Romanized Japanese Bible translation
« previous post | next post »
Stumbled upon an earliest surviving partial bible translation into Japanese made by Manuel Barreto in 1591 (known as バレト写本 or Reg. lat. 459), here's the first paragraph.
Scans can be found here:https://t.co/pW2nH8Udtj pic.twitter.com/HMehP0F6qd
— Maxim Persikov (Panates) (@gyankotsu) December 5, 2024
The Portuguese were the first Europeans to arrive in Japan in 1543, establishing trade and cultural exchange, including the introduction of firearms and Christianity, which later led to persecution and the Sakoku (closed country) policy in the 17th century. (AIO)
For the impact of Portuguese missionaries on the study of East Asian languages and linguistics, see, for example:
W. South Coblin, Francisco Varo's Glossary of the Mandarin Language. Vol. 1: An English and Chinese Annotation of the Vocabulario de la Lengua Mandarina Vol. 2: Pinyin and English Index of the Vocabulario de la Lengua Mandarina (London: Routledge, 2006).
Abstract
See also: W. South Coblin and Joseph Abraham Levi, Francisco Varo's Grammar of the Mandarin Language (1703). An English Translation of 'Arte de la lengua mandarina' (Philadelphia: John Benjamins, 2000).
Selected readings
- "Portuguese words in Japanese, and beyond" (1/8/23)
- "Missionary Linguistics; the joys of interpreting" (12/25/21)
- "Macaque and Old Sinitic reconstructions" (12/17/20)
[h.t. Geoffrey Wade]
J.W. Brewer said,
April 10, 2025 @ 2:37 pm
I can make some guesses as to how one might convert this romanized text into "regular" Hepburn romaji, but it would be nice to see how it looked (w/ someone else having done the work …), and also to understand how much of the differences are simply different choices by Father Barreto (who was presumably more influenced by Portuguese orthographic conventions than Dr. Hepburn was …) versus reflecting any changes in actual Japanese pronunciation over time, or even regional variations if Barreto was in 16th-century Nagasaki but Hepburn in 19th-century Yokohama.
mcur said,
April 10, 2025 @ 5:10 pm
@J.W. Brewer:
The most interesting to me is the use of /f/ for what in Hepburn is /h/. At some point in the past hyaku would have been pronounced fiaku (the sound changed from /p/ to /f/ to /h/ over the course of about 1000 years), but I don't know if that sound was still current in 1591. Also Portuguese has no /h/ sound, so it could equally be a Portuguese influence.
The /x/ in "xengo" and the /tx/ in "goxutxe" are both /s/ in Hepburn, not /sh/. This I can't explain from a either Japanese or a Portuguese angle, but if it's repeated twice in such a short segment it probably isn't merely an error. Very mysterious.
Otherwise, the preview on Twitter (I can't get the actual manuscripts to load) is exactly how I would expect it to be read today.
In Hepburn for you (with wapuro-style long vowels):
JESUS goshussei irai sen gohyaku hachijuu ku nen me Nippon Tenshou juushichi nen shimo tsuki juuhachi nichi IESUS no gotanjou no hi nari
Michael Vnuk said,
April 10, 2025 @ 5:51 pm
I have no knowledge of the languages in question, but the fact that the only word I recognise in the brief extract is written in two different ways (JESUS and IESUS) makes me concerned about the rest. Also, in the third word, are there dots above the letter Y? And why is one letter C in the second line rendered with a cedilla in the transliteration?
Chris Button said,
April 10, 2025 @ 6:21 pm
The "chi" in "nichi" reflects today. But we also have the 1603 Japanese – Portuguese dictionary from only a few years later still showing the vacillation with a plain -t coda, which is crucial evidence for the reconstruction of obstruent codas in Sino-Japanese borrowings.
Chris Button said,
April 10, 2025 @ 6:22 pm
As in nichi ~ nit
Josh R. said,
April 10, 2025 @ 6:52 pm
Michael Vnuk asked: "And why is one letter C in the second line rendered with a cedilla in the transliteration?"
It is representing the "ts" of what is today rendered "tsuki". Apparently, per the Wikipedia article on Portuguese phonology, ç used to represent this /ts/ sound, though in modern Portuguese it has become an /s/. (I of course welcome any correction/elaboration by those versed in Portuguese phonology.)
J.W. Brewer said,
April 10, 2025 @ 7:05 pm
@mcur: Obrigado/Arigato.
Kalov Bliov said,
April 10, 2025 @ 11:31 pm
Related to writing modern /h/ as f, this guy's name was written as Faxicura in the 16th Century. https://en.wikipedia.org/wiki/Hasekura_Tsunenaga
Gokul Madhavan said,
April 11, 2025 @ 3:41 am
Would this use of < x > in Portuguese have overlapped with the use of < x > in Spanish to represent a /ʃ/ as in the early spelling and pronunciation of the name Mexico?
