Genetic Joyce Studies - Issue 7 (Spring 2007)
Japanese in VI.B.12: Some Supplementary Notes
Mikio Fuse

Notebook VI.B.12 has a special appeal to Japanese researchers: if they are curious enough to skim through nearly two hundred pages of the notebook in the Archive volume, they will be surprised to find two consecutive pages (112-13) that bear a pair of large and neat inscriptions of Japanese kanji ideograms (unmistakably unauthorial) representing the names of two famous deities of ancient Japanese mythology. The notebook also includes two sets of Japanese-related entries, the more conspicuous of which is an impressive list of Japanese first-person singular pronouns (14) that is strongly reminiscent of FW 484.26-27: "Washywatchywataywatashy! Oirasesheorebukujibun! Watacooshy lot!" The other, minor Japanese-related unit is a short note on romanization of Japanese (111).

The forthcoming Buffalo Notebook Edition of the notebook VI.B.12 incorporates two significant discoveries that were made in recent years by Japanese researchers. As the agent of the minor of the two discoveries, I will first make a report on the source book for both the list of Japanese first-person singular pronouns and the note on romanization of Japanese. Next I will give a few technical notes on the transcription of a unit on the Japanese mythology pages about which an epoc-making discovery was reported earlier by Yasuo Kumagai in the Spring 2002 issue of this journal. All I write here is to supplement what the editors of the Notebook Edition have chosen to publish in the coming volume. 

VI.B.12.014(e): I. { watakushi / [d?] ore (children) /  boku (soldiers / temaye (beggars) / watashi ([friends]) / watai (W) / washi / wattchi (rustic) / sessha (officer) / jibun (self / oira (fam / Tekurada

It was way back in 1997 that I first brought this notebook entry, along with the one on Japanese mythology, to the attention of the FWAKE-L mailing list. As I wrote in the post, one of the most curious things about the I-word list is the evidently "obsolete, archaic or dialectal" nature of the vocabulary Joyce registered, even if we assume that he read/heard the words in Paris in the pre-WWII years when the language spoken/written by Japanese was indeed markedly different from what it is today. Being notified of the coming publication of the notebook VI.B.12 in the Notebook Edition, I felt impelled to make dedicated research and came to know of the availableness of Stefan Kaiser's The Western Rediscovery of the Japanese Language (1995), that is an eight-volume collection in reprints of early guidebooks and textbooks written between 1861 and 1889 by and for the Westerners who had to study or learn the Japanese language of late-feudal and early-modern periods. (The Yedo era of the Tokugawa shogunate terminated in 1867 to give way to the Meiji era of 1868-1912 that propelled Japan's modernization after the Western powers' models). Consequently I checked the volumes, particularly targeting the sections on pronouns in each guidebook or grammar book, and I found within an hour of my research that Joyce's source was W. G. Aston's A Grammar of the Japanese Spoken Language. 4th ed. (1888), pp. 11-12 [Chapter 4 ("The Pronoun")].

The first three columns of the following table demonstrate the crucial intertextual correspondence between Joyce and Aston. For ease of identification, the words Joyce literally copied from Aston are marked in blue. I have also included what McHugh and Skrabanek, with the help of their Japanese informants, have given in their earlier studies as glosses for the Japanese I-words found in FW 484.26-27. One would note how legitimate genetic evidence not only serves as legitimate annotations but can eliminate unwanted confusions, speculations and anachronism. Luckily the news of source discovery reached McHugh just in time for the last-minute inclusion in his latest third edition of Annotations (2006).

Joyce's note (1927)

Aston (1888), p. 11

McHugh1 (1980)

Skrabanek (1985), p. 14

McHugh2 (1991)



Watakushi, 'I' (plural watakushi domo, 'we'), is the ordinary word for the pronoun of the first person.


used by both sexes, neutral form

I (used with superiors)

[d?] ore


Ore (plural orera) is less respectful, and is the word mostly used by coolies, etc., to each other. To inferiors it is a somewhat hauty word.

O yama no taisho ore hitori [I'm the king of the castle. (in the children's game.)]

