THE CHANGE FROM SURD TO SONANT IN JAPANESE COMPOUNDS.

BY BENJAMIN SMITH LYMAN.

 THE main object of this paper is to place on record in detail the more important facts at the base of certain euphonic rules briefly given in the short published abstract of a paper of mine on "The Japanese Nigori of Composition," read before the American Oriental Society in 1883.

 At the beginning of the second part of very many Japanese compound words the surds ch, f, k, s, sh and t are changed to sonants. The Japanese call a sonant the nigori, that is, the turbid, or impure form, of its corresponding surd. They have at times even insisted that all the sonant consonants of the purely Japanese part of the language are only derived from surds; and, although that has seemed impossible to some foreigners, on account of the occurrence of sonants at the beginning of many apparently simple words, we shall see, in the light of some cases at least, the Japanese view is not so wholly inconceivable.

 It has sometimes seemed to European students of Japanese that the nigori of composition was as inexplicable as it appears to be in our words hurdy-gurdy, hurly-burly and the like, or that it was a mere matter of the ear, and might be used or not at will. But it will be found that its use depends on the meaning instead of wholly on the ear, and that the Japanese do not, like foreigners, use it indifferently or drop it. In some cases, however, both forms may be allowable, according to difference of meaning or derivation.

 The rule in general for purely Japanese words is that the second part of a compound word takes the nigori; that is, if beginning with ch, f, h, k, s, sh or t, those consonants are changed to the corresponding sonant ones ; yet with only a slight preponderance, about 2361 cases against about 2316; and the general rule does not apply : (1) when b, d, g, j, p or z already occurs anywhere in the second part of the compound ; nor (2) when the second part is a Chinese word; nor (3) where the word, though given by Hepburn as a compound, is really made up of words in regular grammatical connection ( without ellipsis ), such as juxtaposed verbal forms, or Chinese words followed by verbal forms denoting doing or action (shi, suru, and the like), or words connected by no or followed by to, te, or any of the syllables used for the terminations of verbal forms ; and (4) there are 1000 other cases where the nigori is not taken against 2220 where it is, or one case out of three.

 It is not probably worth while to record here the very numerous words that conform to these general and special rules, but only the much less bulky lists of exceptions to them. The rules are based on a review, made sixteen years ago, of all the words in Hepburn's dictionary, second edition, and some two or three hundred more, in all about 23,000 words; and though an oversight here and there may have taken place, and though his third edition may have added further material, it is hoped that the present results may be exact enough for practical purposes.

 In reading the lists it is to be borne in mind that under the general rule h, as representing an ancient surd labial, is changed to b, or sometimes to p, "half nigori."-Hu, instead of fu, would correctly give the pronunciation of Tokio; but at Kiyoto the sound is really fu, with the f exactly like the English f; and Kiyoto, from its central situation and other circumstances, rightly gives the standard for the language in general.- In transliterating ( not anglicising), oo (like the other vowels) is not used with the same force as most often in English, but to represent two successive, yet not audibly separated, long o's, as each would commonly be called, much like oo in oolite, oolitic, oological, zoological, zoophyte. Such a mode of writing the sound, so far from being an innovation, as some have considered it, is as old as any systematic rule of Japanese transliteration, and was explicity adopted about two hundred years ago by Kaempfer, and has been in use ever since.-In the lists of exceptions a dash is used to save repetition of the corresponding part of the preceding word.

 1.-B, d, g, j, p, or z in the next syllable (363 cases), or any following one (35 in all 398 cases), prevents the nigori. The only exception is amagappa.

 A sonant in the syllable before has no effect on the nigori (about 150 words with, and about 150 without).

 2.-Compounds with the final part Chinese do not take the nigori in about 2090 cases (besides 81 cases where a following nigori would have prevented at any rate); but in 287(about one case in seven)it is taken, namely:

 (a)-Where immediately preceded by the letter n, in the following 186 cases:

 (aa)-All those (131, and excepting one?-zenhai, which also has zempai,) in which n in the first part of the compound comes before h or f in the second, of which 120 change nh or nf to mp(half nigori), against the 11 following, which change nh or nf to mb : Jimben, mam-, nim-, sam-(4);sam_biyaku, -bon (2); ham_bitsu, hombuku, imban, kembeki, membaku(5).

