Tag Archives: Yosano Buson

A Famous Buson Haiku: Is It ‘Kite’ or ‘Kites’?


One of the peculiarities of the Japanese language is that while it’s does have a plural form for nouns it almost never gets used, mainly because the context that speaker is in dictates if they mean more than one of something. This gets a little tricky when it comes to reading haiku because the reader isn’t in the same physically present context as the speaker. And this haiku by Yosano Buson is a good example of it:

Ikanobori/ kinou no sora no aridokoro

Kites Yesterday’s sky’s having place.

Kites have been in Japan since the 600s when the were imported from China and in the past they were used in many ways as symbols of good luck and talismans for good fortune. Nowadays, we tend to think of kites as something children enjoy, but in the Edo era adults enjoyed them as well. Besides flying them as omens, kite fighting was something that people pursed with enthusiasm as well. Kites became so popular that the Edo government finally banned them except during the o-shougatsu holiday (the few days before and after New Years Day).

The banning of kites caused a problem for haiku. Kites as a “kigo” (season word) were listed as something done in spring, but with what is the now metropolis of Tokyo having forced the flying of them into the New Years’ holiday, the “kigo” had to accommodate that as well. Any seasonal reference books for haiku (‘Saijiki”) are in five volumes, four for the seasons and the extra one for the New Years period which is the most important religious holiday of the Japanese calendar. The old “Saijiki” I have has a double listing for kites, one in spring and the other for the New Year. However, the practice now is to only have kites listed as something done in spring.

Understanding the history of kites in Japan is important in the context of reading plurals into this haiku because, as anyone who lives long enough in Japan learns, there is a right date and a wrong date for everything. My wedding, for example, ending up being on a Tuesday instead of the more preferable Saturday because of it being a more auspicious day, or as it got explained to me “Saturday is no good because it is a bad Buddha day.”

Having good days and bad days makes religious things a bit more communal because people will always do things on good days. So, if someone wants to fly a kite as an omen of something then they would necessarily do it on a good day and not a bad one, meaning that there will be more than one person out flying kites on the good days. Since kites could only be flown during a short period of the year in Edo, probably not longer than a week, then there will be hordes of people out flying them before the ban kicks in again,

The point I am trying to make is that it is hard to read this haiku and not think of there being more than one kite up in the sky and that Buson was actually implying the plural here. The Nihon Daisaijiki, a large full color “saijiki” published about ten years ago, discusses this haiku and states this about the number of kites:

This doesn’t have to necessarily refer to the whole of the sky. It indicates the scene of a part of the sky that has one, or more likely, several kites in it.


Above is the picture that is included in this book under the heading of kites. Because is a wood block print, you can’t say that it is of the whole sky, but it is a part of the sky that has several kites floating in it.

Even when a commentator wants to you read this as being about only one kite, they have to couch it in terms of other kites in the sky. A book titled “Haiku Daikan” published by Meiji Shoin explains it like this:

There aren’t a large number of kites flying in the sky. In the very least, there are no other kites nearby this one, to follow this instruction to its obvious conclusion, there is a solitary one afloat.

Asking the reader to follow this instruction is a dead give away that the writer automatically assumes that the reader is reading the plural for kites.

I don’t find the argument for a single, or actually separated, kite very compelling because it makes the image a very personal one, which is exactly how at the end this commentator reads the symbolism of the kites:

It seems that the loneliness of human existence is suggested with this kite, sadness and at the same time a kind of fond remembrance is to be felt in this expression.

Whenever we feel fondness and sadness at the same time it means we are becoming nostalgic about something. It’s easy to understand how Buson would get nostalgic about his childhood from a kite, but I’d argue that the experience of flying a kite rather than watching one is what would more likely trigger the flood of emotion that would bring the bittersweet memories argued for above. I think it is next to impossible to read that Buson is flying this kite because I don’t think he would be able to make the topographical description of them being in the same spot as they were yesterday if he was.

The commentator in the Daisaijiki makes this conclusion about the haiku

Everything is calm and peaceful, indeed, there is a deepness to the feeling of the mild spring. Getting that as a scene, the impression in a person’s mind too will surely be similar.

