What Is A Tanka?
Richard MacDonald

Before going into the Tanka form itself, there are two other forms that should be understood or at least discussed. These two forms, and the others already discussed, again point to the significance or alternating line lengths of odd numbered onji, specifically, 5-7-7 or 5-7-5 to the Japanese aesthetic. The first form is called a Sedoka; the second is called a Choka.

A Sedoka is similar to a Mondo in that it also consists of two parts, or one pair of Katauta. The difference is that Sedoka were written by a single author, and did not generally consist of a question and answer part. Incidentally, while the norm for writing Katauta was 5-7-7 onji, it was also acceptable and common to see them in 5-7-5 format. Remember, the last line was used to fill out the poem, and while this was generally done with 7 onji, 5 onji (also an odd numbered line) was considered a pleasant variant from the norm. This (Katauta) poetic form was respected for it's variance as well as its conformity. So, it can be said that the Mondo and Sedoka were written by combining two Katauta -- each Katauta consisting of three parts (lines) with two separate rhythms, a consistent overall form length of 17-19 onji, and each part being of an odd numbered line in terms of onji. The major difference between the two being that Mondo were written by two authors and Sedoka were written by a single author.

The next form is the Choka, the Japanese long poem. This poem was structured 5-7-5-7-5-7-5-7-5........7-7 onji in line length, and could be of any overall line total; many of which exceeded 100 lines. As stated earlier the Katauta could be called the basic unit of Japanese poetry. Here in the Choka, the 5-7-5 or 5-7- 7 (17-19) onji pattern is easily found. It seems the preference for ending Japanese poetry in these early days was with the 5-7-7 onji pattern; however, as stated earlier, 5-7-5 was often used as a substitute and was equally acceptable in terms of aesthetics and appreciation. In fact, as time passed, the 5-7-5 onji ending became more prevalent rather than less in forms other than the Tanka, though never quite exceeding the originally preferred ending format in these older poetic form. With all of this in mind, it is easy to see why the Katauta is believed by some poetic historians to be the original basic unit of Japanese poetry, either as a 17 onji unit or a 19 onji unit. This is important to remember not only for understanding and writing Tanka poems but in understanding and writing in any Japanese poetic form, including Haiku.

Below is a Choka taken from Kenneth Yasuda's book, "The Japanese Haiku".

O palace maiden
the daughter of my subject,
Do you bring a wine holder?
If you hold it up,
Oh, hold it in your hands;
oh, hold it firmly,
ever firmly in your hands;
O you wind holding maiden.

In the above, this Choka appears to be written in a 5-7-7-5-7-5-7-7 onji format, varying from the norm in only line three. Actually, what we have here is a Choka composed by combining a Katauta in the form of a question with a Tanka poem in amplifying expression. This is considered to be a Choka, but within this Choka, one can find evidence that the early Japanese poets were not averse to combining poetic forms; so long as the basic unit of structure remained within the 17-19 onji range. So the answer to the question of whether or not Tanka arose from some other poetic form seems to be -- quite possibly so; however, there are no records that concretely say that this is the case.

Now, with a historical background established, we can look at the Tanka itself, and, hopefully, be in a position to appreciate it as a separate poetic art form and as an evolutionary piece of the much larger Japanese literary aesthetic. Of all the poetic forms ever written by the Japanese, Tanka is clearly the most rigidly adhered to form in terms of structure. It is constructed by 5 lines or units, each odd in number of onji, and ending in the traditional 7-7 onji pattern. Further, on the whole, Tanka consist of two separate divisions in terms of rhythm structures, each of about one breath length to recite. The earlier pattern for this rhythm change was:

rhythm unit 1: 5-7 onji -- rhythm unit two: 5-7-7 onji.

Later, the dominant rhythm pattern changed to:

rhythm unit 1: 5-7-5 onji -- rhythm unit two: 7-7 onji.

Of course, there were other variant tried and successfully used, such as: rhythm unit 1: 5 -- rhythm unit 2: 7-50 -- rhythm unit 3: 7-7 onji


rhythm unit 1: 5-7 -- rhythm unit two 5-7 -- rhythm unit three: 7 onji, and

so on with other attempts.

In the earliest Tanka poems, the last line was used primarily as a repetition or summary line to rhythmically fill out the poem keeping the preferred odd number of lines, odd onji count structure, and rhythmic divisions associated with the aesthetics of the Japanese poets of that period:

Many clouds unfurled
rise at cloud-decked Izumo;
Round you spouse to hold
raise many folded barriers
like those barriers manifold.