 (ab)-And the following 55, in which a surd consonant following n takes the nigori (against 515 in which it does not ): Jin_dzuu, yuu- (2); han_goku, hon-, kin-, ou-, ran-, rin-, san-, sen- (8);en_ja, han-, in-, kan-, sen-, shin- (6); ban_jaku, en-, on-, ren-, san-, tan- (6); ren_ji, zen- (2); baken_jo, kan-, kin-, nan-, shin- (5); nan_zan, rin-, san- (3); en_doo, -gi (2); han_dan, -doo, -zatsu (3); sandzui, -gai, -jiki, -zai, -zashi, -ze, -zen (7); sen_zankoo, -zen (2); bushinjin, inju, konjiki, manzai,nenjin, shinzoo, tenden, unjiukitsu,yunzei(9).

 (b)-And the following 106 cases : Do_bei, ishi-, ita-, neri- (4); ashi_biyooshi,ita-, ma-, shira-, te- (5); ue_boosoo, uma-, ushi- (3); go_buku, imi-, ki- (3);cha_dansu,choo- (2);ishi_dooroo, mawari-, taka-, tsuri- (4); boo_dzu, joo- (2);kakure_ga, me-, utsuri-, waki- (4); oyoko_gi, utsuri-na, yowa- (3); Ei_goku, Futsu-, riyoo- (3); kuchi-girei, te- (2); cha_guwashi, hi-, midzu- (3); annai_ja, choo-, moo-, ninsoo-, shugiyoo-, uranai- (6); doo_ji, e-, hana-,hashiri-, hei-, too- (6); e_jiki, kotsu-, moku-, ni-, niku-, so- (6); bareki_jin, sadai-, sui-, ubai-, yoo- (5); kawai_jo, niroku- (2); gin_zaiku, mugiwara-, te- (3); kake_zan, kuwa-, menoko-, muna-yoo, nagare-, sa-, tatami-, wari- (8); hei_zei, fu-, oo-, sei- (4); atsugan, chiwagenka, doozen, gobatsu, funagassen, fuuzetsu, giyodzui, hatsugoori, hayabikiyaku, koogaku, kajichi, katsudatsu, midzujaku, nezoo, otamajakushi, saguwan, shigedoo, soodegooro, tooguwa, usugeshoo, yakiban, yasejotai, yudoofu(23).

 3.-About 670 cases given by Hepburn as compound verbs do not take the nigori (besides 148 similar cases where it would be prevented at any rate by a following nigori consonant), but in the following 35 cases it is taken, namely: Aomi_dachi, hooke-, tsure- (3); mamoro_dooshi, yomi- (2); ike_dori, tsukami- (2); name_dzuri, sae- (2); karon_ji, sakin-, uton-, yasun- (4); shi_bari, -bori, -buri, -bomi, -dare, -goki, -gumi, gure (8); fum_bari, -bataraki (2); degire, iregomi, kikigane, kuribiki, nezame, nibami, oibore, sashigumi,saegiri, tsuibami, ukegai, yasegare(12).