This reading has the scene influencing the feelings in the writer, it’s a lovely spring day and people are enjoying it by going out with their kites and the scene has Buson feeing relaxed and secure about life. The immediate problem with this reading is that there aren’t any emotional markers in the haiku to expresses the “feeling” that the commentator mentions. In haiku, particles of speech are how the writer attaches personal emotions onto images, but in this haiku Buson didn’t use any. If he was feeling the mild spring so deeply you’d expect that there would be a particle used somewhere to express it.

Japanese is my second language, and, so sure, there might be a nuance in this haiku that I am unable to catch because I’m not a native speaker, but for me the lack of a particle shows that Buson is being intellectual rather than emotional here. Saying that the kites are in the same spot as yesterday is not a visceral reaction to the scene, rather it is a measured logical one.

Therefore, the symbolism for me becomes a little more generic and less personal. Instead of being a privatized experience, I tend to see this as a universal one. Everyone has flown a kite or watched kites in flight sometime during their life. Besides the universality, kite flying is also a timeless enterprise. How long have people been flying them? Although the popularity of kites wane, somebody somewhere still enjoys flying them. The kite designs are timeless as well, kites patterned after those made in the Edo era still fly as true today as they did back then, and I’m sure that the even older designs will still be able take to the air.

So, with ‘timelessness’ as the optimum idea in my mind:

    in the spot
        of sky 
           they were  
               in yesterday.

The problem with this literal translation is that the Japanese language doesn’t need adjectives to fill scenes for the reader because connotations between words implies scenes that don’t need to be openly stated. Unfortunately, in English, we need adjectives to paint the scene, and since we use sound as a way to imply emotions, adjectives are also the means of achieving syntax as well.

Kites aloft....
      in the same exact 
              spot of 
                clear sky 
            they wafted in yesterday.

All I have done is put in some adjectives around the images and use a verb that adds a descriptive element rather than a nondescript one. I don’t think I have impinged in on what is in the original is by doing this.


The Spaces of Robert Hass




One of the peculiar things about haiku written in English is that it the people who creatively write it have often followed those who have translated it as the models for what they write. It is natural for translators to have somewhat of an unchallenged status about their knowledge of another language, but what if the translators simply misunderstand what they are translating and generally give translations that completely change what the haiku were in the originals?


Robert Hass is a distinguished poet and critic who has had a successful career that led him to even being the poet laureate of the United States for awhile, but it when comes to his book, “Translations from The Essential Haiku: versions of Bashö, Buson & Issa”  (The Ecco Press, 1994) it can unequivocally be stated that what he has published as the essential part of the first three great masters of Japanese haiku is often flawed by the single worse thing any translator can do, i.e. make the translations read as the complete opposite of what they are in the original language.


Tracy Koretsky has written an article titled “Haiku and its Relationship to Space” and she uses examples from Hass’ book to show how the Japanese masters used space in their haiku and what follows is such a comedy of errors in talking about the Japanese haiku used in the article that one can only shake their head and smile at them. And yes, you can blame Hass for this.


Koretsky starts off interestingly enough when she talks about the tonal quality of the Japanese aesthetic of sabi:

That is exactly the effect the unresolved tanka that is haiku has upon the Japanophone’s ear. This trailing off, this ellipses leading to nothing, effectively imbues haiku with its predominate tonal mode: the untranslatable quality known as sabi. Inadequately understood, sabi is the sadness of aloneness, or perhaps better phrased as the more Zen concept of the solitariness of no-mind.

Notice how space functions to convey this quality in this haiku by Bashö:

         first snow
         on the half finished bridge


The haiku she quotes does add to our understanding of what she has written, but since this is a Hass translation one has to assume that what the haiku aims for in Japanese might be a bit different than what he presents it as being in English.


In the early 1990s a Japanese man named Toshiharu Oseko self-published a two volume called “Basho’s Haiku” where he translates about two thirds of Basho’s haiku into English giving insight and commentary on each haiku. This haiku is in Volume Two on page 548 and this is how Oseko translated it:


Hatsu-yuki ya
hashi no ue ni.

The season’s first snow,
Is falling on the new bridge
Almost completed!


Immediately we are struck by the difference between the “half finished bridge” that Hass gave us and the “almost completed” one that Oseko does. The difference is so striking that there isn’t even a semantical argument about the two ideas.


