KAMINISHI Toshiwo
2003, 2009, 2010, 2013, 2014
This is the transliteration system of the Japanese kana syllabary into Latin alphabet with every effort to stick to the English values as far as consonants are concerned. Vowels are Continental.
The kana syllabary is arranged in the table called fifty-sound chart, for it is made up of ten columns (gyau) and five rows (dan). Each column is called by the name of fuon or the paternal consonant, the representative of the consonants which the syllables situated in it share.
But the first, the sixth and the last column lack such an entity. The first is a vowel column. The sixth and the last columns have a dormant or potential consonant which is activated as a bilabial semivowel only on the first row where the vowel is most open, otherwise activated as an aspirate in the case of the sixth column if initial.
When the last or wa column has an active gwh sound as in transcription of foreign words, it is rendered by ƒE attached with a lowered kana vowel. This is represented by a geminate letter of gwwh.
The consonant of the sixth or ha column is represented with the grave accent or reversed apostrophe (ASCII 0x60). But when initial, it is represented either with ghh or, when the vowel is guh, with gfh. When it is medial, it is a dormant aspirate activated only when read by syllable. So –j (ho`o) is sometimes pronounced as [hoho]. •ê (haha) is an exception. It has no medial ghh in it. ”ª”¦ (ya`ata) is sometimes pronounced as [yawata] and sometimes as [yahata].
It may be interesting to note, though it is beyond the scope of EHS, that the medial ghh can be realized as gbh sound when the syllable is integrated to a compound word, and that when the juncture is a syllabic stop, this gbh is devoiced. H—tŒ´ (aki`abara) is sometimes pronounced as [akihabara], sometimes as [akiwabara] where gwh is barely audible and sometimes as [akibahara]. Or it could be [akibappara].
Sokuon or the syllabic-phonemic stopiƒbj is gth when word-final. When sokuon is medial, it is represented with the fuon. Hence the sokuon preceding gjh is gdh, and that preceding gchh is gth. But in the case of the ha column, the fuon is not adequate to represent stop except when the vowel is guh. The consonant in this case is represented by gfh and the same letter could be used as stop. Otherwise the post-stop ghh is realized as a velar or palatal fricative. So it should be represented by the homo-organic stop gkh.
Sokuon is not supposed to occur immediately before nasals, but remember that the homo-organic stop of gnh is gth and that of gmh is gph, for it is not so uncommon in southern part of Japan.
Hatsuon or the syllabic-phonemic nasaliƒ“ or ƒ€ in classical Japanese) is gnh. But when followed by a bilabial, gmh should be used. When followed by the labial glide of gwh, gmh should be used only when renzhau (see below) produces gmh sound,
Renzhau, an internal open plus juncture or internal sandhi occurs where gwh stands: e.g. ‹Î‰¤ (kinwau), ŽOˆÊ (samwi). But when it occurs without gwh, it can be expressed by a plus sign: e.g. æV‰¹ (kwan+on). ŽOŠ¥‰¤ has an open juncture, but you can use a hyphen to designate that it is composed of two words as sankwan-wau.
Apostrophes should be used to disambiguate prevocalic--including the case immediately preceding a semivowel and ghh of duration.
In transcription methods, macron-capped (or circonflex-capped) letters are used to represent some vowel sequences as a folded unit of a long vowel. In a trasliterational method we need not touble ourselves with the question how to express so-called long vowels. Only we need a symbol to represent the duration mark in katakana. and the letter ghh is assigned for it in our system. But notes on the pronunciation of some vowel sequences would be appropriate here.
The value of geuh is similar to that of gEuropeh and palatalizes the preceding alveolars. The value of gauh is similar to that of gautumnh. The rules also apply when the diphthongs have a visual ghh in between. The hyphen should be used to disambiguate diphthong-like sequences across the morpheme boundary. e.g. gte-usuh, gsono-uchih.
Strictly speaking, vowels are different from diphthongs in that they could remain absolutely constant for any length of time. So if the transliteration gives a geminate vowel or the same vowel twice, it might be taken as a good candidate for a long vowel. But in most cases a geminate vowel represents two distinctive vowels rather than a long unit2.
The Cabinet notification of 1986 is negligent of the distinction and says gwhether they are pronounced as a long ƒI vowel or two distinctive vowels as ƒIEƒIAƒREƒI, they should be written by adding ƒIh.
The sample words and phrases of spurious long vowels given in the Cabinet order are as follows. Here, such written forms are used as hide the elements or kana letters as possible. When kana is used, it is in Historical Kana Orhography.
The second field shows the transliteration of HKO by EHS.