 The following 99 words, given by Hepburn as nouns, of which both parts are verbal, take the nigori(against 96 that do not): Otoshi_banashi, tatoe-, yari- (3);ai_bore, ne- (2); sukashi_bori, uki- (2); (aomi_dachi)are-, suki- (2); ki_dooshi, kiri- (2); kiri_dori, kogiri-ni, oshi-, tsukuri-, uri- (5); sashi_dzume, tashi- (2); baitori_gachi, kane-, itsuwari-, okitari-, wasure- (5); kake_gae, nori- (2); ate_gai, oshi- (2); kaeri_gake, kai-, ki-, nuke-, omoi-ni, tasshi-, tomari-, toori-, yuki- (9); furi_gaki, hashiri-, hikae-, kiki-, misebi-, nijiri-, nuki-, soe-, tsumori-, wari- (10); (ire_gomi),ki-, ue- (2); hanare_jini, kubire, obore, tachi-, ue-, yake- (6); mi_zame, ne- (2); maki_zoe, sashi- (2); de_zome, kaki-, nori- (3); hanarebanare, harebare, karegare, kiregire, shimijimito, taedaeni (6); akegure, aibiki, hanarezakari, kakeberi, kakedzukuri, kaigakari, kaigui, kashidzuki, kiribari, machibuse, makigari, midate, mikakedaoshi, namege, nebie, neboke, nedzumai, negaeri, nurigome, okurebuse, okizari, soibushi, tachigare, tachigiki, tachigie, tachigurami, tachigiri, tatakibarai, uttegawashini, waidame, yoigurui, yukidomari(32).

 The following 31 cases of Chinese words followed by shi or suru take the nigori: Benji(dzuru), danji(dzuru), enji(dzuru), gaenji(dzuru), genji(dzuru), hanji(ru,zuru), henji, junji(ru,dzuru), kenji(dzuru), kunji(ru,dzuru), menji(ru,dzuru), nenji(dzuru), ninji(dzuru), ronji(dzuru), sanji(ru,dzuru), senji(dzuru), shinji(dzuru), sonji(dzuru), soranji(dzuru), tanji(dzuru), tenji(dzuru), zonji(dzuru), (22 ending in n); chooji(dzuru), dooji(dzuru), hooji(dzuru), jooji(dzuru), kooji(dzuru), ooji(dzuru), shooji(ru,dzuru), tooji(dzuru), (8 ending in oo); ei_ji,(1).

 The following 11 words compounded with Chinese ones ending in tsu and the verbal ending shi(suru) do not take the nigori: Besshite, esshi, kesshi(shite), kusshi, resshi, sesshi, sosshi, tasshi, tesshi, usshi, zesshi. Also gese and geshi do not take the nigori. Other Chinese words followed by shi(suru) are not given as compounds, and are not followed by the nigori.

 In about 151 other cases which, though given by Hepburn as compounds, are really words in grammatical connection without ellipsis or contraction, there is no nigori of composition. The six apparent exceptions are : Amanogawa (of which, however, no=prairie?),michinobe, nanigana, osoiba, sainogawara, unabara(for "umi no hara").

 Of so-called verbal terminations, the change from a surd to the nigori occurs in:Ba,in the so-called conjunctive and conditional forms; do and domo, in concessive ones; de, dzu, ji, zaru, in negative ones; de, in affirmative ones where the root ends in gi, and g is dropped in contraction, or where mi at the end of the root is changed to n.

 4.-The following 1000 compounds do not take the nigori (against 2220 that do):

 (a)-353 with verbal endings (against 681 that do take the nigori):

Charumera_fuki, furo-, hai-, hora-, kane-, midzu-, sorauso-, (7); ame_furi, hire-, (2); ei_fushi, hire-, (2); chiri_harai, kushi-, tsuchi-, yaku-, (4); kasa_hari, joo-nokami, taiko-, (3); ami_hiki, edzu-, fune-michi, ha-, midzu-, momo-, mosa-, yado-, (8); ido_hori, kane-, (2); midzu_kai, tsuchi-, ushi-, yak-, (4); fude_kake, hara-, hashi-, katana-, koshi-, mae-, me-, midzu-, rou-, te-, sudzu-, yari-, yodare-, (13); e_kaki, hanshita-, hi-, kago-, kai-, kasa-, koshi-, mae-, masu-, meso-, mimi-, mono-, sumi-, te-, to-, beso_kaku, (16); kugi_kakushi, me-, (2); hana_kami, oo-, yak-, (3); me_kari, midzu-, (2); cha_kashi, gura-, kane-, me-, (4); hi_keshi, sumi-, (2); kuchi_kiki, me-, te-, (3); choo_kiri, en-, ishi-, kama-, kichak-, kubi-, soba-, shin-, yajiri-, (9); cha_koshi, midzu-, toshi-, (3); miru_kui, mono-, mushi-, ki-mushi, (4); ara_kure, chobo-, kai-, nani-, o-, saka-, shiraba-, ta-, (8); ito_kuri, kara-, ta-, wata-, (4); ei_same, haru-, me-, mura-, (4); abura_sashi, bin-, e-, fuda-, hata-, midzu-, mono-, sumi-, tatami-, tori-, zeni-, (11); tadzu_sawari, yu-, (2); abumi_shi, e-, fude-, gura-, hata-, ikada-, ikake-, imono-, ireba-, kagami-, kawara-, kazari-, koshaku-, koto-, kusu-, kuji-, makie-, megane-, nage-, nani-ni, nani-ka, nani-oo, nurimono-, sashimono-, sato-(se), shiru-, sora-, sugo-, yatsu-, makoto-yakani, tai-ta, nami_suru, (32); ato_shiki, kana-, kata-, kore-, kura-, naga-, utto-, ya-, za-, (9); abura_shime, haji-, karo-, midzu-, obi-, soo-, yama-, (7)

