Here's an interesting little article about Japanese phonology by Kawahara Shigeto arguing (eponymously) that "Lyman’s Law is active in loanwords and nonce words".

Lyman's law, as any fule kno (and as Kawahara xplain), is "a general phonotactic restriction in Japanese which prohibits two voiced obstruents within the same morpheme," famously observable via its blockage of rendaku in cases where the second element of a two-morpheme compound already contains a voiced obstruent. Or, as Lyman himself put it, "the second part of a compound word takes the nigori [Japanese for "voiced form"]; that is if beginning with ch, f, h, k, s, sh, or t, those consonants are changed into the corresponding sonant ones ... [but] the general rule does not apply ... when b, d, g, j, p, or z already occurs anywhere in the second part of the compound" (Lyman 1894).

There's more to rendaku than this, of course, and even Lyman observed as much, but the one-voiced-obstruent-per-morpheme insight is the interesting one. (Naturally, Motoori Norinaga and Kamo no Mabuchi reportedly had it first, although I wasn't able to find the exact terms they framed it in.)

Anyway, the question of whether Lyman's Law applies to modern loanwords is a complicated one, not least because said loanwords often already feature multiple voiced obstruents.However, Kawahara cites three pieces of evidence, from three separate papers, supporting the idea that Lyman's Law is still an "active" process at work in loan words and "nonce words" (pseudowords specifically):

1. Vance (1980), finding via wug test that Japanese speakers appear less likely to apply rendaku to compound nonce words when the second morpheme contains a voiced obstruent, i.e. the rendaku would violate Lyman's Law.
2. Tateishi (2003), observing that the English plural suffix -s appears in Japanese as /zu/ in most cases, but /su/ when attaching to words already containing a voiced obstruent (e.g. the baseball team names /faitaa.zu/ "Fighters" vs /taigaa.su/ "Tigers") (there are a lot of other factors at play here and Kawahara goes into them if you are interested)
3. Nishimura (2003), arguing that the general tendency to avoid repeated voiced obstruents is the reason why loanwords like /baddo/ "bad" can be devoiced to loanwords /batto/ while loanwords like /reddo/ "red" cannot be devoiced to /retto/

Kawahara then describes three experiments he performed to investigate further. The first two are updates on Vance's 1980 wug experiment. Kawahara more or less replicates Vance's results, except that (1) where Vance told his subjects that the words were pseudowords, Kawahara also tried telling his subjects that the words were Old Japanese, thus theoretically testing for and failing to find a difference in how subjects treated "Japanese vocabulary" and other words; (2) Kawahara's experiment did not reproduce Vance's finding that the proximity of an offending voiced obstruent to the compound word break can affect acceptability of rendaku.

In the third experiment, Kawahara investigated the /batto/ vs /reddo/ issue, for both real words and nonce words, and came up with surprisingly clean graphs showing that the "naturalness of devoicing" is, from highest to lowest:

1. Geminate consonant + Lyman's Law violation (e.g. /baddo/ "bad")
2. Geminate consonant (e.g. /reddo/ "red")
3. Non-geminate consonant + Lyman's Law violation (e.g. /bagu/ "bug")
4. Non-geminate consonant (e.g. /hagu/ "hug")

Interestingly, the spread of variation is wider in real words than in nonce words. In other words, it's more acceptable to devoice a real "Geminate consonant + Lyman's Law violation" word than a pseudo-word with the same structure, and less acceptable to devoice a real "Non-geminate consonant + no violation" word than a pseudo-word with that structure. Perhaps this represents the effects of a historical "strong Lyman's Law" preserved in the lexicon combined with an ongoing weakening of the law itself. That is, it might be that real loan words were affected more strongly by Lyman's Law in the past, so that /baggu/ → [bakku] is even now considered much more acceptable than /reddo/ → *[retto], but nonce words, having no history, are subject only to a weaker, present-day Lyman's Law, producing less polarized acceptability judgments.

