From Chord to Melody:
Defining Tanka in English

First published in the Tanka Society of America Newsletter 4:2, June 2003, pages 10–12. Also presented as a paper at the Tanka Society of America’s first “Tanka Day,” held on 30 June 2003 at the Collegiate School at 260 West 78th Street in New York City, immediately after the Haiku North America conference at the Dalton School the weekend prior. Written in August of 2000 and June of 2003.       + 

I remember a conversation I had years ago with fellow poets on the subject of tanka—or rather, what tanka is. The definitions we shared all seemed to be missing something. Yes, we knew the history of the genre, that in ancient Japan tanka was primarily a love poem, that in Japanese it traditionally had thirty-one sound symbols in a 5-7-5-7-7 pattern, and that English-language tanka were typically presented without rhyme or titles in five lines with fewer than a total of thirty-one syllables. But most of the definitions we shared or had read seemed to be exactly the same as definitions for haiku, other than regarding length. One could simply insert the word “haiku” wherever “tanka” was mentioned, and change the references to length, and the definition would still seem to be accurate. Yet this obviously felt inadequate. So what was tanka? It’s a question that still needs answering—and with a clear answer perhaps tanka can be revitalized.

        The situation still persists, however, that haiku and tanka definitions are often interchangeable. As a recent example, consider the following judges’ comments, written by Marianne Bluger, about the winners of the 2003 Tanka Society of America tanka contest. Wherever you read the word “tanka” or “poem,” try inserting “haiku” instead:


All fine tanka are lyric poems characterized by their formal structure, and by the immediacy and intensity of one particular moment. They tell us where a body is in the world; and what is going on in the poet’s heart and mind. And then, by the force of the image(s) and by the clarity of both perception and language, the reader is transported from those very particulars to some kind of transcendence. If the tanka poet is gifted, a sense of awe and privilege arises from being drawn into communion with this other life, this other world.

        Note, too, that all fine tanka have an unmistakable freshness about them. They are not obscured with thick imaginings or overlaid with vague, grey thought.

        Every one of these poems achieves “tanka splendour,” that lift-off into truth and loveliness that brings the reader solace and joy, however sad the subject matter. In each case, we get a feel right away for who the poet is; we sense and love the authentic “voice.” Tanka poems are naked poetry. The soul is bare, somehow. There are minor flaws in some of these poems, but in reading each of them, you will experience a deep moment with a fine poet.

        None of these poems is marred by sentimentality, none trades in clichés, none is overwritten and none is merely a one-sided report of either a natural setting, or else an inner event. All are balanced, and all read well. For at the end of the day a real tanka must be a poem and must read like poetry. There must be a natural ease and genuine rightness to the words chosen to express what is deep and what is present in the outer and inner things.


        When considered in reference to haiku rather than tanka, the preceding comments are still amazingly true, thus, if we want to differentiate tanka and haiku, we could easily conclude that something is inaccurate or missing from this description. Furthermore, with Marianne’s use of the phrase “this other world,” I can’t help but think of haiku rather than tanka, thanks to prominent American novelist Richard Wright who used the phrase to title his well-known haiku book. Indeed, what Marianne calls “this other world” of English-language tanka is perhaps not yet sufficiently distinct from the world of haiku.

        The first anthology of English-language tanka that I know of in the United States was Footsteps in the Fog, published by my press in early 1994 (it preceded the estimable Wind Five Folded from AHA Books by four or five months). In the introduction, I wrote that tanka is “a lyrical five-line poem often expressing deep emotions through the use of natural symbolism.” There is more to tanka than this, of course. The late Pat Shelley, an extraordinary tanka writer who would have been delighted to know that the Tanka Society of America now exists, wrote the following about tanka in Footsteps in the Fog. Her viewpoint may serve as a fundamental perception of the genre:


Tanka in English is a small lyrical poem that belongs to everyone. Still written in thirty-one or fewer syllables in five rhythmic lines, as it was over 1,200 years ago, it can embrace all of human experience in its brief space, with emotions of love, pity, suffering, loneliness, or death, expressed in the simplest language. It may sometimes seem fragmentary or lacking in unity because it is more intuitive than analytical, using imagery rather than abstractions. . . . One of the more challenging (and charming) of its elements is the subtle turn at the center of the poem, something unexpected perhaps, usually occurring after the second or third line as two seemingly unrelated events, images, or ideas are brought together, something less than narrative, an elliptical space that adds pleasure to our listening. Tanka is about our everyday lives in its smallest happenings, a little song of celebration.


        Each of the seven poets in Footsteps in the Fog had brief essays on tanka in the book at my invitation. As a publisher, I felt the need to add this element to the book to help clarify this genre of poetry. That clarification, nearly a decade later, is still needed.