; mono_shiri , us-, soo_shiranukao, (3); dara_suke, darani-, fuku-, kumo-, san-, (5); goma_suri, han-, ko-, mimi-, te-, (5); kara_tachi, kit-, kunitoko-, mono-, (4); shiro_tae, uro-, ut-, yoko-, (4); hi_taki, meshi-, (2); hana_tare, shio-, shita-, (3); hachi_tataki, ishi-, ma-, niwa-, shiba-, (5); fude_tate, me-, ya-, (3); shito_tome, sode-, (2); akari_tori, aka-, amma-, ase-, ato-, chiri-, hiyoo-, kaji-, koi-, kuchi-, me-, nomi-, o-, ondo-, sai-, sao-, seki-, shaku-(mushi), shi-, sumi-, sumoo-, tema-, yu-, zoo-, midzutoru-tama, toshitotta, (26); boo_tsukai, hebi-, idzuna-, sora-, (4); bin_tsuke, hada-, hi-, kado-, kako-, kane-, ishi-, jin-, ki-, me-, muku-, ne-, shimo-, te-, (14); aka_tsuki, basa-, beta-, biku-, bira-, biri-, bura-, chira-, fu-ai, fuda-, fura-, giro-, gota-, gura-, gudo-, gidzu-, guta-, hiyoro-, iki-, ira-, jara-, ji-, kabi-, kidzu-, kira-, kitsu-, kiyoro-, kome-, kose-, maga-, me-, na-, nawa-, nicha-, nura-, oyoko-, seka-, sen-, set-, soko-, sowa-, ta-, teratsu-, uka-, uro-, uwa-, kentsuku, shaa_tsuku, (48); hana_tsukuri, niwa-, yumi-, (3);

cha_tsumi, na-, (2); eishire, eitaire, etoki, fusoroi, futemawari, futsuriai, asakaranu, hanahiri, hoofukurashi, hookamuri, iwotsuri, karisome, kikori, komekami, kotokire, kotokawari, kotosaranu, kototari, kubikukuri, kuchisui, mekuramashi, midzusumashi, midzutamari, midzutame, miotsukushi, mukabaratatsu, nedzumikoroshi, netsusamashi, omohoe, saikaeri, sayofuke, shiohi, shirake, shirokae, shitashimi, shitatame, tadzusae, takumi, takuromi, takuwae, tasuke, tasukari, tekihaki, tesuki, tokoroseki, yatsure, yokotawari, yoosuki, yuusuki, yuusari, (50).