David Marjanović said,
April 11, 2025 @ 11:06 am
They're not written in different ways; at the time, the difference between J and I was purely cosmetic. (Likewise U and V.) So you get the more elaborate for J at the beginning of the paragraph and a simpler form, which fits into the line, in the middle. The reinterpretation of two of these forms as separate letters came later.
Shakespeare's The Jew of Malta was once titled THE IEVV OF MALTA.
I think so. The distinction between ii, ij, ÿ and y were fluid; even much more recently, Dutch has ended up with a letter ij (uppercase IJ as in the river IJssel) where Afrikaans systematically uses y instead.
X still mostly means [ʃ] in Portuguese today, and it did in Vietnamese when Portuguese missionaries first wrote it. I don't know what's going on with Japanese here, but the key may be the fact that the /s/-/ʃ/ distinction of Japanese depends entirely on loanwords. Languages without such a distinction, such as Greek, Finnish or Dutch today, or evidently Old French, Middle High German, Classical Latin and many others, generally have a single phoneme with a phonetic range around the middle between [s] and [ʃ]: a "retracted [s]". People familiar with a /s/-/ʃ/ distinction often find that sound at least as close to their /ʃ/ as to their /s/, and transcriptions can vary accordingly. However, in Japanese, 1) the modern /s/ is instead between [s] and [θ], so it would seem unsuitable for a Portuguese x; and 2) the mentioned loanwords are mostly Sinitic, so most of them were already present in Japanese long before the first Portuguese missionaries showed up, right?
David Marjanović said,
April 11, 2025 @ 11:08 am
Maybe Japan has made me allergic to coda consonant clusters. Or the sun is shining too directly into my eyes.
…m
…s
David W said,
April 11, 2025 @ 2:49 pm
Shakespeare?
dictionary.com has a short article on I and J at https://www.dictionary.com/e/j/
PeterL said,
April 11, 2025 @ 4:54 pm
Interesting that long vowels or consonants aren't marked. (I'm assuming that pronunciation hasn't changed too much since 1591) And the use of spacing is somewhat inconsistent (e.g. "finari" ("hi nari" in pre WW2 Japanese).
I wonder if Barreto had learned hiragana/katakana? – because they distinguish long and short vowels. (Although IIRC, no distinction was made between "ho"/"fo" and "po" – the "handaku" was a later innovation)
Yves Rehbein said,
April 11, 2025 @ 5:49 pm
> They're not written in different ways; at the time, the difference between J and I was purely cosmetic.
@DM, they are very different. I might say the initial capital is … (decorative), but font metaphors do not apply to manuscripts.
> I was later adopted by Semitic groups to describe the word “arm”
Weird quote (dictionary.com editors)!
Anyway, iota as it's known to mean a small bit surely was just that: but a scratch. Crooked iota may be reminiscent of the arm hieroglyph, that comes from Old Egyptian d and changed to ʕ? Ayin ‹ע› still represents e in Yiddish orthography, very different from yod https://en.m.wikipedia.org/wiki/Yiddish_orthography The change is evident in Semitic Phonology on account of vowel coloring in Akkadian https://en.wikipedia.org/wiki/Semitic_languages#Phonology
You are missing the part in the story where the Greek invented the vowels and Gesus turned water into genuine horseshit.
Jonathan Smith said,
April 11, 2025 @ 6:55 pm
Re goxutxe, FWIW even modern (go)shussei has "geminate" ss, for which orthographic "ts" would hardly be a crazy choice — plus the middle syllable would have had -t in the Sinitic donor language; maybe such a segment was still around in 16th cent. Japanese.
Chris Button said,
April 11, 2025 @ 9:38 pm
The name of the Brazilian TV star Xuxa (who was famously also Ayrton Senna's girlfriend for a while) is a classic example of how to pronounce "x" as [ʃ] in onset position in Portuguese. It gets a little confusing when it's not an onset though.
When I was learning Portuguese in Sao Paulo, the most salient difference from The Rio accent for me was the pronunciation of syllable final -s as [ʃ] in Rio (which I believe is common practice in Portugal). Online, there's a 1955 article by Hart on 16th century Portuguese that says orthographic "x" was often used for "s" in coda position, which he assumes was still [s].
Perhaps "x" seemed most appropriate to capture the allophonic variation in Japanese at the time. I noted above that Portuguese transcriptions provide crucial evidence for obstruent codas in Sino-Japanese (otherwise a phonotactic violation in Japanese) through variations like nichi ~ nit.