        Others have sought to define tanka also. Most scholarly definitions have focused on tanka in Japan. What has been rare are definitions of English-language tanka. In 1994, the Haiku Society of America definitions committee led by William J. Higginson offered the following draft definition:


TANKA. The typical lyric poem of Japanese literature, composed of five unrhymed metrical units of 5, 7, 5, 7, 7 “sound symbols”; tanka in English have generally been in five lines with a total of thirty-one or fewer syllables, often observing a short, long, short, long, long pattern. Tanka usually need no titles, though in Japanese a “topic” (dai) is often indicated where a title would normally stand in Western poetry. In Japan the tanka is well over twelve hundred years old (haiku is about three hundred years old), and has gone through many periods of change in style and content. But it has always been a poem of feelings, often involving metaphor and other figurative language (not generally used in haiku). While tanka praising nature have been written, and seem to resemble “long haiku,” most tanka deal with human relationships or the author’s situation. In the words of Sanford Goldstein, “behind the scene is the autobiographical moment of the poet” (“Tanka: Off the Back Burner,” Frogpond XV:2 Fall-Winter 1992). The best tanka harmonize the writer’s emotional life with the elements of the outer world used to portray it.


        The Spring 2001 issue of the TSA Newsletter offered the preceding definition to members of the Tanka Society of America, and proposed that TSA come up with its own definition of tanka or adapt the HSA draft definition. But perhaps a fully reliable definition will always prove elusive. As I also wrote in Footsteps in the Fog, “the moment English tanka becomes static and unchanging is the moment its spirit dies.” This is true, I think, as I also wrote, because tanka has “an ineffable spirit, a spirit that defies description but once tasted is never forgotten.” Tanka poets are free to define the mechanics and crafting of tanka, but I propose that one of the ongoing missions of the Tanka Society of America should be not to confine tanka’s spirit but to revel in it.

        Indeed, after I proposed that TSA come up with its own definition, one member wrote to me to say that he hoped we never did so, for he felt that a definition would limit—and thus inhibit—tanka rather than promote and invigorate it. A good definition, though, should never inhibit but inspire. Consider the rules of chess. They are exact and strict, yet the variety and imagination possible within the confines of the game are astounding. As Kōkō Kato once wrote about haiku in A Hidden Pond, an anthology of modern haiku in Japan, “I believe that some sort of framework may be useful for people who compose in English too. The effect it has upon the content is an interesting one, rather like the dohyō or ring where sumo wrestlers engage.” Beyond form, it seems to me that while haiku implies emotions, tanka expresses them more overtly and often autobiographically. Furthermore, as a primarily lyrical genre of poetry, tanka is capable of expressing deep emotions in a variety of personal ways. May whatever definition English-language tanka writers end up with be like the wrestling ring, providing a limit beyond which a poem may not be a tanka, yet also providing an attractive playground for perpetual variation.

        I’d like to conclude by quoting Gerald St. Maur from his 1999 Haiku Canada Newsletter article entitled “From Haiku to Tanka: Reversing Poetical History” (also published in the TSA Newsletter, 2:1, Spring 2001):


In going beyond the experience of the moment, the tanka takes us from delight to fulfillment, from insight to comprehension, and psycho-orgasm to love; in general, from the spontaneous to the measured. To achieve this requires a fundamental shift in emphasis: from glimpse to gaze, from first sight to exploration, and from juxtaposition to interplay; in short, from awareness to perspective. . . . It is thus evident that to compose a tanka is to articulate reflectively. . . . It is a shift which, in general, takes us from the simple to the complex. More pointedly, it moves us from the poetry of the noun to the poetry of the verb; in weaving terms, from the thread to the tapestry; in botanical terms, from seed to plant; in chemical terms, from element to compound; in painting terms, from sketch to picture; and in musical terms, from chord to melody.


        These words may come as close as we in the west have come thus far not just in defining tanka intellectually but emotionally as well, for it is important, I believe, to feel the difference—to know the difference in the heart as well as mind. Or could it be, in contrast, that the only difference between tanka and haiku is that tanka has an extra two lines, and there’s no further difference? Do we resist that idea, if that’s the only difference? Should we resist it? Why? Or should we just accept it and continue writing our poems, whether haiku or tanka?

        Whatever our decision, individually or corporately, here’s to making many wonderful melodies with our tanka poems, and to decades of mutual support in cooperation with writers worldwide of haiku, tanka, and related genres of poetry. And here’s to the revitalization of tanka in English—a revitalization that will not depend so much on definitions but on the poets themselves and on their tanka. It’s a revitalization that I hope will strengthen tanka for decades to come.


Note: At the Tanka Society of America “Tanka Day” in New York City on June 30, 2003, the concluding event is an open discussion on the definition of English-language tanka. The TSA definition committee will take input from this discussion and input from various advisors to the committee (including tanka scholars and translators) to propose a definition to members of the society. All TSA members are invited to offer comments on this topic by writing to Michael Dylan Welch or Pamela Miller Ness.



The Tanka Society of America definition committee (essentially just me and Pamela Miller Ness) never finished its work. This was more the result of having higher priorities than anything else, but also due to the concern of a few members who felt that tanka shouldn’t be defined at all, and others who felt that, for those who wanted definitions, the HSA definition (quoted in this essay) was good enough. I’m not sure I agree, but any attempt to define tanka would surely be problematic. I still have my files of correspondence on the subject. Since our committee’s attempt, others have tried to define tanka also, as have many others before us.

—2 November 2009