰 na na no₂-as-ane (21) More concretely 2 NOM name name-HON-DES • historical-comparative evidence can also shed light on ‘I wish you tell me ’ (mys 5.800) competing synchronic analyses and theoretical issues (see e.g. Loporcaro 2007) 名 兄 乃 君 b. • comparative linguistics allows to make predictions from na SE no₂ KI₁MI₁ one (or several) language about what we can expect in 1 beloved GEN lord another related language ‘my beloved lord’ (mys 16.3885, cf. a se ‘id.’ kk 29) • questionnaires, like those used in Japanese accentology, • in Southern Ryukyuan, it is a reflexive or logophoric - which are based on the comparative method : Irabu naɾa, Ishigaki nâː, náɾà

3 Pellard The comparative study of the Japonic languages

• the questionnaire I created for a survey of the Miyako ‘jaw’ ‘dust’ ‘turtle’ ‘pigeon’ ‘thatch’ dialects was based on my reconstruction of proto-Miyako pR *kakuzu *pokori *kame *pato *kaja (Pellard 2009) and was carefully designed to include Yamatohama kʰaxadzɨ ɸuɸuŋ kʰamɨ hatʰo ɡaja covering virtually all in all environments, and it Shodon kʰaxat ɸuɸum kamɨː hatoː ɡajaː made determining the phonological systems of the dialects Kamikatetsu kʰaːdu ɸuːmu hami hatuː ɡaja surveyed rather straightforward (Pellard & Hayashi 2012) Kametsu kʰaːdzɨ hoːmuɴ kamɨ hatu ɡjaː • through a short elicitation list, it is thus possible to de- Wadomari kaːdʑi ɸuːmuː hamiː ɸaːtuː ɡjaː termine the main features of a linguistic system Yoron kaːdʑi puːmu hami patu ɡjaː • of course, one pitfall is to consider the work done when Ie hakˀudzi pʰukˀui haːmi pʰoːtˀu hajaː the questionnaire is over Yonamine hakˀuːdʑi pʰukˀui haːmiː pʰoːtˀuː hajaː Shuri kakudʑi ɸukui kaːmiː hoːtu kaja (22) Monosyllables in Amami Ryukyuan • open monosyllabic words usually have a long (CVː), Northern Ryukyuan but a small class of exceptions has a short vowel (CV) • in Yuwan, the exceptions are not random but confined to *kame > *kaame *kakuzu > *kakazu words with a fortis onset consonant5 *pato > *pa{o,u}to *pokori > ?*pop(p)omu • comparison shows that all exceptions come from earlier disyllables: Okinawa Amami ‘fish’ ‘grass’ ‘rice plant’ time Kamikatetsu ju sa inɪ *kaja > gja Yuwan jˀu kˀusa nˀjɨ Kikai Ōshima Japanese iwo kusa ine

