Simply Haiku: An E-Journal of Haiku and Related Forms
| Contents | Archives | About Simply Haiku | Submissions |


An Introduction to Déjà-ku
by Michael Dylan Welch

"There is nothing new in the world except the history you do not know."

--Harry S. Truman

For about ten years I've been collecting examples of what I have coined as "déjà-ku"--haiku that bear some relationship to other poems. These relationships are good in some cases, such as parody, homage, and allusion, and not good in other cases, such as plagiarism, cryptomnesia (remembering someone else's poem without realizing that one is remembering rather than creating it), and simply being too similar or insufficiently fresh or original. In June of 2001 I presented a lengthy paper on this subject at the Haiku North America conference in Boston, Massachusetts: "Déjà-ku: The Problem of Haiku Uniqueness." When that paper is published, I hope it will be helpful to the numerous people over the years who have asked about the issues involved and are deeply concerned about whether what they have written is appropriate or not.

We should all relax. Haiku is such a narrow field of poetry that we should expect to repeat each other from time to time. No one owns experience, and we are surely likely to have similar experiences and use similar words to write about them. So long as we write from genuine and heartfelt experience (though this is not the only way to write haiku--and remember that one can write from memory, or make up certain details for the sake of creating literature, as Basho and Buson repeatedly did), our poems will at the very least be well-intended, and hopefully speak to something real and authentic in human life. The fundamental pleasure of haiku is that it gives voice to experience, and we nod or smile in recognition when we read a poem that describes something we've also experienced. Our own feelings in reaction to the same experience magnify the poem, and we complete the haiku, sometimes called an "unfinished poem," by adding our own experiences. The point is not to shy away from shared experiences, but to celebrate them. Haiku could not work if both the writer and reader did not have a significant overlap in experience. The poet, in fact, often brings to the reader's attention something that the reader already knew, but may not have been aware of knowing. If there's any problem with this overlap, it may lie in when two or more poems describe the same experience too similarly. But there, this is only a possible problem if the poems are too similar and if one wishes to publish.

There's the rub. Writing similar poems should never be a problem. Let yourself be free to write. Editing and selection comes later. Only when one wishes to publish might one have to think about whether a poem is fresh or distinct enough. If a poem is too similar to something already published, then perhaps the newer poem should not be published, for only one person can be "first to the patent office." Editors will also help with this matter, as best they can. This step requires a certain degree of "haiku literacy"--knowing the literature well enough to know whether certain subjects have been written about, and in what ways. This is complicated by problems of translation and various wordings for the same poem. No one can be expected to keep up with all haiku in Japanese and English, including all the translations going back and forth. Thus I think we should all be understanding when a poem may get too close to another. There are limits, but unless it's willful plagiarism, I think we should be tolerant and understanding of each other and the similarity of some of our poems.

In the cases of the "good" kind of déjà-ku--parody, homage, and allusion--there is little for the writer to worry about beyond what one should be concerned about in any sort of haiku: Will the reader understand it, get the connection to the other poem, and feel the same feelings that the poet felt in having a given experience? None of these sorts of poems can reach their full potential, of course, unless the reader knows the poem that is being referred to. Consider Alan Pizzarelli's parody:

out of the water . . .
out of her suit

The poet has changed just one word of Nicholas Virgilio's famous poem--from "itself" to "her suit." Yet what a difference it makes. It is not better, of course, but different. The word "Lily" is transformed from a flower to the name of a person, and we see the original poem in a new and amusing light. If we did not know the original poem, the parody would still work on at least a prurient level (one hopes that all parodies would work on their own in some manner or another). But how much better this poem is when one knows the original. So the parodist needs to be conscious of how well known the target poem is by his or her readership, just as with homage and allusion. Readers may or may not "get" the reference, but hopefully they'll still get something else. In fact, if the reader doesn't get anything, even on the surface level, the parody or allusion effectively fails, since ideally it should work in both ways, whether one knows the original poem or not. The main thing that writers of the good kind of déjà-ku should concern themselves with is whether the audience will know the target poem.