References

These are the references from Kawahara's article for the other papers mentioned above, plus Kawahara's paper itself for good measure:

• Kawahara, Shigeto. 2012. Lyman's Law is active in loanwords and nonce words: Evidence from naturalness judgment studies. lingBuzz/001344
• Lyman, Benjamin S. 1894. Change from surd to sonant in Japanese compounds. Oriental Studies of the Oriental Club of Philadelphia.
• Nishimura, Kohei. 2003. Lyman’s Law in loanwords. MA thesis, Nagoya University.
• Tateishi, Koichi. 2003. Phonological patterns and lexical strata. In The proceedings of International Congress of Linguistics XVII (CD-ROM). Prague: Matfyz Press.
• Vance, Timothy J. 1980. The psychological status of a constraint on Japanese consonant alternation. Linguistics 18: 245–267.

No time to write much today, so check out this site a workmate showed me earlier: DictJuggler.net, put together by "Yoichi Yamaoka and his successors, Naoshi Fujimoto, gaia-translators, and Marlin Arms Corporation." The array of dictionary and similar searches is impressive, but what's really interesting is the 翻訳訳語辞典, a dictionary of how various words have been translated in a range of real translations.

For example, searching for 最後 (last, concluding) gives you everything from "(人の)顔だって最後にはどんなふうになるかわからないんだという不安を呼び起こす: make sb worry about sb's face in its closing days" from Murakami Haruki's translation of Mikal Gilmore's Shot in the Heart to "最後に〜を附け加える: conclude the letter with ..." from Edwin McLellan's translation of Natsume Soseki's Kokoro こゝろ. Neat.

Today I present another tale of Man'yōshū editing derring-do, this time featuring poem #1.

Yes, that's right: 1. The very first poem in the entire collection. You might think that copyists would be able to stay motivated long enough to get the very first poem done without transcription errors, especially as it is attributed to Emperor Yūryaku, ancient symbol of the Yamato ascendancy.

But no. The poem is irregular in a number of ways, including metrically, and few editors have managed to resist the temptation to propose a set of regularizing amendments.

Let's compare a couple of translations. First, Helen Craig McCullough's version in Brocade by night: "Kokin wakashū" and the court style in Japanese classical poetry:

O maiden
with a basket,
a pretty basket,
with a scoop,
a pretty scoop,
maiden picking greens
on this hillside:
I want to ask about your house;
I want to be told your name.
In the sky-filling land of Yamato
it is I
who rule everyone
it is I
who rule everywhere,
and so I think you will tell me
where you live,
what you are called.

Next, Earl Roy Miner's translation, from An introduction to Japanese court poetry:

With a basket
A pretty basket,
With a trowel,
A pretty trowel in your hand,
O young maid gathering
Greens on this hill,
Tell me your home,
Tell about yourself.
I hold the power
Over this land of Yamato
Broad under heaven,
And I am a chief known to all;
Famous everywhere,
My power is known to all.
And so to no one else,
Yes, to me alone,
Tell of your home and your name.

There are a few differences that might be explored here, but the one of most interest to me is the first house/name query. McCullough has "I want to ask about your house/ I want to be told your name." Miner has "Tell me your home/ Tell about yourself." The absence vs presence of "name" is not an artifact of translation: it actually reflects different approaches to the text.

Consider the traditional version of the Japanese part corresponding to these two lines, with a fairly standard "reading". Here I'm going to follow Bjarke Frellesvig's practice and use regular text to romanize logographic writing, and italics for phonographic writing:

... 家吉閑名告紗根 ...
... ipye kikana/ norasane ...

So ipye, "house", is written with 家, which is the Chinese character that means house. We therefore don't know for certain that the first person to write it down as part of this poem meant it to be pronounced ipye — but tradition tells us that it is, and that there are no strong completing claims, so we accept it without complaint.

Kikana is written with three characters used solely for their phonetic value: 吉閑名, not their meaning. We are fairly confident that our understanding of how Chinese characters were used for phonetic value is basically correct, so we assume that this part was meant to be pronounced kikana.

Now, if we translate ipye kikana/ norasane literally, we get "I want to ask your house/ Tell me." There is no word corresponding to "name". The character 名 is in there, and this means "name" and is used for that meaning (i.e. logographically) elsewhere in the MYS, but here we are interpreting it as strictly phonetic: the -na at the end of kikana.