(23) Apocope in Shodon Okinoerabu Yoron • some, but not all, final high vowels of HL-ending dissyl- lables alternate with zero: m̂t : mìd̂- ‘water’, but sùd̂ː : (26) Historical-comparative linguistics and typology sùd̂- ‘sleeve’ • linguistic typology is interested in making cross-linguistic • ɨ < *e does not undergo apocope, but ɨ < *u does generalizations about the relative frequency or rarity of ‘water’ ‘sleeve’ linguistic features Shodon m̂t : mìd̂- sùd̂ː : sùd̂- • problem of sampling and genetic bias Kamikatetsu mìdú súdì • is a feature found in many languages due to a universal oj mi₁du so₁de tendency or to the fact that those languages inherited it? • in other words, did that feature arose only once or several (24) Palatalization in Amuro times independently? • some verbs take palatalized allomorphs of suffixes • for instance, M. Shimoji argues that marked nominative ‘untie’ ‘walk’ alignment is cross-linguistically rare but frequent in Ryuky- pres ɸukkˀjuɾ = akkˀjuɾ uan, but did that pattern arise several times independently, pres.neg ɸukkˀaɴ ≠ akkˀjaɴ in which case an explanation is required, or did it arise imp ɸukkˀɨ ≠ akkˀi only once and was simply inherited, in which case it will • progressive palatalization due to an earlier *i remain an isolated exception? ‘untie’ ‘walk’ Amuro ɸukkˀaɴ akkˀjaɴ Conclusions Yoron ɸutukannu aikannu Japanese podokanu arikanu • much remains to be done within the field of Japonic historical-comparative linguistics (25) Classification is not about pigeonholing or drawing trees • after a century of intense philological work and internal re- • a phylogenetic tree represents sequences of shared devel- construction, I expect most future breakthroughs to come opments from comparison with Ryukyuan, Hachijō and Japanese • in contrast with networks or isogloss maps, it has a time dialects axis and is thus the only visualisation that is historical • historical-comparative linguists need to pay more attention • it needs to be based on shared innovations and to exclude to descriptive work on endangered Japonic languages coincidences and contact (comparative method) • descriptive linguists should not disregard historical- • classifications based on typological features classify North- comparative data nor reduce historical-comparative - ern Okinawan and Southern Amami together on the basis guistics to grammaticalisation theory that they share the features *e > *ɨ > i, *k > h, *p ≯ h • it is widely acknowledged that descriptive grammars • earlier classifications are thus unable to explain why all should be typologically and theoretically informed Amami and all Okinawan varieties share mutually exclusive • I argue that they should also be historically and compar- innovations (Lawrence 2006; Pellard 2015) atively, or more concisely diachronically, informed