The third field shows the transliteration of Contemporay Kana Usage by EHS. CKU is often no more than the restrtiction of some kana letters such as ƒa ƒd ƒ ƒ‘ ƒ’ or medial ƒnƒqƒtƒwƒz. But here it tries to capture the result of coalescence or mutual assimilation by replacing the onset kana, often incorporating a palatal foreglide.
The fourth field shows whether the coalescence (assimilation) occurs or rather the sequence represents distinctive (double) vowels.
The fifth field shows transcription of the Contemporary Japanese by the Common Hepburn System (no distinction of ƒaƒWAƒdƒY, longness shown by a macron, rendering of syllabic nasal before bilabials by gmh.
| sample | Historical | Restricted | Assm/Dbl | Common Hepburn |
| ‚¨•ê‚³‚ñ | okaasan | " | Dbl | okāsan |
| ‚¨‘c•ê‚³‚ñ | obaasan | " | Dbl | obāsan |
| ŒZ‚³‚ñ | niisan | " | Dbl | onīsan |
| ‚¨‘c•ƒ‚³‚ñ | ojiisan | ozhiisan | Dbl | ojisan |
| ‚¨Š¦‚¤‚²‚´‚¢‚Ü‚· | osamuugozaimasu | " | Dbl | osamūgozaimasu |
| ‹óŸ† | kuuki | " | Assim | kūki |
| •v•w | fuufu | " | Assim | fūfu |
| Šð‚¤‘¶‚¶‚Ü‚· | ureshiuzonzhimasu | ureshuu- | progressive | ureshūzonjimasu |
| ŒÓ‰Z | kiuri | kyuuri | progressive | kyūri |
| –n` | bokuzhi`u | bokuzhuu | progressive | bokujū |
| ’•¶ | chuumon | " | Assim | chūmon |
| Žo‚³‚ñ | neesan | " | Dbl | nēsan |
| ‚¦‚¦ | ee | " | Dbl | ē |
| ‚¨•ƒ‚³‚ñ | otousan | " | regressive | otōsan |
| “•äi | toudai | " | regressive | tōdai |
| Žál | wakaudo | wakoudo | mutu./regress. | wakōdo |
| ê_–· | aumu | oumu | mutu./regress. | ōmu |
| ”ƒ‚Í‚¤ | ka`au | kaou | mutu./regress. | kaō |
| —V‚΂¤ | asobau | asobou | mutu./regress. | asobō |
| ‚¨‘‚₤ | ohayau | ohayou | mutu./regress. | ohayō |
| î | a`ugi | ougi | mutu./regress. | ōgi |
| e‚é | ha`uru | houru | mutu./regress. | hōru |
| “ƒ | ta`u | tou | mutu./regress. | tō |
| ‚æ‚¢‚Å‚¹‚¤ | yoideseu | yoideshou | mutu./regress. | yoideshō |
| ᢕ\ | happeu | happyou | mutu./regress. | happyō |
| ¡“ú | ke`u | kyou | mutu./regress. | kyō |
| ’±X | te`ute`u | chouchou | mutu./regress. | chōchō |
In Ka`au, the medial ghh is not realized as a bilabial semivowel, for the following vowel is no longer pronounced as the most open vowel. It makes the coalesced sequence gauh which is to be pronounced like that of autumn.
| ƒA | ƒC | ƒE | ƒG | ƒI | ||||
| ia | ii | iu | ie | io | ||||
| ƒJ | ƒL | ƒN | ƒP | ƒR | ƒLƒƒ | ƒLƒ… | iƒLƒFj | ƒLƒ‡ |
| ka | ki | ku | ke | ko | kya | kyu | (kye) | kyo |
| ƒK | ƒM | ƒO | ƒQ | ƒS | ƒMƒƒ | ƒMƒ… | iƒMƒFj | ƒMƒ‡ |
| ga | gi | gu | ge | go | gya | gyu | (gye) | gyo |
| ƒT | ƒViƒXƒBj | ƒX | ƒZ | ƒVƒƒ | ƒVƒ… | iƒVƒFj | ƒVƒ‡ | |
| sa | shi (si) | su | se | so | sha | shu | (she) | sho |
| ƒU | ƒWiƒYƒBj | ƒY | ƒ[ | ƒ] | ƒWƒƒ | ƒWƒ… | iƒWƒFj | ƒWƒ‡ |
| za | zhi (zi) | zu | ze | zo | zha | zhu | (zhe) | zho |
| ƒ^ | ƒ`iƒeƒBj | ƒciƒgƒDj | ƒe | ƒg | ƒ`ƒƒ | ƒ`ƒ…iƒeƒ…j | iƒ`ƒFj | ƒ`ƒ‡ |
| ta | chi (ti) | tsu (tu) | te | to | cha | chu (tyu) | (che) | cho |
| ƒ_ | ƒaiƒfƒBj | ƒdiƒhƒDj | ƒf | ƒh | ƒaƒƒ | ƒaƒ…iƒfƒ…j | iƒaƒFj | ƒaƒ‡ |
| da | ji (di) | dzu (du) | de | do | ja | ju (dyu) | (je) | jo |
| ƒi | ƒj | ƒk | ƒl | ƒm | ƒjƒƒ | ƒjƒ… | iƒjƒFj | ƒjƒ‡ |
| na | ni | nu | ne | no | nya | nyu | (nye) | nyo |
