Friday, October 17, 2008


Opening the current issue of Ribbons , the journal of the Tanka Society of America, I find once again another writer's attempt to suggest something of what tanka in English might be. Of the several points offered, the presence of a poem formatted on five lines is once again offered as part of what defines tanka.

While it is conventional to format tanka in five lines in English, it is not required to do so, and exceptions have occasionally been published. I have addressed alternate lineation in tanka in an article in Modern English Tanka. Since then my understanding of tanka structure and definition have continued to refine.

Specifically, the reason why tanka is written in five lines in English is because that tanka is a poem of five poetic phrases. The easiest way to depict those phrases is with line breaks. However, this leads to poets writing things that don't have five parts, but because they are formatted on five lines, they are accepted and published as tanka. No wonder critics have difficulty in distinguishing between English language tanka and short free verse!

There is also a general misunderstanding between free verse and 'unfree' verse. The opposite of free verse is metered verse, eg, verse in which something is counted. Traditionally in English it has been meter, but it could be morae, sound units, lines, phrases, or any other subdivision. In tanka what is counted is 'poetic phrases.' Thus tanka is not 'free', it must conform to this expectation. Because it conforms, it is 'formal verse.'

Formal verse is verse written to a recognized form. The tanka form is clearly understood -- anyone who is a fan of tanka can recite by heart that it is form that originated in Japan, consisting of five phrases of 5-7-5-7-7 syllables. Once upon a time in literary history Western poets were quite strict about their forms, but they managed to invent new variations of them all the same, with nonce versions of the forms seeing print as well. The sonnet, for example, now has numerous recognized variants, but it is still a sonnet and the relationship between any given variation on the original forms is identifiable.

Just so with tanka; just the numbering of syllables does not serve in English, but I don't believe that formulas based on line length are an adequate definition. It is quite possible to write a poem with something other than five parts and arrange it visually so that it appears to meet the short-long-short-long-long pattern -- but when we are adhering to a format, that is not the same as adhering to a form. Tanka in Japanese are typically written in one or two lines and can be subdivided however the calligrapher pleases. Clearly, the format is not the form.

Having studied the matter extensively, I have noted that certain poems retain their 'fiveness' regardless of how they are formatted, although admittedly, changing the format does change emphasis. The line break is a powerful tool well entrenched in the Western literary tradition. Even so it is quite possible to recognize a poem as tanka that is formatted as two lines, or prose, or even in some other format.

The consequence is a great many short poems are published as tanka that do not adhere to the form. That doesn't particularly bother me since I have always been an advocate for 'tanka and related forms.' Critics of tanka are therefore missing the point. If a journal stated that it published 'sonnets and related forms,' would they claim that it was impossible to know what a sonnet is and that free verse is being published as sonnets? I think not.

On the other hand, there are forums that publish any five line poem and think it to be tanka. Some of these are amateur forums where it is only to be expected that a naive definition prevails, but the larger, more prestigious journals ought to be clear about the matter. Modern English Tanka , for example, is very clear about its expansive scope of using tanka to establish a new lyric poetry in English. I myself am not convinced that lyricism is part of the definition of tanka, but I admit that it is a very common treatment. Lyricism, which is an aesthetic consideration, appears to have replaced form as a defining principle of tanka in many people's minds.

Given the differences between Japanese and English it is inevitable that different treatments will develop in English. How many hundreds of different ways has Basho's frog poem been translated? We cannot say that any one of them is the best translation, but even in the ones we all agree are terrible translations, we can see the shadow of the original. When translating an entire form into English, it is only to be expected that myriad adaptions should result. Therefore, when critics claim to be unable to perceive the difference between tanka and free verse they are coming about it the wrong way. The question is not whether a given poem can be discerned from free verse; the question of whether a given poem's relationship to tanka is detectable.


  1. On an unrelated note, I would appreciate your thoughts concerning the following comment by haiku poet, critic and theorist, Hasegawa Kai, interviewed in the latest issue of "Simply Haiku":

    "Among haiku, there are those that posit one thing (A = B) and those that use juxtaposition (A + B). Of
    the two, juxtaposition is a technique in which two completely different things are combined to describe a single world of harmony. Of course, between these two, 'ma' is born. In other words, juxtaposition is a technique for creating 'ma' in haiku. In contrast, the haiku that posit one thing have 'ma' only on their outside edges.

    "The important thing to remember is that two completely different things are juxtaposed, and not two similar things. That is because 'ma' cannot be born between two similar things."

