Searching out an old letter

Michael Dirda's recent essay on Sei Shōnagon's Pillow book (specifically, the new-ish translation for Penguin by Meredith McKinney) is a fine 101 on the book, complete with balanced appraisal of its treatment in English so far and even suggested European reading along the same lnes. One passage that caught my eye was this comparison of McKinney's style with that of Ivan Morris, previous Penguin translator and go-to Heianographer:

[... A]mong "Things that make you feel nostalgic," McKinney includes this item: "On a rainy day when time hangs heavy, searching out an old letter that touched you deeply at the time you received it." But here is Morris: "It is a rainy day and one is feeling bored. To pass the time, one starts looking through some old papers. And then one comes across the letters of a man one used to love." The McKinney version is doubtless accurate in its succinctness and may even reflect a slightly different original text, but Morris's words catch us by the heart.

I'm not sure which text McKinney used — both Penguin Classics and ANU's Asian Studies department have woefully inadequate web presences — but Morris says that his work is "based primarily on the Shunshō shōhon [春曙抄本] version as edited by Kaneko Motoomi [金子元臣] in 1927 and on the Nihon Koten Bungaku Taikei edition of the Sankanbon 三巻本 ["three-scroll manuscript"] version edited by Ikeda Kikan [池田亀鑑] and Ikigami Shinji [岸上慎二] in 1953." According to Wikipedia, the Shunshō shōhon was edited by Edo scholar Kitamura Kigin 北村季吟 and based on the Nōin 能因 manuscript.

Let's take a look at a few versions of the original from three-scroll- and Nōin-based manuscripts, then:

1. また、をりからあはれなりし人の文、雨などふりつれづれなる日、さがし出でたる。
2. また折からあはれなりし人の文、雨などの降りて徒然なる日さがし出でたる。
3. あはれなりし人のふみ。雨なとふりて。つれ/\なる日さがし出たる。

"1" is from Ikeda Kikan's version for Iwanami, based on the "three-scroll" manuscript. (I have it in bunko form.) "2" is from the University of Virginia's e-text of the Nōin manuscript. And "3" is transcribed from Kyushu University's online scan of an Edo-period published version of the Nōin text.

Both content and form are similar in all three. The main differences are use of kanji, and the missing mata + worikara ("and" + "at the time") at the start of KU's version, which could be scribal error. I suspect that this passage is much the same in all manuscripts, and in particular I doubt that any has it chopped up into sentences the way Morris's translation is. (Note that Edo usage of "。" doesn't correspond to a full-stop "。" today. Think of it as a general "pause" character.)

A painfully strict translation retaining clause order would go something like this:

[And] when a letter from a person that moved one's heart [at the time] is, on a rainy, idle day, searched out.

The succinctness of Meredith's version does, as Dirda suspects, hug the outline of the original more closely. Morris's technique of dividing one polyclausal Heian sentence into several simple English ones makes his version more cinematic and immediate, but does not convey to the reader the texture of Heian literature, the long and twisting chains of thought finally tied off by a verb in the form reserved for ending statements.

The other trade-off is that the assumptions Morris makes to set the scene are not always justified by the source. Here, Morris's image of someone hit by a wave of forgotten emotions when they stumble across an old love letter is certainly more moving than the idea of someone searching out the letter because they want a fix of those emotions to pass the time. But the fact remains that the verb sagasu means "search for" or "search out", and never (as far as I can tell) "come across [by chance]." Maybe Kigin disagreed.

In closing, just because this information should be Googlable, here is McKinney's treatment of the famous first line:

In spring, the dawn — when the slowly paling mountain rim is tinged with red, and wisps of faintly crimson-purple cloud float in the sky.

Popularity factor: 17


I swear that Roy Andrew Miller has this thing in one of his books where he basically says, "Anyone who really thinks Sei Shoganon actually existed is an idiot." What's up with that?


Really? As far as I understand she's about as well-attested as any other non-Emperor from that period. There are even catty references to her in other books.

Ryan posted something about this last year. It sounds like Miller's thing is not that anyone who thinks she existed is an idiot, but that anyone who thinks they know the historical person Sei Shonagon by reading a modern edition of the work attributed to her (much less in translation) is an idiot. Which is true, but trivial, and not really helpful. Of course there's a lot of reconstruction and romanticization involved, same as reading any text from that long ago. The most avid Ren Fair enthusiasts will admit this to you. I bet Morris would have, too, if you'd pinned him down on the topic. Even in modern times, we're all just reconstructing each other in our heads.

Leonardo Boiko:

I miss that missionary guy who started a blog version of the Pillow. A shame he didn’t keep it up. (Archive: http://web.archive.org/web/*/http://blog.simon-cozens.org/shonagon/ )


"but Morris's words catch us by the heart."

Um, if the 'us' is supposed to include me, then I disagree, because it reminds me of something entered into the Bulwer-Lytton contest.

You make a very good point that さがす(both 探す and 捜す, which are two different concepts) and 出くわす have different meanings, and Morris may have gotten carried away in his own mind before penning the translation.

By the way, anyone who thinks they know Sei Shonagon by watching a piss-poor Peter Greenaway flick is probably an idiot. Or an asshole. Or both.