| ƒn | ƒq | ƒt | ƒw | ƒz | ƒqƒƒ | ƒqƒ… | iƒqƒFj | ƒqƒ‡ |
| ha | hi | fu | he | ho | hya | hyu | (hye) | hyo |
| ƒo | ƒr | ƒu | ƒx | ƒrƒƒ | ƒrƒ… | iƒrƒFj | ƒrƒ‡ | |
| ba | bi | bu | be | bo | bya | byu | (bye) | byo |
| ƒp | ƒs | ƒv | ƒy | ƒ| | ƒsƒƒ | ƒsƒ… | iƒsƒFj | ƒsƒ‡ |
| pa | pi | pu | pe | po | pya | pyu | (pye) | pyo |
| ƒ€ | ƒ | ƒ‚ | ƒƒ | ƒ… | i ƒFj | ƒ‡ | ||
| ma | mi | mu | me | mo | mya | myu | (mye) | myo |
| ƒ„ | ƒ† | iƒCƒFj | ƒˆ | |||||
| ya | yu | (ye) | yo | |||||
| ƒ‰ | ƒŠ | ƒ‹ | ƒŒ | ƒ | ƒŠƒƒ | ƒŠƒ… | iƒŠƒFj | ƒŠƒ‡ |
| ra | ri | ru | re | ro | rya | ryu | (rye) | ryo |
| ƒiƒEƒ@j | ƒiƒEƒBj | ƒ‘iƒEƒFj | ƒ’iƒEƒHj | |||||
| wa (wwa) | wi (wwi) | we (wwe) | wo (wwo) | |||||
| ƒNƒ@ | ƒNƒB | ƒNƒF | ƒNƒH | |||||
| kwa | kwi | kwe | kwo | |||||
| ƒOƒ@ | ƒOƒB | ƒOƒF | ƒOƒH | |||||
| gwa | gwi | gwe | gwo | |||||
| ƒcƒ@ | ƒcƒB | ƒcƒF | ƒcƒH | |||||
| tsa | tsi | tse | tso | |||||
| ƒtƒ@ | ƒtƒB | ƒtƒF | ƒtƒH | ƒtƒ… | ƒtƒ‡ | |||
| fa | fi | fe | fo | fyu | fyo | |||
| ƒ”ƒ@ | ƒ”ƒB | ƒ” | ƒ”ƒF | ƒ”ƒH | ƒ”ƒ… | ƒ”ƒ‡ | ||
| va | vi | vu | ve | vo | vyu | vyo |
Parenthesized syllables and those of the last three lines are used only in transcription of foreign words overtly written in katakana.
(EHS is on the historical kana scriptj
| °’£ƒwƒ{ƒ“ | ƒwƒ{ƒ“ | ŠO–±È | V•Äš | •¶‰»œK | š “y’n—‰@ | ¬›{Z | “ú–{Ž® | |
| “¡ | fuji | " | " | " | " | " | huzi | hudi |
| •xŽm | fuzhi | fuji | " | " | " | " | huzi | " |
| •@ŒŒ | hanaji | " | " | " | " | " | hanazi | hanadi |
| ˜c‚Ý | hizumi | " | " | " | " | " | " | " |
| ’û | hidzume | hizume | " | " | " | " | " | hidume |
| —°‰© | iwau | io | ioh | io | ioo | io | iô | " |
| —®‹… | riukiu | ryukyu | ryukyu | ryukyu | ryuukyuu | ryukyu | ryûkyû | " |
| —‰¤ | jowau | joo | jooh | joo | jooo | joo | zyoô | " |
| ¬˜H | kouji | koji | kohji | koji | kooji | koji | kôzi | " |
| b‰³ | ka`uotsu | kootsu | kohotsu | kootsu | koootsu | kootsu | kôotu | " |
| ‘å• | o`oguro | oguro | ohguro | oguro | ooguro | oguro | ôguro | " |
| ‰X | wauwau | oo | ohoh | oo | oooo | oo | ôô | " |
| ‹ß] | a`umi | omi | ohmi | omi | oomi | omi | ômi | " |
| ”Ñ“c | i`ida | ida | ida | " | iida | ida | îda | " |
| ’²•z | teufu | chofu | chohfu | chofu | choofu | chofu | tyôhu | " |
| ŽU•à | sampo | " | sanpo | " | " | " | " | " |
| ŽOˆÊ | samwi | sammi | sanmi | " | " | " | " | " |
| ”½œä | han+ou | hanno | hannoh | hanno | hannoo | hanno | hannô | " |
| ‹Î‰¤ | kinwau | kinno | kinnoh | kinno | kinnoo | kinno | kinnô | " |
As for ƒwƒ{ƒ“ method, see –¾Ž¡›{‰@š¤‘ŠÙƒfƒWƒ^ƒ‹ƒA[ƒJƒCƒuƒX"
As for ŠO–±È method (or passport method), see “Œ‹ž“sƒpƒXƒ|[ƒgƒZƒ“ƒ^[
As for the method of •¶‰»œK, see the chapter 4 of the Handbook of Learning Japanese and Life in Japan by Agency for Cultural Affairs
The pp.164 and 165 carry romanization tables for Hiragana and Katakana and stipulate to use only gnh for syllabic nasal with no provision for the case of pre-bilabial position. But the writer betrays his good command of alphabet by rightly committing the error of using the homo-organic nasal before labials without exception unless separated with a space: e.g. 110 ban mo (p.140); Kombanwa, bumboogu, empitsu, kompasu, bempi, densembyoo, Gum'ma-ken, hinambasho, bumbetsu, tempura, hambaagaa.