So was McCullough's "name" added as something to be implicitly understood in the text, like her "O maiden"? No! In fact, she was working from a different interpretation of the text. Check out the romanized "originals" that McCullough and Miner supply:

McCullough: "... ie kikana/ na norasanu ..."
Miner: "... Ie kikana/ Norasane ..."

McCullough includes a note saying that her line 9 has been "[e]mended to follow the [Shōgakukan] Nihon koten bungaku zenshū edition; the NKBT [i.e. Iwanami Shoten Nihon koten bungaku taikei] edition omits the particle [sic!] na."

That is, the 1957 NKBT edition (ed. Takagi et al) reads the text in the "standard" way I described above: ipye kikana/ norasane (which they write, in modern orthography, "家聞かな/告らさね". In a footnote, they explain that there is a tradition of reading the phrase "家聞かな/名告らさね" (i.e. with /na/ appearing twice in a row), but they don't feel it is supported: for that, you would have to get the /-kana/ out of 閑 alone, and for complex reasons that I won't go into they argue that this is implausible.

I don't know what the NKBZ edition McCullough was working from says, but in the Shin Nihon Koten Bungaku Taikei (ed. Satake et al) edition, which is to say Iwanami's update on Takagi's 1957 NKBT edition, the position is that "家吉閑名告紗根" is a corruption of "家告奈名告紗根", which they read "ipye norana/ na norasane". Thus, not only do they find the extra /na/ for "name", they also change the verb used in the first part: it goes from "I want to ask your house" to "Tell me your house". (They have a couple of older sources for this amendment.)

If we grant that the mistranscription from 告奈 to 吉閑 is possible (and it's not as crazy as it seems, if you consider the calligraphic forms rather than the print ones), this actually is an attractive theory in some ways. First, it forms a better parallel with the end of the poem, where "house and name" are both explicitly mentioned. Second, it unifies everything to the verb noru, which I suppose earns it points for neatness. Third, it removes the slightly suspicious situation whereby 閑 appears in this poem but nowhere else in the entire MYS. And fourth, it brings the use of norasane in line with the other two instances in the MYS which are both "汝が名告らさね": na ga na norasane, "Tell me your name!", with the object explicitly included.

On the other hand, the "con" column has two big issues. The first is that getting the meaning "Tell (me) X" from X norana is a bit of a stretch. Satake et al assure us that it is possible, but it's more usual to see the bare -na used to mean "I want to X" (as in kikana) or "I hope that (third party) Xes."

The second is similar to the one discussed in my last post. The poem isn't obviously broken in its na-less form. It was good enough for Takagi in 1957, for example. Satake et al's proposed amendment is a little too ingenious. It takes me back to studying Old English in university, where the bolder and more ingenious a proposed "fix" was, the more likely it was to be proven inaccurate by later developments in the understanding of the language itself and/or meticulous examination, with better technology, of the primary sources.

Academics standing up publicly to Elsevier! This is very exciting. Thanks, Tim Gowers! Thanks, everyone else who has been working for open access for quite some time now!

On the topic of open access, I thought I might briefly address some of the most common non-financial/technical objections I hear raised about it. (I'm excluding the financial/technical ones because I'm not qualified to argue them in detail; let's just take it as given that the truth lies somewhere between "the current situation is the best of all possible worlds" and "we could switch to an everything-is-free-for-everyone model overnight if it wasn't for greedy capitalists", and deal solely with objections to open access as an ideal.)

Hardly anyone outside academia wants to read academic research anyway.

First of all, this can't just be asserted. Every year, JSTOR turns away 150 goddamn million search requests. Even if 99% of those search requests were from idly websurfing bozos who would have immediately clicked the "Back" button upon not seeing a picture of Katy Perry, that still leaves one and a half million serious requests.

Secondly, even if it were true, that's not an argument against allowing access to people who do want to read the stuff. And if the argument is supposed to be "Hardly anyone wants to etc., so it hardly seems worth addressing the problem" then clearly, the only option people like me have is to keep agitating until some worth is perceived in addressing the problem.

If we allowed everyone to read our papers, we'd have to dumb them down and include so much catch-up information that journals would become bloated and progress would be slowed.