I (vulg.)

neutral or used to inferiors

I (used to inferiors)



Students and soldiers say boku for 'I' [...].


used by children

I (generally used)



Temaye is a humble word for 'I,' much used by the lower classes of Tokio in addressing their superiors. [...]






watashi (familiar)

I (abbr. of watakushi)

[watachi (SIC):] a neutral form, abbreviation of watakushi

I (abbr. of 'watakushi')



watai (by women)


used by prostitutes

[watay:] I (familiar, used by women)



washi (very familiar)


used by senior or superior

I (abbr. of 'watashi')



wattchi (rustic)


a neutral dialect form

[wachi (SIC):] I (Dial., used by men)



sessha (formal)

I (epistolary style of elderly man)

[seshe (SIC):] archaic, used by samurai

I (official, obsolete)



jibun (properly 'self')

[hibun (SIC):] myself

used by soldiers in Meiji and Taisho eras

I (obsolete)



oira (familiar)


a variant of ore

I (form of 'ore')

Except for the reversal of the last two terms, jibun and oira, Joyce's arrangement of the I-words faithfully follows the order in which they occur in Aston's text. For the word starting with the letter "d" Joyce attempted to write as a second term before he corrected it to "ore," Aston's explanation of the first I-word watakushi would be a clue. Suppose Joyce entered "watakushi" in the notebook and looked for the next italicized I-word that comes after the first one watakushi, he might well have been tempted to erroneously pick up domo that does immediately follow watakushi in Aston's parenthetical note for watakushi: "plural watakushi domo, we."

Until the discovery of the source, the last item of this unit, "Tekurada," had remained adamantly impenetrable, for it is unlike any Japanese first-person singular pronoun. To me it looked rather like a proper noun, either a place name or a family name, but its relevance to Japanese grammar seemed hopelessly tenuous. As it turned out, Joyce's "Tekurada" is certainly a proper noun. It comes from one of Aston's example sentences:

Joyce's note

Aston, p. 12


Watakushi wa Tekurada Futoshi de gozarimasu. Hajimete o me ni kakarimashita. [I am Tekurada Futoshi. I have the honour of meeting you for the first time.]

"Tekurada" is very rare for a Japanese family name but, as any genetic source hunter knows, rare words in Joyce's notebooks always positively contribute to source identification because of the most uniquely unequivocal intertextual correspondence they betray with the source.

Although the whole unit is undeleted, it calls for a careful genetic treatment. In the same year that he composed the notebook, or at the latest before April 1928, Joyce came back to this I-word list as it had been transcribed by Madame Raphael in VI.C.6. It was then that the first term "watakushi" singly made it in III:3:

  • VI.C.6.004(j): bwatakushi

    • 47484a-128: ^+Watacooshy lot!+^ | JJA 58:250 | May 1926-Apr 1928 | III:3A.6/3B.6 | FW 484.26-27

About a decade later in late 1938 Joyce came back to the list in the C notebook again to recycle and transfer the rest of I-words onto VI.B.30 (composed late November to late December, 1938, according to Rose). On pp. 72 though 75 of this notebook Joyce experiments with amulgamating all the rest of I-words into a couple of impossibly hybrid word strings. The final result of this experiment, 075(a), was soon to be used, with slight modification, as a holographic overlay to the last but one galley proof of III:3A:

  • VI.B.30.072(b): gwashi / wattchi / sesha / oira / jibun

  • VI.B.30.073(b): gore  boku[?] / [temaye] / Patrick / watai / watashi

  • VI.B.30.074(a): g[T]e[m]ayore[kubu] / jibun

  • VI.B.30.074(b): gOirasesha

  • VI.B.30.075(a): gWashiwattchi / wataiwatashi / oirasesha / temayeoreboku / jibun

    • 47487-193v & 194: ^+Washiwattchywataywatashy! Oirasesheorebokujibun!+^ Watacooshy lot! | JJA 62:356 & 357 | 22 April and 31 May 1937 (printer's date of draft; date of insertion to be confirmed) | III:1A.13+/1B.4+/1C.10+/1D.13+//2A.14+/2B.12+/2C.14+//3A.11+/3B.18+//4.8+ | FW 484.26