    As a gay writer of both tanka and haiku, this sort of statement gives me pause. Is the comment subtly hetero-biased if not heterosexist? I'm no activist, but am left to wonder: is it true that two similar things can only produce the outer edge of "ma"? Or is the analogy merely convenient, even esoteric? I understand the inherent need for creative tension but am troubled by the lack of nuance in the haijin's statement.

    Also, Hasegawa Kai is speaking about haiku. Do you perceive a difference in application (or none at all?) where tanka is concerned?

    Sorry to take up blog space. I could think of no other way of contacting you regarding the matter. Feel free to delete from your comments section and contact me directly at, if you feel inclined to answer at all. Your past post on homoerotic poetry as well as your own poetry and scholarly essays prompt me to seek your reaction.


    Brian Zimmer
    Ontario, Canada

  2. Thanks for your comment, Brian. It's an interesting one, but since I'm not a haiku poet, I really can't address the critical remark you quote. I think tanka is a very different animal from haiku and the vast majority of haiku criticism simply doesn't apply to tanka.

    We can have tanka of one, two, three, four, five or more parts, if by 'part' we mean 'structures of meaning.' And these parts may be similar or different. There is no parallel in sexuality, unless we conceive of a more varied sexuality (which I do). Gay, straight, bi, transgendered, celibate . . . Sexuality is not a duality any more than poetry is.

    The heterosexual bias of haiku and tanka is far more flagrant than analysis of esoteric literary terms. It is difficult to get homoerotic poems published as such; it was the realization that when placed in standard tanka journals, profoundly homoerotic poems of mine would be interpreted as heterosexual poems. Hence I was pleased to publish them in 'Love Letters : Homerotic Poems of Love and Friendship' at 3Lights Gallery, and I'm grateful to Liam Wilkinson for being open-minded enough to publish it.

    Returning to the comment you quote; I don't see a heterosexual bias in it because I can conceive of two males or two females or two of anything being sufficiently different from each other to provide the necessary 'ma' he cites as desirable in haiku.


  3. Hi Kei:

    Thanks for your thoughtful comments on the Ribbons editorial. Since I'm one of the people who talks about the free verse connection, and expresses the view that much of what is published today as Tanka would be better understood as free verse, I thought I would comment. First, I have no difficulty with the idea of "Tanka and Related Forms"; makes sense to me. But the focus of the discussion is, I think, on the form of the Tanka. The concept "related forms" is not generating discussion.

    My basic view about free verse, and I think you agree, is that free verse is poetry that does not use counting as a mechanism of organization. The Tanka tradition does use counting as a method of organization and has done so for a very long time. But my sense is that most western Tanka poets do not use counting, and it is for this reason that I think they are writing free verse and not Tanka. At this point I always feel constrained to say that I have no objection to free verse since the Psalms, Whitman, and Dickinson are some of my favorite poetic sources. It's just that I don't think it is the case that Tanka are free verse.

    After reading your post, in some ways I think in the past we may have talked past each other and it feels to me that we may be more in agreement than I had previously suspected.

    Best wishes,

    Jim Wilson

  4. Jim:

    The average tanka poet isn't very well educated about the form and genre of tanka. They read things they like and try to write like that. In a very real sense, tanka is largely an 'amateur' field. That is to say, it is the work of people who love it and their skills and knowledge may vary widely.

    While I explicitly embrace 'tanka and related forms', I do it to avoid the war over tanka definition. I don't think there is a universally accepted and satisfying definition of tanka out there. As noted, when translating from another language, considerable leeway must be granted. Therefore, I'm willing to accept as tanka as a poem that shows its tanka ancestry. That poem might very well fall into other categories as well, such as free verse, but so what?

    I have written tanka with the meter of old English folk songs and pop tunes; that doesn't mean that they aren't tanka, it means that they are both tanka and something else. But then, I'm a person of mixed blood, part American Indian and part Euro-American, so resist the notion that things, whether they be people or poems, must be either A or B, and not both at once.

    I'm not sure it helps us understand tanka to say this one is really free verse and that one is really tanka. What do we accomplish with such distinctions, aside from empowering somebody to be the judge of what is 'real' tanka and what is not? I find it more useful to analyze structure and to see how the poem works and why it does or doesn't work.

    For example; if I find a poem written on five lines that has only four poetic parts, I'm probably going to find it incomplete and/or vague. That's because it has substituted format for form, which is rarely satisfying.

    Likewise, if a poem is padded to have five parts, one or more of which is superfluous, I'm going to tell that poet: this isn't a tanka. Let it be a quatrain or senryu or whatever it is that it really wants to be. Sure I could chuck them aside and go, "Ew, this is free verse, not tanka," but that doesn't really explain anything.