 (b)-83 reduplicated words (against 67 with the nigori):

chikuchiku, chirachira, chirichiri, chirochiro, chokochoko, furafura, fuwafuwa, hakihaki, haraharato, hatahata, hekoheko, hetahetato, hihi, hirahirato, hirihiri, hitahita, hiyokohiyoko, hiyorohiyoro, hokohoko, horohoro, hotehote, hotohoto, kachikachi, kakukaku, karakara(to), katakata, kechikechi, kirakirato, kirikiri(to), kiyakiya, kiyorokiyoroto, kokekoke, korokoroto, kosekose, kosokoso, kotekote, konkonto, kurakura, kurukuruto, kushakusha, kusukusuto, kutsukutsuwarau, kuyokuyo, sakusakuto, sarasara(to), sashitsumesashitsume, satemosatemo, satesate, sawasawato, saetsuosaetsu, sekaseka, sekiseki, sewasewashii, shaashaa, sharisharito, shikashika, shikushiku, shioshioto, shitoshito, sokosokoni, sokusoku, somosomo, soresore, sorosoroto, soosoo, sowasowa(shite), soyosoyoto, surasurato, surusuruto, suyasuyatoneru, takatakayubi, takatakatsuki, tamatama(tamadama), taratarato, teratera, teriteriboodzu, torotoroto, tonton, tootoo, tsukatsuka, tsuratsura, tsurutsuru, tsuyatsuya.

 (c)-34 compounds with adjective endings(against 106 that do take the nigori):

akarui, anakashiki, aoshiroi, arakuroshii, aramahoshii, furukusai, futokutakamashiki, hashikashii, hinatakusai, ikikusai, ikuhisashii, jimankusai, kashikamashii, katakurushii, kirakirashii, kogarekusai, mimahoshii, mimishii, mudzukashii, musakurushii, semahoshii, shibutoi, shiohayui, shiokarai, sharakusai, shisomanai. shitsukoi, tattoi, tootoi, tsumetai, utsukushii, wakawakashii, yofukai, yuyushii, besides others compounded with mahoshii, shii, tai, and toi, which do not appear as separate words.

 (d)-29 juxtaposed words of allied or contrasted meaning:

achikochi, anakashiko, atosaki, hirarikururito, iroka, itokake, kagehinata, kakasoso, kakute, karekore, muchakucha, musakusa, norarikurari, norakurato, oyako, sakoso, sosokusato, tokaku, tokoo, tomokakumo, tomokoomo, tonikakuni, toosamakoosama, tosenkakusen, toyakakuto, toyakooto, unekune, ushitora, uwoosawoo.

 (e)-Also the following 501 words (against 1366 with the nigori):

a_chi, ko-, nama-, shira-, so-, idzu-, (6); haya_fune, hiki-, kawa-, yo-, (4); de-ha, ori-, saka-, shira-, yudzuru-, (5); naga_hama, shio-, yoko-, yoshi(دد), (4); aka_hara, ato-, hi-, kata-, name-, suki-, ura-, (7); kata_hashi, me-, (2); kiza_hashi, mi-, sori-, (3); iri_hi, tobi-, (2); kumi_himo, uchi-, (2); ma_ho, midzu-, tsugi-, (3); hanashi_ka, hoshi-, (2); kawa_kami, kaza-, kome-, (3); kiri_kami, ori-, shibu-, (3); furu_kane, shiro-, midzu-, (3); ashi_kase, kubi-, maro-, te-, (4); aburakasu,cha-, soba-, tabe-, tare-, (5); ai_kata, ara-, ato-, se-, fuchi-, funa-, haha-, hake-, hiyooro-, hisa-, idzu-, kari-, kashi-, koshi-, kure-, kawase-, mae-, me-, mi-, moto-, ni-, mochii-, ori-, oya-, sabake-, saki-, sato-, sen-, shiire-, shi-, shitate-, shite-, tana-, tsukai-, uchi-, uma-, ura-, ya-, yu-, yuu- (40); nari_katachi, shina- (2); abura_kawa, atsu-dzura, kata-, ni-, oo-, shibu-, togi-, totsu-, tsukuri-, usu-, uwa- (11); abura_ke, ara-nai, chiri-, hata-, iro-, kawara-, koshi-, midzu-, mukai-, mushi-, nebari-, nigo-, nodo-, oomi-, saku-, shiru-, sori-, tawa-, tsuyu-ki, ubu-, yata- (21); chi_kemuri, midzu-, uma- (3); kabu_ki, karasu-, kare-, koshi-, kuchi-, kusu-, ma-, maru-, masa- nadzu-, nama-, nami-, saka-, shira-, taru- tori-, tsugi-, ubu-, ue-, waka-, yak- (21);