The great thing about Oseko’s book is that he lays out and explains what the verb is in each haiku and how the conjugation of them works out in English. The way verbs are conjugated in classical Japanese are difficult enough for the average Japanese person, so this is an invaluable guide to how verbs work and the effects they give in the originals. He explains the verb conjugation here as:

Kake = kakeru: to build (a bridge)
kakari= kakaru: to start  ~ing
tari= taru: almost finished


Which is easy to see that this literally translates as “building bridge starting to finish” which although might seem like a tortured phrase to us in English is something that is natural to the Japanese. Sure, the language of Hass’ translation is more poetic, but there isn’t any doubt that Oseko correctly understands the grammar of the original and is able to accurately render the sentiments Basho expressed in this haiku.


Oseko also notes how the haiku has a preface that says “When Fukagawa O-hashi is almost finished” and also more information about what bridge it is:

The official name of the bridge was Shin Ohashi (New Great Bridge). The construction started in July and finished in five months on Dec. 7 1693. It was about 200 m. long. People living on the east side of the Sumida River were anxious for the completion, because they expected much easier access to the city center.


The woodblock print that is the picture on this article is of the Shin Ohashi and it is an imposing structure due to it having the height to let ships pass underneath it. Basho lived on the east bank of the Sumida River and no doubt having this bridge built affected his life and this haiku is attaching the thrill and excitement of the first snowfall with the feelings that accompany seeing a much welcomed bridge nearing completion. The woodblock print makes it easy to imagine what a sight it would be (and if you click on it you’ll get a fuller sized picture.) Which of course makes Koretsky’s conclusions on this haiku completely superfluous:

The bridge, the only mode of connection between people of the day, is not only unfinished, but, because of the snow, will remain so for the long winter ahead. The image confirms an unbroken emptiness of space and time lying ahead.


The only way she can get the reading of this haiku so backwards is because Hass has made the translation into something that the original never was. It also needs to be mentioned that although Tokyo does get some snowfall during the winter, the snow never accumulates and within a few days what has fallen melts away.


Yosano Buson is famous for writing some panoramic haiku that are able to convey an astonishing wide sense of view and space, so it is a bit surprising to have Koretsky arguing the opposite when talking about him:

One way, to borrow a term often applied to the paintings of Edgar Degas, is to “break the frame”. He was the first painter, at least in the Western tradition, to “crop” in ways that implied extension or continuation beyond the composition. For example, by depicting only the thigh and forearm of one of his famous dancers, Degas left the viewer to complete the partial limbs mentally. This is, in a sense what the second of the great classical masters, Yosa Buson (1716-1783) is doing in this haiku:

field of bright mustard,
       the moon in the east -
the sun in the west.


The problem with Hass’ translation is that he added the image of a “field” to it which isn’t in the original:


Na no hana ya Tsuki ha higashi ni hi wa nishi ni


The opening phrase of the haiku simply states “the mustard flowers” so the image isn’t delineated into the space of a field as Hass gives us, and although the translation faithfully follows the original after this, it is easy enough to see how inserting a field into the reader’s view crops the imagery. As Koretsky writes:

Notice that in Buson’s haiku above there is just pure description, and sparse description at that. Not only must the continuation of the mustard field be supplied by the reader, so too must the emotional content, the sense of awe, or perhaps humility, the poem instils. By the way, take care not to read meaning into the moon coming before the sun. The haiku has been translated both ways.


But the problem is that the “pure” and “sparse” description occurs because space has been defined by the translator’s decision to crop the mustard flowers into a defined place and not by the person who wrote the haiku. The idea that the emotional content of the haiku is supplied by the reader and not the writer is also another thing that comes about by another translation decision to completely ignore the spoken particle of “ya” altogether. In fact, the particle of “ya” is the thing which makes this whole haiku come together for the Japanese because it indicates where the writer is placing their emotional content in the haiku.


Toshiharu Oseko in the introduction in Volume One defines how “ya” works a particle:

It is an interjection for a strong impression, emotion, excitement and exclamation. This is used most often as a “kire-ji” (cutting word) to cut a haiku into two sections giving a pause for intensifying the impression and emotion giving more depth and the expansion of imagination.


An good example of how this is used is in Haruo Shirane’s “Claasical Japanese, A Grammar.” (Columbia University Press, 2005). It states, using a quote from the Genji Monogatari (Yugao, NKBT 14:139):

If “ya” has an exclamatory function, it is an interjectory particle. An example of “ya” as an interjectory particle is

あはれ,いと寒しや     Aware ito samushi YA.

