I have heard that くゎ died last in the Kyoto dialect, so it sort of makes sense that the Emperor would use it.

Kwannon is definitely a cooler name than Kannon.

Also, I live in Toyama and sometimes hear tiny y-s in 円. Particularly if the word before ends in ん, you get a kind of "ni sen'yen" sound.

"Were the morals ... in the original, or did the Jesuits add them for the benefit of their local audience?"

I can't comment on the text in question, but there is a long history of "Aesop" translations, mostly adaptations. Sometimes a familiar proverb was substituted in the course of translation, even if the "fit" wasn't very good.

I don't know if the Jesuits were working with a moral-in-final-position text, which is what we now think of as standard, but they were certainly familiar with that format. I would not be surprised if they followed precedent in adapting Japanese proverbs, or even tried to work up their own proverbial-sounding formulations.

However, there is considerable variation found in the long history of the "Fables." The name "Aesop" is used to cover both Greek versions, in verse and prose, many anonymous, some with credited authors, and Latin adaptations (notably the verse "Phaedrus'), as well as innumerable translations into other languages.

There is a well-informed and lucid short discussion of the problems by Laura Gibbs in "Aesop's Fables" (Oxford World's Classics, 2002), which contains her translations of 600 fables drawn from all of the main ancient versions.

Some tellings of fables contain an internal "moral" expressed by one of the characters. Others have an external commentary. This is sometimes in the form of a title, and sometimes in the now-standard final position. And some are found with both internal and external formulations, but they do not necessarily agree on the "real" message!

Some translators try to impose uniformity; P.F. Widdows' translation of "The Fables of Phaedrus" (University of Texas, 1992) shows instead the considerable variation in a single ancient collection.

There is actually a translation called "Aesop Without Morals" (by Lloyd W. Daly), which tried to get past the childhood associations by omission. And, of course, there are versions of some genuine fables which don't provide any such neat formulation, so this wasn't completely arbitrary.

By the way, I would avoid the attractive-looking 1998 Penguin Classics translation by Robert and Olivia Temple, as "Aesop: The Complete Fables." It is a "complete" translation only of Chambry's edition (Paris, 1925-26) of the diverse "anonymous" Greek versions; alas, both the translation and their commentary are less than reliable.

Carl: I hear you. I also like "Kwaidan" more than "Kaidan". Re the extra /y/, do they say "yedamame" instead of "edamame" too? Because if so, that's totally awesome and you should put some sound files online.

Ian: Thanks, that was very informative. Clearly the Aesop situation is more complex than I thought. (Bummer about the Penguin edition, though.)

This throws what I thought I knew about the は行 into chaos. Cool.

So was the whole は行 bilabial like modern ふ? I assume we're not talking about lip-biting English-style "f"s, right?

Yeah, as I understand it, there was never a labiodental "f" like the one we have in English. It went /p/ (bilabial plosive in ancient Japanese) → /F/ (bilabial fricative, like the sound in modern ふ, starting in the Heian period sometime IIRC) → the situation today (a bunch of different realizations, some of which have fossilized and are not considered to have anything to do with the は行 any more, like the /w/ in 会わない)

Matt,
I actually wrote a short piece about the history of Japanese /p/->/F/->/h/ for a historical linguistics class. Poke me at my geemail (silentdibs) and I'll dig it up and post.

Found it...
AND my IPA fonts...

Thanks. I could have been clearer about the source of the Penguin translation; it is based mainly on the 1927 second (and shorter) edition of Chambry, a collection which is restricted to prose texts.

For those interested in numerous versions of the fables, including ancient and medieval original-language texts, there are invaluable resources at Laura Gibb's website:
http://www.mythfolklore.net/

These include BESTIARIA LATINA and 1000 LATIN PROVERBS, and, most notably, AESOPICA.NET. The latter includes Gibbs' published translations, although not her notes and commentary, along with four other English versions, starting with Caxton in 1484.