; arai_ko, arashi-, asu-, dada-, funa-, hari-, haru-, iri-, ishi-, ko-, kumi-, kushi-, mai-, mama-, midzu-, mi-, migaki-, moro-, nama-, ne-, nicha-, nuna-, obo-, shiro-, shiru-, so-, tana-, te-, tera-ya, tori-, tsure-, udon-, uji-, uro-, yak- (35); hiki_koto, kata-, mi-, tawa-, uwa-, wabi- (6); ai_kuchi, de-, ho-, iri-, karu-, kata-, mitsu-, mochi-, muki-, oo-, ore-, sabake-, sode-, tobo-, ure-, uri-, waru-, yatsu-, yoi-, yomi- (20); ashitaka_kumo, mura-, shira-, yami- (4); kami_kura, kari-, nama- (3); haya_kusa, kara-, midzu-, mi-, omo-, saki-, shichi-, some-, to-, ume-, yake- (11); kuchi_kuse, shi-, te- (3); hana_kuso, kani-, kana-, me-, mimi-, mune-, mushi- (7); abumi_kuwa, kuro- (2); aka_saka, ko- (and kozaka), kudari-, nobori-, tama-, to- (6); akari_saki, he-, hoko-, kuchi-, mi-, muna-, te-, toto-, uri-, ya-, yoo- (11); ari_sama, akara-ni, ashi-ni, ika-, midai-, mina-, nani-, ne-, ni-, noke-ni, oku-, saka-, saki-, tono-, too-koo-, toto-, yoko-, nesan, nisan, obaasan, okamisan, ototsan (22); hi_sao, kara- (2); ima_sara, nao- (2); furu_sato, tori- (2); asa_se, fuka-, hada-, kugu- (4); saka_shima, te-ishi, yoko-, (3); chi_shio, ha-, hi-, hiki-, michi-, sashi- (6); kawa_shiri, mayu- (2); kawa_shimo, kaza- (2); kata_shiro, nawa-, toji-, uri- (4); chi_shiru, hana- (2); kaze_shita, me-, obi- (3); ami_so, nanori- (2); ao_ta, ara-, are-, fuke-, kawa-, midzu- (6); ko_tachi, kodomo-, kunitoko-, nan-, omae-, yakunin- (6); ara_kata, kuma-, ashi-kumo (3);

; ari_take (and aridake), hana-, hatsu-(and hatsudake), iwa-, kawa-, kure-, matsu-(and matsudake), mimi-, shii- (9); ara_tama, kin-, kuro-, kubi-, midzu- (5); kakobi_te, hama-, hiki-, hineri-, hon-, ho-, i-, kai-, kara_me-, kara-, kata-, kawariban-, kiri-, kit-, ko-, me-, naka-, nawa-, oi-, oku-, oo-, saka-, saki-, sawa-, sen-, shimo-, shita-, shi-, sho-, tori-, tsukai-, tsuri-, uri-, uwa-, yaki-, yari-, yose- (37); ao_to, e-, mune- (3); kana_toko, niwa- (2); ko_tori, niwa-, oo- (3); ma_tsuchi, masa-, neba-, yase- (4); mu_tsuki, shimo-, sa- (3); kiba_tsutsu, ko-, moto_gome-, o-, oo-, tan- (6); han_shita, -toki (2); kara_kami, -kane, -kusa(-sao), -sumi (4); kata_ho, -kana, -sumi, -toki (4); ko_sawa, (-saka and -zaka) (1); mama_chichi, -haha (-ko), -samurai (3); mi_hakase (-hashi), (-kata), (oo-ke), -koshi (koto), -sora, -takara, -tama (iki-tama), -tarashi, -toohoo, takimikura (8); o_fukuro, -hayoo, -hari, -hiya, -hiyarakshi, -hie, -kan, -ketsu, -tamaya, -tori (totsan), (-tsutsu), -tsuyu (11); (oo_kawa), -kimi (-kuchi), -kurashoo, -sawa (3);