No. This is a straight-up misconception and I don't know why it's so common. Maybe the people who believe this are confusing the call for "open access" (that is, letting non-academics read academic journals) for "greater accessibility" (that is, writing for the general public rather than fellow researchers). If so, let me reassure them: we want the former, not the latter. We don't want things rewritten For Dummies. We don't want editorial control. We don't want everything put on Reddit and voted up and down based on how cute the lead author's cat is. We want academic research to go on just as it always has, jargon and all, except with us allowed to read the results too. The fact that the material is written at a high level is exactly why we want to read it. If we just wanted to read summarized and sensationalized highlights, we'd be content with Slate.

If we allowed everyone to read our papers, the information would be twisted and misused by anti-vaccine activists, creationists, demagogues, bloggers, etc.

The world is already full of con artists, rabble rousers and dupes. There is no conceivable mechanism by which making better information more available worsens this situation. If anything, it should ameliorate the problem: the activists outside academia trying to refute the tidal waves of bullshit sloshing around the noosphere would certainly benefit from being dealt in to the non-bullshit game.

Arguments like these boil down to one single, patronizing principle: non-academics can't handle academic writing. This isn't true, and it's getting falser every day. We're allies, not enemies. One day things will be better, and I mean for everyone.

A haiku by Akutagawa Ryunosuke 芥川竜之介, 1919:

竹切れば寒き朝日や竹の中
take kireba/ samuki asahi ya/ take no naka
Cut the bamboo:/ a cold dawn sun/ inside the bamboo

From the Tale of the bamboo-cutter (Donald Keene's translation):

Many years ago there lived a man they called the Old Bamboo Cutter. Every day he would make his way into the fields and mountains to gather bamboo which he fashioned into all manner of wares. His name was Sanuki no Miyatsuko. One day he noticed among the bamboos a stalk that glowed at the base. He thought this was very strange, and going over to have a look, saw that a light was shining inside the hollow stem. He examined it, and there he found a most lovely little girl about three inches tall.

(Note that she's actually from the moon, not the sun.)

A couple of weeks ago, in the course of translating Kokin shū poem #189, L.N. Hammer noted that itsu wa to wa toki wa wakanedo ("Even though we can [feel this way] at any time", in his translation) is, as constructions go, hard to get your head around.

This is true. From the modern Japanese perspective, it can't be parsed from first principles at all: the particle wa isn't supposed to attach to interrogatives like itsu, presumably because, since they don't double as relative pronouns, they make no sense as topics. So you know it has to be some older, fossilized construction that must simply be remembered as it is, like bekarazu.

And, indeed, if you look it up in a big-enough-ass dictionary, there is often a definition. The Nihon kokugo daijiten 日本国語大辞典 offers two, in fact: one likening it to itsu to itte (which sort of means "at (some/any) particular time" but usually combines with a negative form to mean something like "there is no particular time when...") and another to itsumo wa ("usually", "normally"). However, the latter is only attested from the Edo period, while the former goes back to the Man'yō shū; it also seems closer to what we are looking for in terms of understanding poem #189 (remember poem #189?).

What this suggests to me is that itsu wa is a case where an interrogative is functioning to represent a non-specified but specific member of the class of things it refers to (in this case, times). Here's an example of the sort of thing I'm thinking of, in modern Japanese:

料理を教えるのに、塩何グラム、砂糖何匁などと、正確に出すなら、ねぎを適宜に刻み、塩胡椒少々などというな。なになにを何グラムというような料理法を、科学的文化人の生活だと思っている人がある。科学的文化人とは、塩何グラムではなく、科学する生活態度を身につけた自由人のことである。

If you use precise expressions like 'salt, so-and-so many grams; sugar, so-and-so many monme' when teaching cooking, don't also say 'chop the onions nicely' and 'add a little salt and sugar.' Some people think that living scientific person of culture means cooking methods specifying so-and-so many grams of such-and-such. But being a scientific person of culture isn't about so-and-so many grams of salt; it's about being someone who has mastered the scientific approach to life but remains unfettered by it. (Mikaku baka 味覚馬鹿 ("Mad About Flavor"?), by Kitaoji Rosanjin 北大路魯山人. Undated but presumably from sometime in the first half of the 20th century.)

