Forms of 'In a Station of the Metro' (Accidental Poetry Month, Part 18)

One of Ezra Pound's most famous poems, "In a Station of the Metro", is only two lines long. (I'll quote it in a bit, but I want to hold off for now, for reasons that will momentarily become clear.) Pound himself described it as a "hokku" (which seems to be an archaic term for what today we call (in English) haiku). In describing the poem's composition, Pound quotes two "hokku" (naming neither poet). The first he describes as "the substance of a very well-known hokku":

The fallen blossom flies back to its branch:
A butterfly.

The second Pound describes as written by a "Japanese naval officer":

The footsteps of the cat upon the snow:
plum-blossoms.

I must admit that those two poems, read in sequence, just reinforces something I had long thought: that I don't quite see what the deal is about this particular poem. It's not that I don't like Pound at all -- this, for instance, is a great poem (although not one whose view I agree with, and which is all the more disturbing given where that view ultimately lead Pound to go, politically). I just never thought this quintessential "imagist" poem was all that much. Reading those two hokku, it seems even less special, since it seems that Pound was walking (however well) oft-trod ground.

But a few weeks ago, preparing for one of the lectures on modernism I'm giving in my class this term, I came across, in Jo Anna Isaak's The Ruin of Representation in Modernist Art and Texts (1986), a reproduction of what she presented as the long-lost original form of this poem. Here's how she gave the poem:

            IN A STATION OF THE METRO

The apparition       of these faces       in the crowd:

Petals       on a wet, black    bough.

Stunning. At least for me, it was almost a revelation: a whole new poem. The adding of the spaces, the forced pauses, the breath of the page, brought what had been, for me, a fairly flat two-line image to life.

So I looked around. Most reprintings seem to have ordinary (flat, dull) spacing, but this web site reproduced almost the precise same spacing as did Isaak (the difference being that at the end of both lines they put extra spacing between the final letter and the punctuation mark too).

Just to be sure, however, I tracked down a (scanned page image of) the original magazine publication. Here's a jpg of that:

-- So Isaak was right -- except insofar as the web site editors were right, and there was, indeed, space before each closing punctuation mark, too.

Fabulous!

Alas, however, the story does not quite end there. This web site quotes a chunk of an article called "The Punctuation of ‘In A Station of the Metro’", by one Steve Ellis, which tells more of the story. (The journal the article originally appeared in is given as Paidenma, about which I have found out nothing -- googling turns up mostly reprints of this article. Anyone know what its story is? Update: A commentator says that it was a typo on that page, and that the name of the journal is actually Paideuma, which explains why I couldn't find any information before. Here's Paideuma's home page; here's what the word "paideuma" means (more).)

Ellis does quote the original, white-space bejeweled version of the poem -- although he quotes it without any punctuation mark in the first line -- not, in an article specifically about the punctuation of the poem, a reassuring sign about his accuracy (perhaps the reproducing web site is at fault here? I can't think of another charitable explanation.) But setting that concern aside, Ellis reports that Pound reproduced this version precisely once, in another magazine -- but that subsequent reprintings of the poem (and, indeed, in one that came between the two space-inclusive versions as well) he omitted those spaces. There was a coma added, and then removed again, after "petals", in various versions; but the spaces never reappeared.

A few years later, Pound made one final punctuation change, changing the colon into a semi-colon. Ellis writes of this change that it was

The final and most important change Pound had made to the punctuation... [T]his alteration makes the relationship between the two lines appreciably more subtle and suggestive than was previously the case: the colon tended to subordinate the first line to the second by indicating that by itself line one was incomplete, its function being primarily that of introducing the "Image" in line two which the colon informs us is necessary to complete the first line’s meaning. With the semi-colon the first line is, so to speak, less definitely a "prologue" to the second, the linkage between the two lines being insisted on less emphatically. The relationship between them can be said to be not only more subtle but even more equivocal, and the cost of not foregrounding the "Image" is the possibility, as some of my sample readings indicate, that the semi-colon assists the first line in overturning its subordinate position and becoming foregrounded itself.

I can see Ellis's point about why the semi-colon is better, although the colons in the other hokku Pound quoted would seem to argue the other way -- if, indeed, the colon is traditional in the form. (Do they have colons/semi-colons in Japanese? Or, rather, did they when Pound was writing? I haven't the foggiest idea.) At any rate, I would certainly argue against that being the "most important change": I think the most important change was his decision -- his quite unfortunate decision -- to give up those spaces. Which, for me at least, made the poem.

Yet it does seem -- assuming that our doubt about Ellis's reliability doesn't come to anything -- that Pound himself gave the spaces up. And it's his poem, alas, so I suppose they're gone. Although Auden revised many of his poems in versions that most poetry lovers simply ignore --- indeed, he cut from his "complete poems" one of his very best poems, which most Auden fans (I among them) count among their favorites. If we reject Auden's revisions, why not Pound's? But I don't think Ellis is right about the semi-colon; there's no question, certainly, that I would trade the semi-colon (however much it adds) for the spaces. If that's the trade -- original versus final -- then I'd go with the original.

But do we have to trade? To be sure, Pound himself never published a version with both semi-colon and spaces. But the poem is there, in the language, waiting to be read. See?:

            IN A STATION OF THE METRO

The apparition       of these faces       in the crowd  ;

Petals       on a wet, black    bough  .

Maybe Pound didn't write that. But if it's a better poem then the almost-identical two he wrote, do we care?

A complex philosophical question, that, and quite possibly not one with a determinate answer.

For my own part, however, that is the version of this cluster of nearly-identical poems that I like the best; and the one that I will most often reread in my own memory.

(File under 'possibly-interesting-links-I-found-while-researching-this': Metafilter discusses the poem here.)