5. However, there are also CˀVː words, e.g. tˀɨː ‘one’, and the generalization cannot thus be reversed.

4 Pellard The comparative study of the Japonic languages

Abbreviations and sources Haspelmath, Martin. 2018. What is the name of my subfield (or subcommunity): Language typology, linguistic typology, Amuro (Amami, Uke-jima: Shirata & Shigeno 2016) or comparative linguistics? Diversity linguistics comment: emj Early Middle structures throughout the world. 27 February Hachijō (Hirayama 1992) 2018. https://dlc.hypotheses.org/1022. Hirara (Miyako: Nevskij 1922; Hirayama 1992) Hattori, Shirō. 1959. Nihongo no keitō. Tokyo: Iwanami Ie (Okinawa, Ie-jima: Oshio 1999) Shoten. Irabu (Miyako, Irabu: Tomihama 2013) Hattori, Shirō. 2018. Nihon sogo no saiken. With annotations Ishigaki (Yaeyama, Ishigaki: Miyagi 2003) by Uwano, Zendō. Tokyo: Iwanami Shoten. kk Heinrich, Patrick & Miyara, Shinsho & Shimoji, Michinori Kamikatetsu (Amami, Kikai: Kibe et al. 2011) (eds.). 2015. Handbook of the Ryukyuan languages: Kametsu (Amami, Tokunoshima: Hirayama 1986) History, structure, and use. Berlin: De Gruyter Mouton. mys Man’yōshū https://doi.org/10.1515/9781614511151. Yonamine (Okinawa: Nakasone 1983) Hirako, Tatsuya & Pellard, Thomas. 2013. Hachijōgo no fur- oj Old Japanese usa to atarashisa. In Kibe, Nobuko (ed.), Shōmetsu kiki pIE proto-Indo-European (Rix et al. 2001; Fortson 2009) hōgen no chōsa, hozon no tame no sōgōteki ken- pJ proto-Japonic kyū: Hachijō hōgen chōsa hōkokusho, 47–67. Tachikawa: pR proto-Ryukyuan National Institute for Japanese Language and Linguistics. Shodon (Amami, Kakeroma: Karimata 1995; 1996) http://pj.ninjal.ac.jp/endangered/report/hachijo.html. Shuri (Okinawa: Kokuritsu Kokugo Kenkyūjo 1963) Hirayama, Teruo (ed.). 1986. Amami hōgen kiso goi no Wadomari (Amami, Okinoerabu: Hirayama 1986) kenkyū. Tokyo: Kadokawa Shoten. Yamatohama (Amami, Ōshima: Osada & Suyama 1977) Hirayama, Teruo (ed.). 1992–1994. Gendai Nihongo hōgen Yonaguni (Yamada et al. 2013; 2015; personal fieldnotes) daijiten. 9 vols. Tokyo: Shoin. Yoron (Amami, Yoron: Kiku & Takahashi 2005) Hock, . 1991. Principles of historical linguistics. Ber- Yuwan (Amami, Ōshima: Niinaga 2014) lin: Mouton de Gruyter. https : / / doi . org / 10 . 1515 / 9783110219135. References Kaneda, Akihiro. 2001. Hachijō hōgen dōshi no kiso kenkyū. Tokyo: Kasama Shoin. Bloomfield, Leonard. 1928. A on sound-change. Language Karimata, Shigehisa. 1995. -ken Ōshima-gun 4(2). 99–100. https://doi.org/10.2307/408791. Setouchi-chō Shodon hōgen no fonēmu (jō). Nihon Tōyō Bopp, Franz. 1820. Analytical comparison of the Sanskrit, Bunka Ronshū 1. 1–23. http://ir.lib.u- ryukyu.ac.jp/ Greek, Latin, and Teutonic languages, shewing the original handle/20.500.12000/2290. identity of their grammatical structure. Annals of Oriental Karimata, Shigehisa. 1996. Kagoshima-ken Ōshima-gun Literature 1(1). 1–65. https://books.google.com/books?id= Setouchi-chō Shodon hōgen no fonēmu (ge). Nihon Tōyō EPk9AAAAcAAJ. Bunka Ronshū 2. 1–57. http://ir.lib.u- ryukyu.ac.jp/ Buchi, Éva. 2012. Des bienfaits de l’application de la méthode handle/20.500.12000/2292. comparative à la matière romane: L’exemple de la recon- Keiler, Allan R. 1970. A phonological study of the Indo- struction sémantique. In Vykypěl, Bohumil & Boček, Vít European laryngeals. The Hague: Mouton. https://doi. (eds.), Methods of etymological practice, 105–117. Prague: org/10.1515/9783110812480. Nakladatelství Lidové Noviny. https://www.academia.edu/ Kibe, Nobuko & Kubozono, Haruo & Shimoji, Kayoko & 14435886/Methods_of_Etymological_Practice. Lawrence, Wayne & Matsumori, Akiko & Takeda, Kōko. Chamberlain, Basil Hall. 1895. Essay in aid of a grammar 2011. Shōmetsu kiki hōgen no chōsa, hozon no tame and dictionary of the Luchuan language (Transactions no sōgōteki kenkyū: Kikai-jima hōgen chōsa hōkokusho. of the Asiatic Society of Japan 23, supplement). : Tachikawa: National Institute for Japanese Language and Kelly & Walsh. https://archive.org/details/in.ernet.dli. Linguistics. http://pj.ninjal.ac.jp/endangered/report/ 2015.70910. .html. Chamberlain, Basil Hall. 1897. A preliminary notice of the Kiku, Chiyo & Takahashi, Toshizō. 2005. Yoron hōgen jiten. Luchuan language. The Journal of the Anthropological Tokyo: Musashino Shoin. Institute of Great Britain and Ireland 26. 47–59. http: Kokuritsu Kokugo Kenkyūjo (ed.). 1963. Okinawago jiten. //www.jstor.org/stable/2842331. Tokyo: Ōkurashō Insatsukyoku. Cole, Francis . 1944. A history of comparative anatomy: Kuryłowicz, Jerzy. 1927. ə indo-européen et ḫ hittite. In From Aristotle to the eighteenth century. London: Mac- Taszycki, Witold & Doroszewski, Witold (eds.), Symbolae millan. https://archive.org/details/in.ernet.dli.2015. grammaticae in honorem Ioannis Rozwadowski. Vol. 1, 20232. 95–104. Kraków: Gebethner & Wolff. Dobzhansky, Theodosius. 1973. Nothing in biology makes Lawrence, Wayne P. 2006. Okinawa hōgengun no kai kubun sense except in the light of evolution. Biology ni tsuite. Okinawa Bunka 40(2/100). 101–118. Teacher 35. 125–129. https://doi.org/10.2307/4444260. Lindeman, Fredrik Otto. 1987. Introduction to the ‘ Laryngeal Fortson, Benjamin W. iv. 2009. Indo-European language and Theory’. Oslo: Norwegian University Press. culture: An introduction. 2nd edn. Chichester: Wiley- Loporcaro, Michele. 2007. Facts, theory and dogmas in histor- Blackwell. ical linguistics: Vowel quantity from Latin to Romance. In Salmons, C. & Dubenion-, Shannon (eds.), His-