Parody should not be confused with allusion, where a smaller part of the target poem is referred to, or perhaps where the form is overlaid in some way to bring about an awareness of the earlier poem. Allusion, or a variation of it in which part of another poem is directly quoted (though not attributed), is generally known as honkadori in Japan. This practice requires that the target poem be sufficiently famous, and one really needs to know the literature and the audience to know if the poem will connect. If the poet has insufficient experience with the literature or the audience, he or she is likely to write parodies or allusions that fail, thus it's no wonder that such poems tend to be written by poets with greater experience or knowledge. They can easily fail or be seen as elitist, though, when the references are too obscure. But when a parody or allusion succeeds, one of the pleasures of reading such poems is noticing the reference or similarity to the original poem, and entering into the poet's game in which both the reader and writer know something together that lies beyond the present poem. It adds an extra layer of intellectual or emotional gratification for the reader. The writer takes a calculated risk, but in the hands of a skilled writer, the risk is not too great.

As for the other kinds of déjà-ku--plagiarism, cryptomnesia, and being insufficiently fresh or original--things get murkier there. No one has any trouble accepting the validity of parody, homage, and allusion, and no one has trouble rejecting the act of deliberate plagiarism. No one wants to accept such willful deception, and haiku editors and poets should not tolerate it. But things become grey with cryptomnesia and insufficient originality.

Cryptomnesia, technically, is still plagiarism. Somebody already wrote the poem, and the offending person has put his or her name to it, without realizing the mistake, and called it his or her own. I've seen this done with Hackett's famous "bitter morning" poem, with poems by John Wills, and several other famous haiku. And the poem doesn't even have to be famous for cryptomnesia to be a problem. Even the editors involved didn't catch the mistake. But I'd venture that it's still plagiarism, and editors, writers, and readers should not consider it acceptable. A famous case of cryptomnesia in the music world involved George Harrison, who was found guilty of plagiarism for using the Chiffon's song "He's So Fine" in his song "My Sweet Lord." With haiku, too, cryptomnesia, though not deliberate or willful, is still plagiarism and should not be tolerated. Nor, though, should we be too harsh. I think it's appropriate for a reader to bring the problem to the attention of the offending party, and perhaps for a correction to be published in the same journal where the offending poem appeared, but then to leave it at that. The poet may also wish to make private or public apologies, but there's little need to make too much of the accident. These things happen, and I've accumulated a number of striking examples. The poets involved in these examples, on both sides of the fence, would likely surprise many haiku poets. It happens to the best of us. It has happened to me.

The remaining problem is perhaps the most difficult to deal with, and that's insufficient originality. It's even hard to define. What is it? What it's not is failed allusion or parody, where no one knows the original poem being referred to. It's not willful or accidental plagiarism. But it may be using words that are too similar to another poem, though written independently and not under the influence of cryptomnesia, about a similar experience. Deciding how similar is "too similar" is a subjective problem, but one can at least accept that the problem exists. There are two causes for this problem. One is that many of us have the same experiences and can't be blamed for writing about them when the experience is touching, profound, striking, or seemingly unique (though obviously it isn't). The other cause is that we don't see freshly, or don't write with fresh-enough words. But how does one define that, or say what is fresh or original? The edges are just too blurry.

Yet the concept is hopefully still valid, and poets would do well to read as much as they can so that they can know what else has been published on specific topics. Indeed, in both cases, knowing the literature helps safeguard against a lack of originality. Fortunately, if one writes honestly about experience, one can be reasonably sure that the poem will not fall victim to the problems of the bad kind of déjà-ku, for the seasoned haiku voice is typically unique or quirky enough to not see or describe things in quite the same way as other poets. Furthermore, good intentions go a long way, though of course there's a limit. It's just hard to define that limit regarding some aspects of déjà-ku.

I want to emphasize that the term "déjà-ku" should not be considered pejorative. I intend it as a general term for a range of effects in haiku, some positive, some negative. The fact that déjà-ku encompasses plagiarism should not taint the good kinds of déjà-ku, such as parody or allusion or two poems uniquely expressing a similar experience. The word "déjà vu" means, simply, "already seen." That's vital to remember with déjà-ku--the poem brings to mind another poem in some way, thus it is wholly or partially seen again. There are also psychological meanings to the term "déjà vu" that may or may not be applicable to the adoption of the term relative to haiku.

