Why Say More? The Problem of Titling Tanka

First published in the Tanka Society of America Newsletter 3:1, March 2002, pages 12–15. While focused on tanka, the insights I hope this essay offers apply equally well to haiku.


In the Autumn 2001 issue of the Tanka Society of America Newsletter, Gerald St. Maur proposed that tanka poems should have the option of being titled. I would like to respond to this proposal because I think certain issues need to be clarified. On one hand, I agree with him—the author of any poem should feel empowered to do whatever he or she feels is necessary to create a good poem. If putting a title on a tanka achieves this goal, then I’m all for it as an option. No one has ever taken this option away from Western poets, including Western writers of tanka. On the other hand, I’m not so sure I agree with the idea that titles can be appropriate for tanka (or haiku, for that matter). There are many reasons why titles are problematic and undesirable in tanka, and these reasons lie behind why most tanka editors and contest coordinators advise against titles. Avoiding titles is not just an unthinking habit or, as St. Maur seems to dismiss it, merely unexamined “conventional wisdom.” To the extent that it might be an unexamined issue, I would agree that the matter does need attention, whether on an individual basis or as a tanka community as a whole, so I’d like to explore this issue in some detail.

        But first, some background. Earlier in 2001, Gerald St. Maur, in conjunction with the Alberta Poetry Festival Society, spearheaded an international contest called the North American Tanka Contest—with “traditional” and “modern” categories. I was honoured when St. Maur asked me to judge the “modern” category. The “traditional” category was defined as following a 5-7-5-7-7-syllable structure and presenting the “traditional topics of nature, love, loneliness, etc.” (one wonders about that “etc.”). The “modern” category was defined as “five lines on contemporary topics expressing the spirit of tanka” (one wonders about that “spirit of tanka”). I received the submissions for both categories, and quickly observed that many poems in the “traditional” (syllabic) category were about modern/contemporary topics, including some of the poems eventually chosen as winners. Similarly, many of the “modern” (presumably free-form) poems dealt with the “traditional” topics of love and loneliness—again including some of the poems picked as winners. I mention these categories because they seemed essentially arbitrary and ultimately unhelpful in distinguishing the poems and how they were submitted. Better, I think, to have had just one contest and let the poets decide for themselves if they wanted to write syllabically or not, and what topics to write about. In his article on titles for tanka, St. Maur proposes that the tanka poet should not be straightjacketed into avoiding titles, but should be allowed to use them if he or she so wishes. One would think, then, that these same poets should be empowered to write syllabically or not and to write on so-called “traditional” topics or not in their tanka. But apparently the freedom St. Maur wishes for titles is somehow different from the freedom to write syllabically or not.

        More troubling in the contest’s rules—and it did trouble me, for tanka’s sake—was the submission requirement that “Each poem must have a title or a brief headnote.” This rule was disturbing enough to at least some poets, so I’ve learned, that they chose not to enter simply because they objected to this requirement. At the very least, the rule struck me as puzzling. Consequently, I wrote to St. Maur, whom I had the pleasure of meeting in September of 2000 when I attended the Japan Tanka Poets’ Club Third International Tanka Convention in Vancouver, British Columbia. In St. Maur’s essay in support of the option for titles in tanka, some of his words directly arose out of my email to him objecting to titles in tanka. Titles in such short poems, I said, were either cheating (adding an extra line) or redundant (repeating something already in the poem), arguments that I do not feel he countered sufficiently in his essay. Of course, these objections to titles could apply to any poem of much longer length. However, with a poem as short as a tanka or haiku, the title is overpowering.

        The insistence on titles or headnotes, however, could have been simply an administrative request to help manage the many poems received for such a contest. For many years, though, the Haiku Society of America and other haiku organizations have managed much larger contests with titleless poems with no visible problems. The poems, submitted in duplicate or triplicate, are simply numbered to maintain the anonymity conventional in such contests, and when discussing the poems, a judge and the contest coordinator typically refers to each poem by number, by first line, and/or by quoting the entire poem. It has long been customary in academic circles, too, to simply refer to titleless poems of any length simply by their first lines.

