Saturday, October 14, 2006

Counting Syllables

Recently some one on a tanka workshop email list pleaded to be told the 'correct' number of syllables to be used in writing English-language tanka. I posted the following response:

There is no correct syllable count for English-language tanka, because tanka are not built on syllables. They are built on 'on' which is translated as 'syllable,' but this is an error. The correct translation is 'mora,' which is a unit of sound. For example, the English word 'stretch' is only one syllable, but three mora. str-e-tch. In Japanese, it would be four or five morae.

Since English prosody is built on the syllable, not the mora, it is impossible to make any direct correlation between Japanese 'on' and English syllables. In short, it is linguistically impossible to count syllables correctly in English for the simple reason that syllables are not what the Japanese count.

Japanese and English are extremely different languages. An English verse contains approximately 35% more information than a verse of equal syllable count in Japanese. Therefore, your syllable count in English should rarely exceed 31. Usually English-language tanka fall into the range of 21 - 27 syllables, but there are no hard and fast rules about this.

Some people adopt various alternative schema to mimic the Japanese form, such as the rubric of 'short-long-short-long-long', the adoption of metered feet of 2-3-2-3-3, and various other forms. Most people writing in English settle for the pattern of five phrases on five lines.

Yet even this has problems; Japanese tanka are written in one, two, or three lines, or whatever lineation happens to suit either the style of the calligraphy or the space available. Since Japanese is structurally different than English, the precise arrangement of lines and words on the page does not have the same impact that the choice of line breaks does in English; the calligrapher is free to place line breaks in Japanese where he pleases, but the same is not true in English.

A small number of English-language poets regard the 'on' or mora as not being the most important unit of the tanka; some poets regard the division of the poem into two parts to be the defining feature. Two- part structure is certainly common in tanka, but not universal; tanka have been written with one, two, three, four, and five part structures.

If two part structure is used, English-language tanka frequently places the break at the end of L3, probably due to the influence of haiku, but it could just as viably occur at the end of L1, 2, 3, or 4 and often does in Japanese. Three part structures are not rare in tanka, four and five parts are fairly uncommon, but unitary tanka (no breaks), also referred to as a 'rush of five lines down,' are well enough known to have inspired the name of the first tanka-only journal in English, Five Lines Down, edited by Sanford Goldstein (one of the most prominent translators of modern Japanese tanka) and Kenneth Tamemura.

In short, the translation of a Japanese poetic form to English is not easy, and no one can make it simple for you. In fact, given the profound differences between English and Japanese, I would argue that 5-7-5-7-7 syllables is about the only wrong way to write a tanka. Poems of this length are far too wordy. The Japanese originals are lithe, supple poems.

My advice is: Aim for the greatest possible meaning in the fewest necessary words, pleasingly arranged on five lines.

~K~

References:

Nakagawa, Atsuo. Tanka in English: In Pursuit of World Tanka. Tokyo: New Currents Internataionl, 1990 [1987].

Gilbert, Richard. 'Stalking the Wild Onji: The Search for Current Linguistic Terms Used in Japanese Poetry Circles.' AHA Books Online. Previously published in Language Issues: Journal of the Foreign Language Education Center.(1999, Vol. 5, No. 1). Prefectural University of Kumamoto, Kumamoto, Japan.

No comments:

Post a Comment