Here, nani ("what?") is reduplicated to give naninani, meaning "such-and-such": some unspecified (but specific!) member of the class of "things", and nan guramu ("how many grams?") and nan monme ("how many monme?") are placeholders for members of the class of "amounts in grams/monme of a given ingredient (that a recipe might specify)".

So maybe itsu wa to wa toki wa wakanedo is working similarly: "Time isn't divided up so that at such-and-such a time [there is a division]". I seem to recall Ōno Susumu 大野晋 mentioning something about this use of interrogatives, especially in combination with wa, but I can't remember where.

Here's another interesting itsu wa-related story. As noted above, itsu wa can be found in the Man'yō shū. There are four instances, and here, courtesy of the Man'yō shū Search System and the Oxford Corpus of Old Japanese (Frellesvig, Bjarke; Stephen Wright Horn; Kerri L Russell; and Peter Sells, 2012), they are:

#2373 (Kakimoto no Hitomaro):
何時不戀時雖不有...
いつはしも恋ひぬ時とはあらねども...
itu pa si mo/ kwopwinu toki to pa/ aranedomo...
"There is no time at which I do not love [her], but..."

#2877 (anonymous):
何時奈毛不戀有登者雖不有...
いつはなも恋ひずありとはあらねども...
itu pa namo/ kwopwizu ari to pa/ aranedomo...
"There is no time at which I am not in love, but..."

#3329 (anonymous):
... 何時橋物不戀時等者不有友...
... いつはしも恋ひぬ時とはあらねども...
... itu pa si mo/ kwopwinu toki to pa/ aranedomo...
"... there is no time at which I do not love [her], but..."

#3904 (Ōtomo no [Sukune] Fumimochi):
宇梅能花伊都波乎良自等伊登波祢登...
梅の花いつは折らじといとはねど...
ume no pana/ itu pa worazi to/ itopanedo...
"The plum blossom: there is no particular time at which I refrain from picking them, but..."

(I'm thinking of moving the whole blog to Frellesvig/Whitman romanization for OJ, and another equivalent system [probably Frellesvig's, again, for consistency] for MJ; what do you think?)

You will notice that itu pa is used in three ways:

1. itu pa (#3904)
2. itu pa si mo (#2373, #3329)
3. itu pa namo (#2877)

This namo of the third example is, in fact, typically cited as the only namo of its kind in the MYS. There are other namo in there — the sentence-final namo, the Eastern dialect auxiliary verb namo, cognate to ramu in the standard dialect — but they are mere homophones. This namo is a conjunctive particle, the one that grew up to be namu and overrun Heian literature completely.

And, despite being quite common in the Senmyō (a collection of 62 "Imperial edicts", i.e. senmyō 宣命, embedded in the late 8th-C history Shoku Nihongi 続日本紀 and written in Old Japanese prose), this namo only appears once in the MYS... or does it?

Non!, cry Satake Akihiro, Yamada Hideo, Kudō Rikio, Ōtani Masao, and Yamazaki Yoshiyuki, editors of Iwanami Shoten's Shin Nihon koten bungaku taikei 新日本古典文学大系 ("New Compendium of Japanese Classical Literature") edition of the MYS. Invoking an "old interpretation" 古義, specifics not given, they amend the 奈毛 /namo/ of the text to 志毛 /si mo/, citing #2373 and #3329 as examples of the itu pa si mo form, and thereby (a) tidying up the use of itu pa in the MYS, and (b) removing namo entirely from the MYS (making it, perhaps, a form used exclusively in prose).

It's an interesting idea, but it's hard to build a case stronger than the circumstantial one above. In particular, while si mo does appear as 志毛 four times in the MYS (and 志母 another half-dozen or so times), 奈毛 appears in the Senmyō, like, a million times, as does 奈母. (I don't have a good scholarly edition, but something claiming to be the full text of the Shoku Nihongi can be found at j-texts.com.) So it's certainly possible that si mo was intended, but it would be by no means unusual in terms of spelling to have written 奈毛 and meant namo: it's not an obvious corruption in the text.

Personally, I'm not confident enough in my understanding of namo/namu to declare its appearance in MYS #2877 a more egregious instance of entities being multiplied unnecessarily than the proposition that we should amend a text that looks perfectly fine.