        At any rate, it soon became apparent from my discussion with St. Maur that requiring titles or headnotes was not just an administrative request. Any contest organizer can set rules however he or she likes, so I won’t quibble with his choice to require titles or headnotes for this contest. I cannot help but wonder, though, how many more people would have entered or how the poems might have differed in quality if titles or headnotes had not been obligatory.

        What I can quibble with, though, is the aesthetic effect of promoting titles in tanka (the same quibbles would also apply to haiku). Upon receiving the poems for the contest, my reading of all the poems (both “traditional” and “modern”) immediately and uniformly confirmed for me that the titles were superfluous. A few poets clearly chaffed at adding titles to the poems, expressly identifying what was required of them as a “headnote” (and all of which added nothing worthwhile to the poems). Since I’ve never thought tanka benefited from titles, and feeling sympathy for the poets who were required to tack on titles or headnotes, I attempted to review all of the poems as if they had no titles or headnotes. Indeed, I thought the tanka were practically all significantly weakened by their titles—again because they were either redundant (repeating something in the poem) or cheating (adding an extra line).

        I’d like to expand on this problem of adding an extra line (where the title does not merely repeat something in the poem). When I say it is “cheating,” I mean it in the sense that one is turning a supposedly five-line poem into a six-line poem. In much Western poetry, a title serves not just to identify the poem (as a label) but often to summarize it, analyze it, or direct the reader’s attention. In all of these cases, the title, if it is not merely adding an additional line to the poem, allows the writer’s ego (intellectual processes) to intrude. The title then becomes an intellectualization, something that may immediately be at odds with the tanka aesthetic. It is clearly at odds with our prevailing haiku aesthetic, which aims, among other things, for egoless objectivity. I believe contemporary tanka shares a similar aesthetic in English and Japanese, but even those poems that assert a subjective intellectualization in the poem itself (if at all) do so without needing or relying on a title.

        For me, a chief problem with St. Maur’s argument in favour of the option of titles in tanka is that he seems to overlook centuries of Japanese precedent. In all of 1,300 years of Japanese tanka and waka, one seldom sees titles. The extent of this much precedent may not be sufficient reason on its own to avoid titles in English-language tanka, simply because the languages are different. However, the reasons why there has been 1,300 years of precedent in Japan is sufficient—and the reasons are essentially applicable to English-language haiku. Tanka in Japan sometimes have headnotes as opposed to titles (where they have either at all). As St. Maur says in his article, headnotes explain the “circumstances of the poem”—the context or setting. Absolutely these contexts can be helpful, especially across the cultural and language gap between Japanese and English (some headnotes, though, have been added by scribes or translators, and are not always part of the original poem). Again, though, headnotes are not at all like titles, and one need not be a particularly astute reader of tanka to spot the difference; headnotes are typically factual, locational, or anecdotal, whereas titles are often symbolic or intellectual. The headnotes, if used at all, do not fight with the tanka aesthetic, whereas titles typically do.

        Paintings benefit from titles (or numbers) simply because one has no other way to efficiently refer to the painting in words. Some artists have objected to any sort of titles for their work, preferring that only the image itself should be experienced, and stating that the “representativeness” of a title diminishes the artwork. The name of a thing, of course, is never the same as the thing itself, just as a map is not the country. It is for a similar reason, I think, that haiku poets avoid metaphor and simile, because these devices indicate the ego (intellectual processes) behind the poem, add an additional layer of representationality to the poem, and are detours around or away from the thing itself. Likewise, movies (mostly visual experiences) and songs (aural experiences) need titles as a means to refer to them in words—and perhaps also as shorthand (it is easier to say “Citizen Kane” than to have someone watch the entire movie each time you want to refer to it). Similarly, for long poems, the title usually functions as an identifier as well as shorthand. For poems as brief as haiku and tanka, however, a line is crossed. No shorthand is needed because the poem itself is short enough. In terms of maps representing countries, the map is no longer needed because the detail of the map is so great as to make the map redundant. The poem is already in words, so no title is needed to convey or represent in words any nonverbal artistic expression (in fact, some poets such as E. E. Cummings eschewed titles for nearly all of their poems because of their redundancy and for their representational limitation). I do not wish to make an argument against titles in Western poetry, however, just that the Western notion of classification and representation as is manifested poetically in the use of titles seems at odds with the Eastern poetic aesthetic evident in tanka and haiku.