; aburahi, aohiki, aosora, aoto, asahaka, edaha, fusasakura, hakoromo, hanafuyu, inukoro, irotsuya, i(h)e, kamashika, kamisakayaki, katatsumuri, kirikishi, marutoshi, mekao, morotomoni, muneto, narisoo, norikumi, okusokonai, orifushi, orihima, ototoi, ototsui, raiharu, sahachi, sahari, satsuhito, shookachi, wakatono, dzukuni, jisaka, midzukame, midzusaki, midzuseki, sabitsue, shattsura, shinobitsuma, shiotsu, shiratsuya, tobihi, tookarasu, uminechima, ubusuna, yabuka, yobikoe, yohoro, yubukarashi, yumahiko, yurumekusuri(53).



 If the complete lists of compounds with the nigori and without be carefully examined, it is found that:

When the first part indicates the origin, source, cause or the like, possession or ownership, superiority, prevalence, pervasion, inclusion(either physical or ideal or a classifying feature) of the second part, in short domination over it as a subordinate thing, there is no nigori of composition. These are the very qualities possessed in English by a substantive following the word of, as compared with the one that precedes.

 But when those qualities are rather possessed by the following part of the compound,of which the first part indicates a subordinate or a more or less imperfectly, partially, superficially, temporarily, occasionally applying characteristic or feature, there is nigori. When, for example, the nigori compound has an adjective ending, the first part shows in what respect the quality is meant; and when both parts are verbal forms, the first likewise shows with reference to what the being something else to which both actions concomitantly refer.

 It is clear that the nigori invariably arises from the disappearance of a sonant consonant, almost always an n, and generally the word no(of), but sometimes ni(in, to, especially in re-duplicated words), sometimes the negative n, and sometimes other sonants or syllables, as perhaps occasionally de(at or with), which appears to be on the same principle a contraction either of nite(with, by, in) or of motte(having). It can now be understood why the sound n is so often heard in colloquial and rustic Japanese before a dental nigori and m before a labial one, and still oftener the sound ng instead of simple g. The significance of such sounds is a very strong argument for specially marking them in any system of transliteration in Roman letters; and for writing, say, nangasaki,in the time-honored European way, instead of the recent Nagasaki. The very existence of the argument, too, is proof that investigations like the present one, though seeming perhaps remote and trivial, may nevertheless have useful bearings upon a question of such pressing importance as the best method of adapting our alphabet to the use of Japanese.

 It is probable that some of the Japanese themselves are not altogether conscious of any difference in meaning, owing to the presence or absence of the nigori of composition, or disregard it on account of inability to explain it or formulate it. At any rate the famous spot for the manufacture of porcelain called generally by the Japanese Kudani(taht is, not nine valleys, as some one has mistakenly imagined, but ku-no-tani, or ninth valley, corresponding to the uncontracted ichi-no-tani, first valley, and several other numbered small valleys that with it branch out of a single large one) is in the neighborhood itself called Kutani, without the nigori. It seems to be an illustration of fact that the attempts of the partially informed to carry out what they conceive to be grammatical rules, are often less correct than the unquestioning instinct if the wholly ignorant.

 The real significance and character of the word no, of such extremely frequent occurrence, is of some interest and consequence. It appears to be the last syllable of the last part of the word mono(thing); for in Japanese not merely is the last part of a word dropped in derivation, as in many western languages, but it is very common that the first part is dropped ; as Mr. E.M.Satow has also remarked (Trans. As. Soc. Jap., VI, 472). The form no is very often used after adjective and verbal forms (frequently contracted to simple n), with obviously the same meaning as mono(thing). It is plain that in the form of the postposition no(of) it has in reality the same meaning(thing), and helps to carry out the universal Japanese rule of letting general words precede the particular. If a subordinate feature has to precede, it is brought about through the interposition of the word no(that is, mono, itself a particularizing word in reference to the foregoing one), in order to make the expression so general that the otherwise principal word may follow as a subordinate, or a possession, or a limiting or defining word. This corresponds well with common idioms in so distant a language as Chinese, and supports the view that even in western language the likewise essentially the same meaning(thing).