5 Pellard The comparative study of the Japonic languages

torical linguistics 2005: Selected papers from the 17th In- Pellard, Thomas & Hayashi, Yuka. 2012. Miyako shohōgen no ternational Conference on Historical Linguistics, Madison, on’in: Taikei to hikaku. In Kibe, Nobuko (ed.), Shōmetsu Wisconsin, 31 July - 5 August 2005, 311–336. Amsterdam: kiki hōgen no chōsa, hozon no tame no sōgōteki kenkyū: John Benjamins. https://doi.org/10.1075/cilt.284.23lop. Minami Ryūkyū Miyako hōgen chōsa hōkokusho, 13–51. Meillet, Antoine. 1925. La méthode comparative en lin- Tachikawa: National Institute for Japanese Language and guistique historique. Oslo: H. Aschehoug. http://gallica. Linguistics. http://pj.ninjal.ac.jp/endangered/report/ bnf.fr/ark:/12148/bpt6k934150p. miyako.html. Miyagi, Shin’yū. 2003. Ishigaki hōgen jiten. 2 vols. Naha: Rix, Helmut & Kümmel, Martin & Zehnder, Thomas & Lipp, Okinawa Taimusu. Reiner & Schirmer, Brigitte. 2001. Lexikon der indoger- Morpurgo Davies, . 1998. Nineteenth-century linguistics manischen Verben: Die Wurzeln und ihre Primärstamm- (History of Linguistics 4). London: Longman. bildungen. Wiesbaden: Reichert. Nakasone, Seizen. 1983. Okinawa Nakijin hōgen jiten. Tokyo: Saussure, Ferdinand de. 1879. Mémoire sur le système prim- Kadokawa Shoten. itif des voyelles dans les langues indo-européennes. Nevskij, Nikolai A. c. 1922–1928. Materialy dlja izučenija go- Leipzig: B. G. Teubner. https : / / gallica . bnf . fr / ark : vora ostrovov Mijako. Edited in Jarosz, Aleksandra, 2015, /12148/bpt6k729200. Nikolay Nevskiy’s Miyakoan dictionary: Reconstruction Schleicher, August. 1861. Kurzer Abriss einer Lautlere der from the manuscript and its ethnolinguistic analysis, indogermanischen Ursprache, des Altindischen (Sanskrit), Poznań: Adam Mickiewicz University, (doctoral disserta- Alteranischen (Altbaktrischen), Altgriechischen, Altitalis- tion), http://hdl.handle.net/10593/14699. chen (Lateinischen, Umbrischen, Oskischen), Altkeltischen Niinaga, Yuto. 2014. A grammar of Yuwan, a northern (Altirischen), Altslawischen (Altbulgarischen). 2 vols. Wei- Ryukyuan language. Tokyo: University of Tokyo. (Doctoral mar: Hermann Böhlau. http://www.deutschestextarchiv. dissertation). http://hdl.handle.net/2261/59732. de/schleicher_indogermanische01_1861. Osada, Suma & Suyama, Nahoko. 1977–1980. Amami hōgen Shimoji, Michinori & Pellard, Thomas (eds.). 2010. An in- bunrui jiten. 2 vols. Tokyo: Kasama Shoin. troduction to Ryukyuan languages. Tokyo: Research In- Oshio, Mutsuko. 1999. Okinawa Ie-jima hōgen jiten. 2 vols. stitute for Languages and Cultures of Asia and Africa. Ie: Ie-son Kyōiku Iinkai. http://lingdy.aacore.jp/jp/material/An_introduction_to_ Pellard, Thomas. 2009. Ōgami: Éléments de description Ryukyuan_languages.pdf. d’un parler du Sud des Ryūkyū. Paris: École des hautes Shirata, Rihito & Shigeno, Hiromi. 