        To turn more thoroughly to St. Maur’s essay, I wonder if, as he puts it, “the majority of Western tanka poets have never really thought about [titles].” That could be so, and to the extent that they have merely followed the injunctions of tanka editors or contest coordinators, they owe it to themselves to give the matter some consideration. However, I think many tanka poets have thought about the issue. Most Western tanka poets come to tanka through haiku, and for both haiku and tanka the Western poet must unlearn certain Western poetic habits—not because there’s anything wrong with them, but because the Eastern poetic aesthetic demonstrated in haiku and tanka is simply different. In teaching haiku workshops to children and adults for many years, I have found that the metaphorical and analytical tendency is enormously apparent in initial and repeated attempts at haiku, a tendency that is sometimes insurmountable. Even in children, the Western dialectic that flowered with the ancient Greek poets and scholars and their dialogues is still with us. All of civilization, especially medicine and the sciences, owes much to this Western perspective, where the self is promoted as independent of nature, and where nature is to be subjugated and classified. The Eastern perspective is generally more holistic, where the self is part of the larger whole, dependent on nature rather than independent of it—part of nature rather than apart from it. Titling a poem is, in many cases, simply an extension of the Western dialectic—the need to identify, to classify, to analyze. In the Eastern tradition, this is considered unnecessary, like adding legs to a snake. In Raking Sand (Press Here, 1993), Virginia Brady Young comments on this Western predisposition by discussing Robert Frost’s poem, “Dust of Snow.” The first half of the poem presents a striking objective image, but this image, from a haiku/tanka perspective, is marred in the poem’s second half by an explanation of what the poet thinks about the scene. A title is equivalent to this sort of commentary, and, for the tanka or haiku aesthetic, is it unnecessary.

        Indeed, it is not simply “conventional wisdom” (seemingly unexamined) that tanka do not have titles, as St. Maur says. Rather, titles being superfluous in tanka (in both English and Japanese) is an extension of a deep-rooted aesthetic. St. Maur says in his essay that “Sometimes the title is said to simply repeat words in the poem” (expressing my objection to him that a title is often redundant in a tanka). He counters by saying “how better to emphasize a key ambiguity or turning point?” But to me this is the worst thing a poet could do in a tanka (or haiku). If, as St. Maur says, one has to “point the mind of the reader” with a title, then I believe the poem itself has failed. Rather, the tanka, like haiku, should aim at suggestion and subtlety. A bad title simply hits the reader over the head, and I’ve yet to see a good title for a tanka. Titling a tanka is directly at odds with the holistic Eastern aesthetic manifested in tanka and haiku. Poems in these genres can certainly be attempted in modes other than strictly the prevailing Eastern mode. However, one can quickly run afoul of the many good characteristics that are strongly effective in the Eastern aesthetic in successful tanka and haiku (witness the pseudo-haiku of Spam haiku or road-rage haiku—and thank goodness that tanka seems not to be plagued by a corollary).

        In his essay, St. Maur also repeats my statement to him that adding a title is “cheating, by using more space than is allowed in the poem.” He counters by saying “it is hardly fraud or deceit” because “the title is openly declared right in front of the reader.” However, this misses the point. The sense of “cheating” I mean is that it lengthens the poem, making it something other than what the tanka has so effectively been in Japan for centuries. My reference to “cheating” is not in the sense of fraud or deceit. Moreover, a title in a tanka does not just cheat the reader; it cheats the poem. Titling a tanka is like changing the dimensions of a tennis court. A tennis player could openly get out his or her paint and redo the lines to make the opponent’s side larger, and thus easier in which to land serves and volleys. No one would be “deceived” by this paint job. The effrontery of it would be too blatant for anyone to feel deceived. But it’s still a sort of cheating. And it would no longer be tennis.