While this poem seems to consist rhythmically of two parts divided 5-7; 5-7-7, it actually has three units, 5-7; 5-7; 7, with the last unit being used as a repetition and summary line common in the earliest of recorded Tanka. In the poem above there are two distinct main rhythmic parts separated by a major stop at the end of the 12th onji. From there the rhythm starts out again and continues to the end of the poem; however, the repetition of the last line causes it to stand out in isolation from the remainder of the second part; thereby, giving the second part a technical second internal rhythm found by the ear. Remember that Japanese poetry is syllabic by nature and not metrical or rhymed. This is because, like the French language, the Japanese language lacks stress accents where the uniform stress on the last syllable of each word makes it impossible for poets to observe the kinds of metrical patterns favored by Western poets since the ancient Greeks. Rhyme is not used in Japanese poetry, not because it is too difficult, but, on the contrary, because it is too easy. Japanese words, for the most part, all end in one of 5 open vowels; therefore, without trying, a poet has a 20% likelihood of achieving rhyme. The other European metrical scheme based on quantity is also not possible since in *classical* Japanese vowels all end with equal weight. This may be a possibility in modern Japanese which has both long and short vowels, but has rarely been attempted. This left the Japanese with syllabics as the only true choice and has remained so ever since.

So, how does one write Tanka in English? This depends on how closely one wants to stay to the Japanese model. The language differences between Japanese and English, touched on slightly above, are vast and complex. For this reason, most writers feel that converting onji to syllables is not a one for one process. English syllable are far too long and carry too much information to equate to the Japanese onji. With rare exception, a Tanka written in English would be difficult to recite properly (in proportion to Japanese speaking lengths) in two breath lengths. Haiku written in 17 syllable would be as difficult to recite using the same comparatives as 17 onji.

This should not deter Western poets who would like to write in a Tanka style from using English syllables in a one for one format, or from using rhyme, meter, or accent weighting. Again, this is a matter of personal taste and is reflective only as to whether one wishes to stay close to the Japanese model, or stray from it for personal reasons or aesthetics to incorporate the heritage of the West into their poetic works. If you are concerned with your English based Tanka being translated to the Japanese language in a format that readers in that language are accustomed to, then one may wish to stay closer to the Japanese model and modify the structure somewhat to make it work. If this is not you choice, then in order to keep the poem resembling Tanka, one must as a consequence of using freeer choice in language become more restrictive in tone and theme to have ones poems recognizable as Tanka in the Japanese tradition. Of course, other possibilities exist, but making a claim that the outcome is in fact a Tanka poem becomes much more difficult to defend.

Tanka, at least in the classical sense, was used to touch on all sorts of subjects; especially after the decline of the Choka. However, the tone of Tanka poems has always reflected the tone of the Japanese Imperial Court and its courtesans. This tradition is carried forward today in Japan for the most part, although there are exceptions. So, if one were to emulate the Japanese aesthetic, one would write Tanka in an elevated tone in English avoiding harsh epithets and vulgarities and themes of a similar nature. Some may find this too restrictive and wander from the established tone of Tanka; this is purely a matter of choice and personal taste, but either way, one can work in relative freedom and originality and achieve remarkable results in just 5 short lines of poetry. Tanka is a wonderful medium of expression and because of its short form, it is a poetic form in which poets can produce highly memorable and memorizable works.

Tanka is also, but not always, used in a manner that includes nature in the expression of thought or feeling, similar to haiku, but because of its extra length, Tanka allows for deeper thought and expression of themes that would be too burdensome for haiku to carry. Again, the decision to use nature as a backdrop for expression is a personal choice. Those who think staying close to the Japanese model will do so with a great sense of commitment; while others who choose to stray will argue their view points with equal vigor. Whatever choice you make, it should be a choice that you are happy with and that provides you the means of expression that you seek. Whatever choice others make should be viewed as simply an alternative choice and not an attack on your personal perspectives or tastes. Like Haiku, Tanka is in its infancy in the West and its development will be determined by those who utilize its heritage and manner of presentation to greatest impact on readers both in the East and West. By trying to appreciate the aesthetic of Japanese poetry and thinking and incorporating it into the West, we bring the two cultures closer in terms of understanding and fellowship. There is nothing to lose in this process. All can only gain by the experience. So, when embarking on this journey, remember the goal is fellowship and learning how to best live our lives by incorporating poetic experiences into them.

In conclusion, I sincerely hope that this introduction to Tanka has achieved its goal of enabling writers, previously unexposed to Tanka to appreciate it as a poetic form and as a remarkable expression of the Japanese aesthetic. The only true way to begin seriously writing Tanka poems is to first understand what it was, how it developed and changed, and what it is now by reading as many Tanka poems as you can, both classical Tanka and modern Tanka. Hopefully this piece will encourage others to try their hand at Tanka and to share their feelings and insights into the living world of which we are incorporated for better understanding and achievable peace on the planet by expanding fellowship throughout.

Copyright © by Richard MacDonald, 1995

You may email Richard at

More on renga .