In addition to this basic textual history, it is noteworthy that in the upper half of  pp. 72 and 73 of VI.B.30 we have entries written in Mme Raphael's hand, which bear close resemblance to (but are not exactly the same as) what Joyce himself wrote earlier on pp. 62-63 of the same notebook. These authorial and scribal entries prove to have been based on Ernest Fenollosa's article edited and posthumously published by Ezra Pound, first in the Little Review of 1919 and in Pound's Instigations in the following year, under the title "The Chinese Written Character as a Medium for Poetry" (Thanks to Geert Lernout for googling to find the source for me). Seeing that in the source passage for the second cluster of Joyce's notes (p. 63) Fenollosa specially focuses on the five forms of Chinese I-words, it is not difficult to conjecture why Joyce asked Mme Raphael to redo this particular note on p. 73 and started on the same and following notebook pages to work afresh with the Japanese I-words in VI.C.6. Mme Raphael's copy of the Fenollosa notes on p. 73 is deleted in green, that is the same colour in which the Japanese I-words on pp. 72 through 75 are deleted. Surely both the Chinese and Japanese I-words, along with what Fenollosa explains as Chinese verbal prepositions, were used for revision of two consecutive galley sheets (47487-193 and 194) in the same subsection 3A of Book III:

  • VI.B.30.062(a): oby = to cause / to =  fall [toward] / in = remain / from = follow / because = to use / and = to be / incuded / and = be parallel / or = partake / if = to permit

  • VI.B.30.063(a): I = {[handspear], emph / 5 + mouth hold off / crowd by speaking / to conceal = selfish / cocoon egoistic / [|] self [--] to self

  • VI.B.30.072(a): gby = cause / to = to fall toward / in = to remain, to dwell / from = to follow / because = to use / and = to be incuded under one / -- [=] to be parallel / or = to partake / if = to permit

  • VI.B.30.073(a): gI { spear in hand = emphatic / five and a mouth = weak and / defensive / conceal = selfish and private / cocoon sign and a mouth = / egoistic / selfsign -- speaking of / oneself

    • 47487-192v & 193 | JJA 62:354 & 355 | 22 April and 31 May 1937 (printer's date of draft; date of insertion to be confirmed) | III:1A.13+/1B.4+/1C.10+/1D.13+//2A.14+/2B.12+/2C.14+//3A.11+/3B.18+//4.8+ |  FW 483.15-484.14

VI.B.12.111(d): Jap { si = shi / hu = fu / wu = u / we = ye

To Japanese researchers the implication of this unit is seemingly self-evident. It apparently compares two different systems of romanizing Japanese, say the Hepburn system and the Kunrei ("Cabinet-ordered") system. Indeed, in the former system the sets of five sounds the five Japanese vowels (a-i-u-e-o) make with the initial consonants s and h are respectively spelt as sa-shi-su-se-so and ha-hi-fu-he-ho, while in the latter system they are spelt as sa-si-su-se-so and ha-hi-hu-he-ho. The Hepburn system was proposed by an American Presbyterian missionary James Curtis Hepburn in his first ever Japanese-English dictionary in history, A Japanese and English Dictionary (1867). The Kunrei system was actually not introduced by the Japanese government until 1937, nearly a decade after the composition of the notebook VI.B.12, but there is the earlier Nippon system, originally proposed by Aikitsu Tanakadate in 1885, whose spellings sa-si-su-se-so and ha-hi-hu-he-ho were adopted by the 1937 Kunrei system.