 The rule of the nigori in composition helps very much towards tracing the derivation and primitive meaning if many Japanese words. For example, terashima would be an island belonging to a temple on it. Akindo(trader) is akinai no hito(man of trade); shirooto(one not skilled in a profession) is shiro-hito(man of whiteness); while kurooto(one skilled in a profession) is kuro-hito(man of blackness). But kuromboo(negro)is perhaps kuro no hito (a man that has become black or tanned); and likewise akamboo(baby) is aka no hito(red man, but not permanently or fully so); and shiwamboo(miser) is shiwa no hito. It should be remembered that the Japanese h in these cases is to be reckoned as a labial. The last syllable of karriji(return journey), kawaji(river road), mikkaji(three day's journey), and kooji(small streets) is clearly michi(road). The first part of kadzu or koodzu (the paper mulberry) is apparently derived from kami(paper). Koodzuke, the same of a province, is evidently Kami-tsuke(this kami meaning upper),correspomding to Shimotsuke(shimo meaning lower), without the nigori. Koobe,the name of a town, would be Kami-he (upper place or dwelling). Oozaka, the name of the great city, is ooki na saka(the great steep-road); whereas Oosaka,as it is often called, would be Ooki saka, nearly the same in meaning, but perhaps differing in the degree of emphasis. The monosyllable ga, pronounced nga,may be derived from no ka, with the ka meaning emanation. 

Ga, like ji from michi, also given as a separate world, and like de, already mentioned, is an instance where the nigori begins a word ; and it seems not wholly impossible that all the comparatively few cases where purely Japanese words so begin might have some similar explanation, and that the other cases of nigori, in the middle of a word, may have arisen from compounding.

 The word hidari(left hand), often hindari in the country, appears to be the direction of the sunrise, hi no detari, while migi(right hand), often in the coutry migiri, is possibly miru no o kiri, or miru n' kiri, the direction of the cutting(kiri) off of seeing (miru no), or sunset ; or from mi kagiri (the august setting, or the god's setting). The derivation that has been proposed (As. Soc. Jap., VI,473) from nigiri, to grasp, is rather impossible ; for besides the difficulty of changing n to m in such a case, the word nigiri as a concrete substantive applies to the part of the bow that is grasped, and that with the left hand. The words for left and right in Japan appear, then, to be derived from the position of the sunrise and sunset, with reference to the favorite and ordinary outlook of dwellings there. This would seem to suggest a reasonable and natural explanation why in India the South is reckoned to be on the right hand ; not by any worship of the rising sun, such as exists even in Japan, but by the fact, discovered with little camping experience in those tropics, that tents or other dwellings, whenever possible, are made to look towards the east, so as to have the rising sun take off the morning chill, and to be in the shade the rest of the day. It seems to be one of those cases where points in one language are made clear by the investigation of another very distant one.

 It is certain that a through collation of what may seem very dry Japanese grammatical facts, aside from mere euphonic changes, would lead to the elucidation, not only of the derivation and true meaning of words, but to a better understanding of the structure of the language ; so that the acquisition of the tongue could be made easier for future students. It can hardly be doubted, too, that useful light would be thrown in many ways upon the derivations and grammar of our western languages, and on grammar in general. It is highly probable, moreover, that research of that kind would uncover several more or less hidden grammatical features that would guide towards a more satisfactory method than any yet common for the rational and completely practical phonetic adaptation of Roman letters to Japanese, a matter of the greatest moment. But perhaps that might require first the still more needed improvement of the transliteration of Chinese, considering the very large number of words that have been taken from Chinese into Japanese, especially among scholars.



(PAPERS OF THE ORIENTAL CLUB.)P160-176

http://kuzan.f-edu.fukui-u.ac.jp/