Well (aware), it’s very (ito) cold (samushi)!


Buson is using “ya” here to intensify his emotional response to the mustard flowers in the same manner it used above to state the speaker’s reaction to the cold. Because of it, the reader picks up the cue and reacts to it the same way we would if someone had said “Ah, the mustard flowers!” Since the writers attention has been captured by it, the reader then fills in the space between the eastern horizon the moon is rising in and the western horizon the sun is falling on by realizing “oh, the mustard flowers!” and thus filling out the panoramic view that Buson expresses by placing an interjection on the flowers.


To put it a bit more succinctly,  the interjection attached to the flowers piques the reader into wondering why Buson has had a reaction to them, and then, because of it, come to understand that his emotional response was to the vast swarth of yellow. The use of “ya,” the personal emotion of the writer, is where the poetry of the haiku blooms from.


Cheryl A. Crowley in her book, “Haikai Poet Yosa Buson and the Basho” (Brill, 207), on page 278 says that this haiku was the opening hokku of a linked verse sequence and notes that “scholars have pointed out the similarity between this verse and the second of Tao Yuanming’s “Untitled Poems:”

The white sun sinks into the western slopes,
the pale moon rises over the eastern peaks.
For ten thousand leagues the light shines,
Over a great distance the sky is bright.


The allusion is quite apparent and this isn’t the only haiku Buson wrote which is indebted to Chinese poetry. Crowley also notes that there are two separate headnotes for this haiku in different sources that have “Spring scene” and “Scenery outside the capital” which “later critics have tended to ignore.” (pg 279). Anyone who has spent some time in Japan can attest that there are many amazing places where the scope and amount of the same wild flowers in bloom can be a sight to marvel at, but, considering how clear the allusion is in this haiku one has to wonder if Buson had actually witnessed this or not.


Robert Hass is an accomplished poet and his ability with language is shown in this translation:

A poem from the third of the triumvirate of classical masters, Kobayashi Issa (1763-1827) demonstrates the point:

        the snow is melting
and the village is flooded
        with children



He made a great connection that catches the idea of the melting snow, tying the water that comes out of  it together with the image of children out to play in it. However, the word “flooding””isn’t in the original at all. Rather, all Issa wrote was “full of” (ippai):


Yuki togete mura ippai no kodomo kana

And, as in the other haiku above, Hass simply ignored the exclamation that Issa included, which here is the particle of “kana.”


Toshiharu Oseko’s introduction in volume one of book on Basho explains “kana” as a “conclusive particle” that is “also used very often as a “kire-ji” like “keri” which is an auxiliary verb, cutting haiku not in the middle, but at the end, giving a strong feeling of exclamation.”


Haruo Shirane’s grammar book on page 241 states:

In the Heian period, the Nara period exclamatory final particle “kamo” was replaced by “kana,” which derived from the exclamatory final particle “ka” and the exclamatory final particle “na.”

And in separate entries on pages 239 and 240, ka” is listed as “final particle” and an “exclamation” and “na” is listed as a “final particle” that has two separate functions, one as an “exclamation” and the other as “seeking assurance, confirmation. “(The note below at the end of this article quotes the example usages the book has for them.)


So again, it is the use of an exclamation to bring the poetry into this haiku; by strengthening the the image of the village full of children with an exclamation, the reader understands what the most important part of the haiku is for Issa and now can share his wonder at the joy that the children are expressing as they play in the melting snow. Although Hass’ translation does perfectly mirror the sentiment that Issa expresses, it is important to understand that he still did add his own imagery in and ignored the verbal context that the exclamatory particle sets into the haiku.


Koretsky’s comments take everything that Hass did as being a literal translation of the haiku:

Notice how the poem sets up its first two ominous lines and then cuts, like a punchline, to its resolution. In Japanese, Issa would have had a variety – nearly 50 – ways to punctuate the end of the lines to build toward the joke, then to let us know it was time to relax and smile. Western poets, on the other hand, have about four end punctuations from which to choose. There is the dash, the ellipses, the comma, the colon.