But here again, as in the case of the I-word list, we are faced with a curious instance of archaism recorded by Joyce in the latter two pairs of spelling variants (wu = u and we = ye). The case in point is the sounds the consonant w makes when followed by the five vowels, a-i-u-e-o. Theoretically, there should be five different sounds, wa-wi-wu-we-wo, to spell them in the Nippon system, but in practice the sounds wi, wu and we have been assimilated to the pure vowels i, u, e. As for wo, although the spelling is still extant, the sound has become indistinguishable from o, except in dialectal and archaic pronunciation. In order to ascertain how the sound set wa-wi-wu-we-wo has been spelt in different systems of Romanization, I have compiled the following comparison table based on the tables of key Romanization systems available at the Meiji Gakuin University Library's digital archives (manuscript and different editions of Hepburn's Dictionary) and Simizu Masayuki's Romazi Bunko [The Library of Romanized Japanese] that includes historical documents regarding the Nippon, Kunrei and Revised Kunrei systems:

Hepburn MS












Revised Kunrei


? / ?








? / ?








? / ?








? / ?








? / ?








It would be noted that not only the 1937 Kunrei system but also Hepburn himself preferred the spelling e to ye after the 1872 second edition of his Dictionary. With the help of the above table of comparison we can clearly point out where exactly Joyce's note is found oddly anachronistic: while its awareness of the Nippon system indicates that Joyce's source should postdate 1885, its inclusion of the spelling ye indicates that the source should refer to a not up-to-date version of the Hepburn system that was current only in the 1860s. How can we solve this curious anachronistic confusion? As it turned out, we can only do so decisively by means of genetic evidence. Joyce was again reading Aston's A Grammar of the Japanese Spoken Language. 4th ed. (1888):

Aston, pp. 2-3














































































It will be seen that there are a number of irregularities and repetitions in the above Table. These are owing to the circumstances that there are certain sounds which a Japanese cannot, or at any rate, does not pronounce. For si, he says shi, for hu, fu, for yi, wi, wu and we, i, i, u and ye, and so on.

It was Aston who, in his book of 1888, still used the older version of Hepburn system Hepburn himself no longer used. Aston's apologetic remark attached to the table reads rather odd. For one thing, Aston talks as if he himself has adopted the newly proposed Nippon system (si ... hu ... yi, wi, wu and we) while still sticking to the outdated version of Hepburn system in his table. For another, Aston puts the blame of "irregularities and repetitions" in the table on the Japanese speaker's inability to pronounce "si ... hu ... yi, wi, wu and we," whereas it was actually for the sake of English speakers that Hepburn introduced the irregular spellings shi and  fu. In fact, in the period between the 1920's and 1940's, which includes the very period Joyce composed Finnegans Wake, Aston's remark would have already sounded absurd not only to the more or less nationalistic Japanese promoters of the "native" Nippon system but also to Western linguists and phonologists of world renown, such as Daniel Jones and Otto Jespersen, because they found the Nippon system phonologically more logical and consistent than the Hepburn system that allows for irregular variant spellings of the same phonemes only because the spellings like shi and fi facilitate English learners to pronounce them (Matsuura).

VI.B.12.112(g): ????? / 1 2 3 4 ^+?+^ 5 / Amé no minaka nushi no / mikoto / (1) [(2)  (3) ^+(2)^+???+^+^] (4) / (5) / (heaven's middle master / ^+of+^ / Amé no minaka nushi / no / ^+of+^ / mikoto } >>

VI.B.12.113(a): Takaoki / Katta >

VI.B.12.113(b): ????? / 1  2 3 4 5 / Ama terasu oh mi kami / (1) Heaven (2) shine (3) great / 4) pre honorific / (5) god(dess)

Yasuo Kumagai's breathtakingly intriguing discovery of a crucial countertext to these entries, i.e. Takaoki Katta's personal notebooks recording the Japanese scholar's interview with Joyce on 15 July, 1926, is well documented in his GJS article, complete with the transcription of the key Katta notebook pages. My supplementary notes here are not to add anything redundant to Kumagai's excellent presentation, but to clarify two technical issues I discussed with the editors of the Notebook Edition regarding the transcription of the first of these three units, i.e. 112(g). In this unit there are two types of emendations, both concerned with the Japanese postposition no (A no B = "B of A").