There is no explanation of the apostrophe in Gum'ma.
As for the š “y’n—‰@ method, see ’n–¼‚̃[ƒ}Žš•\‹L š “y’n—‰@
V•Äš method (ALA-LC Romanization) is based on the table of Kenkyusha's New Japanese-English Dictionary (in p. xiii for 4th edition).
The original version was carried by Nihongo Gaku January 2003 published by Meijishoin with the title of gKwakuchau Hebonshikino Teishauh. It was in Japanese.
This was made into PDF and published on the web site of ƒ[Žš‘Š’kŽº by courtesy of Mr KAIDZU on March 16th, 2003. The former arrangement of the site was institutionalized systems first and then personal or original systems and I appreciate it very much that this distinction was dicarded on the arrival of EHS.
The introduction of the reversed apostrophe as a medial ghh was made in Eigo`a Nihonzhin Kyoushidakara Woshi`erareru published by Yousensha in April 2004.
The English version was first written on December 11. It was a working paper. I made several copies of it and distributed among some of my friends. Dr KATSURA Shigetoshi, Professor Emeritus of physics, Touhoku (Tohoku) University critisized severely differentiation of ƒEƒB and ƒ by case and suggested the digraph gwwh instead. to which I yielded. It was the revised version of EHS which was published on the same site December 19 of the same year.
Sokuon seldom occurs before nasals. But in KAGOSHIMA, you can hear kagopmabeniŽ™Z“‡ç‡jor sutnawonagoiD‚Á‚È—Žqj. The stipulation was added in December 2009.
Renzhau is a phenomenon usually masked by kanzhi logographs and kana rendering in reformed script fails to capture it. It was at the end of November 2010 that the author noticed mispronunciation caused by this on radio. I do not know whether the text was in reformed script or not, but the reader obviously took the kana string statically and missed its dynamics. Hence generalization of renzhau to cover the implicit cases.
In May 2011, YABUKI Susumu, Professor Emeritus of Yokohama City University told me ASAKA`A Kwan'ichi (ASAKAWA Kan'ichi) used the same method in Documents of Iriki Yale 1929, and in October 2012, I had a chance to read Ernest Satow's ON THE TRANSLITERATION OF THE JAPANESE SYLLABARY published in the Transactions of the Asiatic Society of Japan 7, 1879, and found his method identical. The only feature EHS introduced is the differentiation of medial ghh.
In December 2013, the section on vowel sequences is added.
Sample transcription:iroha-uta
Iro`a ni`o`edo chirinuruwo Wagayo tarezo tsunenaramu Uwino okuyama ke`u koete Asakiyumemizhi we`imosezu
The medial ghh of the inverted apostrophe should be pronounced as ghh or if immediately followed by guh as gfh and the voiced kana as the unvoiced counterpart when read by syllable.
This document was generated using the LaTeX2HTML translator Version 2008 (1.71)
Copyright © 1993, 1994, 1995, 1996,
Nikos Drakos,
Computer Based Learning Unit, University of Leeds.
Copyright © 1997, 1998, 1999,
Ross Moore,
Mathematics Department, Macquarie University, Sydney.
The command line arguments were:
latex2html xtn_hbn
The translation was initiated by toshiwo on 2014-02-06
A typo was corrected and kagotmaben was changed to kagopmaben on 2014-12-18