In lieu of an abstract, here is a brief excerpt of the content:

  • Forms Transformed: Japanese Verse in English Translation
  • Hiroaki Sato (bio)

In translating poetry, no one is wrong, except when the literal deciphering is. I remember, for example, a translator working on some poems of Hagiwara Sakutaro (1886–1942) and misreading the Chinese character (kanji, in Japanese) for sara (“plate” or “dish”) as the one for chi (“blood”). This was years ago, and I no longer remember the lines, as translated by him, that contained the misreading, although, checking my own translation, I see it must have occurred in places such as “I would like to steal and eat that love-plate of skylarks, which gleams in the sky,” “I ate too much of the plate of cabbage this morning,” and “I looked through the whitened plate.” Sakutaro’s imagery in these descriptions may be odd enough for this particular misreading not to matter much—at least to the reader ignorant of the original; still, it is an error.

But Hagiwara Sakutaro is mainly known for his jiyu-shi (“free verse”), which does not employ any discernible syllabic patterns. My focus in this essay is on two traditional forms: the 5-7-5-7-7-syllable tanka and, Japanese verse forms having developed genealogically, its grandchild, the 5-7-5-syllable haiku. Most translators routinely render these in five and three lines, obviously because the two forms consist of five and three syllabic units. The question is: does the 5- or 7-syllable unit constitute a “line”? Also, in view of the recent emergence in the United States of translators who employ the same syllabic count in their translations of classical Japanese verse, you might ask what happens when they do so.

The first thing you find when you step out of the realm of traditional forms, where inherited notions may hold sway, is that, yes, the 5- and 7-syllable units can each be a “line.” Miki Rofu (1889–1964), for example, [End Page 100] wrote “Furusato no”—a poem that became famous because it was set to music—in lines that alternate 5 and 7 syllables:

Furusato no  ono no kodachi nifue no ne no  urumu tsukiyo ya.

In her village   in a stand of trees by a field a flute’s sound   blurs in the moonlit night.

For that matter, the rendition by Ueda Bin (1874–1916) of Paul Verlaine’s “Chanson d’automne,” likely the most memorized French poem in Japanese translation, is done in a series of 5-syllable lines.

Yet, at the end of a prolonged period of verse experimentation—from the latter half of the Meiji Era (1868–1912) and well into the Taisho Era (1912–1926), when Western notions of poetry and poetics swamped the land in one wave after another—Hinatsu Konosuke (1890–1970), a scholar of English literature and a poet, concluded that neither the 5- nor the 7-syllable unit had “a general suitability in engendering poetic effect in Japanese.” Indeed, among non-tanka and non-haiku poets, the usual practice was to compose poems with lines variously combining the two classical syllabic units. For example, the lines in the poem “Isago wa yakenu” (The sand is burnt), by Kambara Ariake (1876–1952), employed 7-5-7, 7-5-5, 5-5-7, or 5-7-5 syllables. In another famous translation of his, Ueda Bin gave 7-5-7-5, or a total of twenty-four, syllables to each line of Baudelaire’s sonnet “L’albatros.” Also, before free verse originating in France reached Japan and prevailed, poets worked out new syllabic patterns, such as 6, 8, and 9, and combinations thereof.

So, first, it may legitimately be asked: if the 5- and 7-syllable units in tanka and haiku are to be automatically regarded as “lines” because each forms a pattern, what to do with those “lines” Ariake composed? What to do with newly created syllabic patterns?

Second, free verse also affected the realms of tanka and haiku, prompting a substantial number of poets to stop using syllabic units and counts. I’ll give one example from each genre.

Toki Aika (1885–1980)—who critically influenced the famous tanka tri-lineator Ishikawa Takuboku (1885–1912) and continued...

Share