Though the word "ku" literally means "verse," I should also add that I find it tedious and cutesy when people to refer to haiku as "ku." It is not merely a "verse," nor do I wish to refer to haiku in such a diminutive fashion. I also dislike the term "ku" used as a suffix in such monikers as "spam-ku" and "sci-fi-ku." The term "déjà-ku" may add to this malaise, but I hope the term is at least memorable, and that its possible cleverness does not detract from its usefulness. It is an efficient way to refer to poems that are "seen again" in some way, and to refer to the range of emotions and issues that arise relative to the good and bad kinds of déjà-ku.

One final thought. In Japan, I'm told that when one unintentionally repeats too much of another person's poem (when it's not intended as an allusion or parody), the offending poet, no matter how senior, apologizes and "withdraws" his or her version. This is apparently done routinely enough that it is given little thought, though tongues may wag a bit until the matter is resolved. Such overlap and repetition is simply an occupational hazard of such a small genre of poetry. In Haiku World, William J. Higginson quotes two remarkably similar poems on the subject of "pheasant." He does not banish either one to any dungeon of shame because it is similar to the other. Instead, his including both is a form of praise for the shared experience.

Likewise, Robert Spiess sometimes published remarkably similar poems side by side in Modern Haiku, citing them as being written independently. And Japanese poems have been remarkably similar as well, as I've seen in translations by R. H. Blyth and Makoto Ueda and others where they have expressly discussed the topic of such similarity. We should be aware of limits, of course, but not paralyzed by them. We should, ultimately, celebrate what we can of shared experience that not only leads us to sometimes write similar poems, but also to find value and resonance in the poems we read by other poets--we enjoy and value them because we have had a similar experience.

Cor van den Heuvel commented on this potential for celebration despite its accompanying difficulties in the introduction to the third edition of The Haiku Anthology, and they are fine words to conclude with: "The writing of variations on certain subjects in haiku, sometimes using the same or similar phrases (or even changing a few words of a previous haiku), is one of the most interesting challenges the genre offers a poet and can result in refreshingly different ways of 'seeing anew' for the reader. This is an aspect of traditional Japanese haiku which is hard for many Westerners, with their ideas of uniqueness and Romantic individualism, to accept." We might do well to remember these words when we next encounter a déjà-ku poem. Can we rise to the challenge?

Note: If you know of examples of déjà-ku you would like to share (please include all relevant publication details), or if you have any other comments on this introductory essay, please contact the author at

Michael Dylan Welch says about himself:

"I've always had a sense of poetry. Being named after Dylan Thomas may have had something to do with that! I was born in 1962 in Watford, England, and grew up there and in Ghana, Australia, and the Canada prairies. In college I majored in communications/media and English, and I received an MA in English in 1989. I focused my graduate studies on twentieth-century poetry and fiction, and wrote a thesis on Anthony Burgess and his sense of play with words--something akin, I think, to the sense of play that pervades haiku. I delight in the fact that "haiku," literally translated, means "playful verse."

"My path to haiku began in a high school English class, where George Goodburn introduced haiku as a seventeen-syllable nature poem. I've long preferred short poetry, so I immediately gravitated towards this form. For years all of my "haiku" were rather ill-formed and ill-informed. About a decade later I bought my first haiku book at a Japanese bookstore near St. Paul's Cathedral in London, a collection of Basho's haiku translated by Lucien Stryk. Shortly thereafter I started buying every haiku book I could find (I now have some 3,000 haiku books and magazines). When I encountered Cor van den Heuvel's The Haiku Anthology, however, my perception of haiku shifted radically, thanks most particularly to the work of Marlene Mountain. No longer did I see "haiku" as whatever words I could squeeze into an arbitrary cookie-cutter shape. Rather, the poems in Cor's collection showed the value of content over the so-called form."

[Biography taken from his website:]

Copyright 2003/2004 Simply Haiku