As is illustrated by the black and gray cells of the table below, when the god's name, Amé no minaka nushi no mikoto, is written in kanji (or Chinese) ideograms, the postposition no is not spelt out but is tacitly understood. In other words, when the Chinese ideograms are read or spelt out in Japanese, the tacit postposition has to be automatically supplemented:

To explain the literal word-for-word correspondence between the Chinese ideograms and the Japanese transliteration, Katta first used parenthetical Arabic numerals to number the words in the transliteration corresponding to the numbers assigned to the Chinese characters, but in doing so he erroneously wrote "(2)" above the first no. As he noticed the error, he deleted the "(2)" and added a new "(2)" above minaka. (See the red cells in the above table.) However, that didn't fix his error, for minaka actually consists of two elements, mi (honorific prefix) and naka ("middle"), and should be associated with the second and thrid Chinese characters respectively. When Katta realized this new error and renumbered the words in the transliteration, he used encircled Arabic numerals. The "4" of the fourth Chinese character was probably encircled at the same time to confirm the correct sequence of correspondence. (See the yellow cells in the above table.)

Katta's painful correction of these errors evidently triggered Joyce's positive interest in the Japanese postposition no, for in the latter half of this unit we have Joyce's own copy of the Japanese transliteration, with special emphasis on the two parts of it where no is concered. The second addition "no / ^+of+^" simply glosses no, while the first addition of "of" overwrites "'s" (Amé no ... =  heaven's ... = ... of heaven ).

While Kumagai has given life to the Japanese mythology units by discovering a historical document that testifies to the lively (and very much humane) interaction between Joyce and Katta, my small contribution here would supplement Kumagai's work by highlighting some subtle "penmarks" on the notebook page where their interaction is dynamically traceable.


Works Cited

Aston, W. G. A Grammar of the Japanese Spoken Language. 4th ed. Yokohama: Lane, 1888. Rept. in Stefan Kaiser (ed.), The Western Rediscovery of the Japanese Language. vol. 6. London: Curzon, 1995.

Fenollosa, Ernest. "The Chinese Written Character As a Medium for Poetry." Ed. Ezra Pound. Originally published in Ezra Pound, Instigations (1920).

Fuse, Mikio. "Exotica Nipponica (VI.B.12)." The F-WAKE-L Maiking List. 13 August 1997.

Hepburn, J. C. A Japanese and English Dictionary. Meiji Gakuin University Library Digital Archives.  (明治学院大学図書館 『「和英語林集成」デジタルアーカイブス』)

The Japanese Government. "Naikaku Kunrei Dai 3-go" [Cabinet Order No. 3]. 1937. Reproduced in Shimizu. (内閣訓令第3号)

---. "Naikaku Kokuji Dai 1-go" [Cabinet Notification No. 1]. 1954. Reproduced in Shimizu. (内閣告示第1号)

Kumagai, Yasuo. "'Takaoki Katta' (VI.B.12:113)." Genetic Joyce Studies 2 (2002).

Matsuura, Shiro. "Hyaku-Yonen Kakatta Hyojun-ka" [One-hundred and Four Years' Meandering Journey towards Standardization] (松浦四郎 「104年かかった標準化」)

McHugh, Roland. Annotations to Finnegans Wake. 3rd ed. Baltimore: Johns Hopkins UP, 2006.

---. Annotations to Finnegans Wake. 1st ed. London: Routledge, 1980.

---. Annotations to Finnegans Wake. 2nd ed. Baltimore: Johns Hopkins UP, 1991.

Rose, Danis. The Textual Diaries of James Joyce. Dublin: Lilliput, 1995.

Shimizu, Masayuki. Romazi Bunko [The Library of Romanized Japanese].

Skrabanek, Petr. "St. Patrick's Nightmare Confession." Finnegans Wake Circular 1.1 (1985): 5-20.

Tanakadate, Aikitsu. "Rigaku-kyokai Zasshi o Romaji nite Hatsuda suru no Hotsugi oyobi Romaji Yoho Iken" [A Proposal to Publish the Journal of Scientific Society in Romanized Japanese and Observations on the System of Romanization]. Reproduced in Shimizu. (田中舘愛橘 「理学協会雑誌ヲ羅馬字ニテ発兌スルノ発議及び羅馬字用法意見」)