”As far as I understand she's about as well-attested as any other non-Emperor from that period. There are even catty references to her in other books."

That's just in Murasaki's diary, correct? Perhaps: conspiracy!

I'd say it's more "she's about as well-attested as minor nobility from that period." We have much more on Michinaga and Sanesuke, Fujiwaras extraordinaire. And there's more Sei (admittedly, in different sorts of sources) than, say, Otsuki such-and-such from secretarial.

And more on the Kamo Saiin (Senshi at the time, I think?), than on Sei.


I started reading that review and danger, and ouch. A prime example of why literature specialists should be banned from the Heian Period until we finish decontamination. Just, ouch.


Oh, come on, I winced at the "fairy-tale" thing, but thought it was balanced out by the acknowledgment that only a couple of thousand nobles got to enjoy the delightful moping. That's more than most overviews of the scene will admit.

Maybe I should have said "non-Fujiwara"...


Also, if it were for the Bulwer-Lytton contest, it would surely start "It was a dull and gloomy day..."

Sgt. Tanuki:


I'd say the problem isn't literary specialists, but literary generalists: I don't think many Heian specialists would repeat some of the canards in that review. But nobody's going to ask a Heian specialist to write for a mainstream publication. Instead they ask a generalist like Dirda... You get what you pay for.

Actually thanks to your exploration of the original, I think I prefer McKinney's now over Morris's. The succinctness of it heightens the poetry, rather than diffusing it like Morris. And I like "a rainy day when time hangs heavy."

But, is it just me, or is not pretty clear that "あはれなりし" refers to the "人," not the "文"? Why does McKinney decide to eliminate the guy altogether?


Morris's version doesn't "catch me by the heart" either. The use of "one" is distancing, especially when repeated over and over (I know, it's also a European thing.) And personally I prefer the succinct to the wordy. Morris makes me go tl;dr.

"At the time" is what really grabs me. Yeah, back then you thought it was a big deal.


Sgt. Tanuki: Point taken, yes, and that's the problem of painting with a broad brush.

But I do get into debates with literary specialists on the quality of lived experience in the Heian Period, and I think it's due to the type of sources. Where they see more aesthetic concentration and polishing, I tend to see people going through forms and checking to see the signatures match; and in illness, they tend to see more cases of possession than I do. These are minor points (to an outsider), and easily traced to 栄華物語 versus 朝野群載. (Or The West Wing versus CSPAN, perhaps.)

The generalists are still going back to Morris' World of the Shining Prince when they feel obliged to do some research, and perhaps its us bureaucraticy sorts who need to step up and write actually readable books about document checking. (Perhaps, the impossible dream?)


I would rush out and buy two copies of The World of the Inkstained Prince, personally. Also, West Wing vs CSPAN is a great analogy.

> But, is it just me, or is not pretty clear that "あはれなりし" refers to the "人," not the "文"? Why does McKinney decide to eliminate the guy altogether?

Secretly I was hoping that MMS would answer this. I think there is enough ambiguity to potentially refer to the letter rather than the person. Maybe she figured in that case that "人の" was unnecessary because "letter" + "received" in English already implies that the 文 came from another person?


I don't think I'm any authority on classical Japanese semantics. Short answer is that I do think that there might be some ambiguity in the sentence between [あはれなし人の][文] and [あはれなし][人の文] parsings (although in general, the latter seems less likely to me). But for more certainty, I'd want to do some corpus analysis.


This is a fun problem. I guess I like McKinney's translation, because "aware" as an adjective is used to describe linguistic objects so frequently (あはれなりし御一言、あはれなる歌、あはれなる声), so it seems more natural to parse it as describing the letter rather than the person. Genji has the suspiciously similar phrase 「をりからの御文いとあはれなれば」 in the Suma chapter, which Tyler translates as "...made her timely letter especially moving." But looking around I did find one clear instance of 「あはれなりし人」 referring to a past lover, in Utsuho monogatari, which lends some support to the "man one used to love" interpretation.

Boy, I wonder if there are any other instances of irresolvable syntactic ambiguity in this work!


"... and ANU's Asian Studies department have woefully inadequate web presences". True, but is not a department. Did you want to ask her? Her work email is meredith.mckinney@anu.edu.au and she is around at the moment.


Surely a faculty (he said, after Googling) should have even MORE resources than a department!

I sent Dr McKinney an e-mail. Let's see if she has time to drop by.

P.S. I forgot to mention that my sample translation was actually designed to preserve (what I saw as) the ambiguity in the original via careful placement of the relative clause.


Thanks for the interesting discussion. It's wonderful to get some good, informed feedback.

Just to mention first that I didn't use the Noinbon but the Sankanbon text of Makura, so there are numerous differences large and small from Morris' version.

I've checked the relevant sentence in Sankanbon and it's the same. In the Shokakan edition I based my translation on,, there's a stern note here saying that aware narishi refers to the letter not the hito, so I guess I followed the scholars on this rather than pause to question. Myself, I think the sentence should be treated as ambiguous. (Hence Matt's is a good solution.)

Comment season is closed.