2016. Kita Ryūkyū études en sciences sociales. (Doctoral dissertation). https: Amami Uke-jima Amuro hōgen no on’in suketchi: Kei- //tel.archives-ouvertes.fr/tel-00444150. tai on’inronteki kōtai o chūshin ni. Ryūkyū no Hōgen 41. Pellard, Thomas. 2011. The historical position of the Ryukyuan 165–204. http://hdl.handle.net/10114/13850. languages. In Historical linguistics in the Asia-Pacific Thorpe, Maner L. 1983. Ryūkyūan language history. Los region and the position of Japanese, 55–64. : Angeles: University of Southern . (Doctoral dis- National Museum of Ethnology. https : / / hal . archives - sertation). http : / / digitallibrary . usc . edu / cdm / ref / ouvertes.fr/hal-01682958. collection/p15799coll3/id/505374. Pellard, Thomas. 2013. Nihon rettō no gengo no tayōsei: Tomihama, Sadayoshi. 2013. Miyako Irabu hōgen jiten. Naha: Ryūkyū shogo o chūshin ni. In Takubo, Yukinori (ed.), Okinawa Taimusu. Ryūkyū rettō no gengo to bunka: Sono kiroku to Watkins, Calvert. 1973. Language and its history. Dædalus keishō, 81–92. Tokyo: Kuroshio Shuppan. 102(3). 99–111. https://www.jstor.org/stable/20024148. Pellard, Thomas. 2015. The linguistic archeology of the Ryukyu Whitman, John. 1999. shift in Japanese: A Islands. In Heinrich, Patrick & Miyara, Shinsho & Shimoji, case study in lexical change and point of view. In Kamio, Michinori (eds.), Handbook of the Ryukyuan languages: Akio & Takami, Ken-ichi (eds.), Function and structure: History, structure, and use, 13–37. Berlin: De Gruyter In honor of Susumu Kuno, 357–386. Amsterdam: John Mouton. https://doi.org/10.1515/9781614511151.13. Benjamins. https://doi.org/10.1075/pbns.59.16whi. Pellard, Thomas. 2016a. Nichiryū sogo no bunki nendai. In Yamada, Masahiro & Pellard, Thomas & Shimoji, Michinori. Takubo, Yukinori & Whitman, John & Hirako, Tatsuya 2013. Dunan (Yonaguni)-go no ’i bunpō to shizen danwa (eds.), Ryūkyū shogo to Kodai Nihongo: Nichiryū sogo shiryō. In Takubo, Yukinori (ed.), Ryūkyū rettō no gengo no saiken ni mukete, 99–124. Tokyo: Kuroshio Shuppan. to bunka: Sono kiroku to keishō, 291–324. Tokyo: Kuroshio Pellard, Thomas. 2016b. Typological and historical- Shuppan. comparative perspectives on tone and in Yamada, Masahiro & Pellard, Thomas & Shimoji, Michinori. Ryukyuan. (Paper presented at Japanese and Korean ac- 2015. Dunan grammar (Yonaguni Ryukyuan). In Heinrich, cent: Diachrony, reconstruction, and typology, Tokyo. Patrick & Miyara, Shinsho & Shimoji, Michinori (eds.), 2–3 July 2016). https://hal.archives- ouvertes.fr/hal- Handbook of the Ryukyuan languages: History, structure, 01416041. and use, 449–478. Berlin: De Gruyter Mouton. https : Pellard, Thomas. 2019. Ryukyuan and the reconstruction of //doi.org/10.1515/9781614511151.449. proto-Japanese-Ryukyuan. In Frellesvig, Bjarke & Kinsui, Satoshi & Whitman, John (eds.), Handbook of Japan- ese historical linguistics. Berlin: De Gruyter Mouton. http : / / www . academia . edu / 35301776 / Ryukyuan _ and _ the _ reconstruction _ of _ proto - Japanese - Ryukyuan. (Forthcom- ing).

6