        Here, too, St. Maur seems to confuse headnotes with titles, saying that “providing contextual space around the poem does not alter the poetic space.” However, contextual space is what headnotes can provide, and I have no objection to them in English tanka if kept to a rare minimum. A title does something different, though, moving beyond context to assert the Western analytical ego in ways that contradict the Eastern aesthetic typically presented in the rest of the tanka, and thus it detracts from the rest of the poem. Because a title is commentary or symbolic, adding a title to so short a poem as a tanka or haiku strongly risks destroying the very moment of perception, awareness, and understanding that the poem is trying to convey or imply.

        As St. Maur says, he does not claim “that tanka must have titles; only that they may have them.” His fundamental argument for this seems to be that “To discourage titling is to lay a bias against it and to restrict the artistic freedom of the poet.” I’m all for encouraging and maintaining the poet’s artistic freedom, but I do not believe that one can include a title with a tanka and have it still be an effective tanka unless (if anything) one ignores the title. And if one has to ignore the title, then why have it?

        If nothing else, St. Maur is wise to question the wholesale practice some beginners and even more experienced poets may adopt in unthinkingly eschewing titles in their tanka. It is good to think about the reasons to do something, or not do it, in one’s poetry, and thus to shape one’s aesthetics. However, Gerald St. Maur goes beyond merely questioning what he calls “conventional wisdom” to assert that tanka may have titles in English. Yet he presents not one tanka that is better off for having a title (as opposed to a headnote). Only empirical evidence would possibly bolster his argument. In contrast, from seeing so many unnecessarily titled poems in the North American Tanka Contest (as well as occasionally in other contexts), I can say that for me their titles all failed. Since so many poems could be presented as failures with a title, and since it is incumbent on poems with titles to prove that they do work (if they do), there is nothing to be gained by my showing empirical evidence of tanka that fail because of being titled (nor would I want to embarrass their authors).

        Actions, they say, speak louder than words. The poem, the tanka, is action. The title, unfortunately, is either an extra line of poetry or merely words about the poem, and typically doesn’t accomplish anything of benefit to the poem, at least one as short as a tanka or haiku. Only for longer poems might the title serve a useful role as shorthand to represent the whole poem, something that isn’t necessary in a short poem—and in many cases might not even be necessary in a longer poem where the first line can easily be used to identify or represent the poem. At any rate, I’m pleased to report that none of the tanka have titles in Countless Leaves (Inkling Press, 2001), the anthology resulting from the North American Tanka Contest that St. Maur edited. Each poem, and the whole book, in my opinion, is better off for it. And, in fact, the entire book is a remarkably strong publication from its graphic design and production values to its choices of poems and introductory material.

        To conclude, I recall a relevant remark made by Makoto Ueda in Modern Japanese Tanka (Stanford University Press, 1996). In commenting about the tanka of one Japanese poet, Nobuyuki Okuma, who chose to title many of his tanka, Ueda notes that the poet’s tanka have titles “like English poems.” He emphasizes that titles are rare and unusual in tanka by explaining that, “Following the waka tradition, tanka by most other poets bear no titles, although they are sometimes preceded by headnotes that explain the circumstances of composition” (page 135). Certainly, one can tack titles on tanka if one wishes, but such “artistic freedom” has a cost. As Ueda says of Okuma’s work, the poet “began to turn his attention more toward formal innovation in tanka, exploring the possibilities of free-style tanka” (page 133). Ueda concludes that, influenced by European poetry, “his tanka grew longer in length and more irregular in prosody, until they became indistinguishable from ordinary free verse.” Ueda reports, sadly to me, that “At this point he lost many of his readers, until he stopped writing poetry” (pages 133–134). What more need be said?