It’s hard to understand what is “ominous” about the first two lines because it is hard to imagine that a village would be seriously flooded by the snow in it melting, but of course in the original this is something that isn’t there anyway. Nor is the break in the same place either, it is after “melting.” It is interesting to see in her list of end punctuational that she left out the exclamation point and the period, which are ones we often use in English.


This punctuation list is wherein the problem lies with this idea about the Japanese haiku always being “unresolved, (“the predominate tonal mode of the Japanophone’s ear” as Koretsky calls it.) Imagine if all you ever spoke could never come to a full punctuation stop, and that every verb you used had to be in present tense and you couldn’t express any self-emotion, then everything you said would be the ellipsis of some idea that left the resolution of what you were saying all in the listener’s mind because all you could ever present was an image of what you wanted to express. Wanting to write this style of haiku is one thing, trying to say that this is the only kind of haiku that the great Japanese masters produced is another.


The problem with haiku in the English language is that it has always been a monolith to mistaken ideas that translators have propagated about it. Robert Hass’ translations aren’t literal representations of these great writers in English, they are faint copies by someone who didn’t have the grammatical background to properly get all the poetry in them into his translations; because of this he was only left with the imagery and if there wasn’t enough imagery to fill out the poetry in the original he added in his own.


Hass wrote in his preface, “When the “hokku” became detached from linked verse….what was left was the irreducible mysteriousness of the images of themselves.” It is better to take this as an unconscious admission of his own inability to understand how  haiku, because of the imbedded grammar of kire-ji, layers poetic meaning onto the imagery. He only saw the concrete parts of haiku and never understood how these great writers were filling in the spaces around the imagery.


I’m not arguing that these three great haiku writers never wrote “unresolved” or “open ended” haiku if it was the sensibility they had at the time the wrote it, but they certainly wrote a lot more “connected” haiku with as much personal emotion and poetic technique as any conventional poetry in the west did.


The reason why they could do this is that they understood haiku as a verse form and not a single style of writing,  which is what what Hass’ translations aimed for and what Koretsky’s  article is arguing for.  The first hundred years of haiku in our language has been hampered and held down creatively by this and the wrong ideas about Japanese haiku it profligates. One has to wonder if the second century of it in our language will ever pull itself from the sand it continues to keep its head stuck into.


Note on Shirane’s examples.

On page 238 “ka” is listed as “final particle” and an “exclamation” with this example from the Kokinshu (no. 73, NKBT 8:117):

うつせみの世にも似たるか  Utsusemi no yo ni mo nitaru KA

How it resembles (nitaru ka) the ephemeral world (Utsusemi no yo)!


On page 240 “na” is listed as a “final particle” that has two separate functions, one as an “exclamation” and the other as “seeking assurance, confirmation.”

The example for the exclamation comes from the Kokinshu (Ono no Komachi, Spring 2, no. 113, NKBT 8:124):

花の色はうつりにけりな   Hana no iro wa utsurinikei NA

The colors (iro) of the cherry blossoms (hana) have faded completely (utsurinikerina)!

And for assurance and confirmation it is from the Gengi Monogatari (Yomogiu, NKBT 15:152:

ここは常陸の宮ぞかしな.  Koko wa, Hitachi no Miya zo kashi NA

This is (the residence of) the Hitachi prince, am I right?


“Kana” is still used in modern Japanese. This entry in the “A Dictionary of Intermediate Japanese Grammar” (The Japan Times, 1995) pg. 90 explains its usage:


Konshumatsu ni wa nani wo shiyo KANA

I wonder what I should do this weekend.


If you read “kana” in the Issa haiku as being spoken by a writer who isn’t looking at the scene he is describing, then it is possible to get the reading that Issa is wondering if the children are out at his old village which he has left. As Koretsky noted, Issa moved from his home in the mountains into the city at an early age so he could have been wondering about the action of this haiku instead of seeing it.

There are two extant versions of this haiku. One has “town” (町) instead of “village” (村) and the other has “town” as well as “sparrows” (雀) instead of “children” (子ども). Town indicates a more urban setting than village does so this might give credence to Issa being away from the mountains when he wrote it. It could also just prove that he just rewrote it and put kids and village in when he had made back to a rural area. You would need to know when and where Issa was at the time.


Buson’s Hokku: Butterfly, Temple Bell and the Taira Clan



釣鐘にとまりて眠る胡蝶かな          蕪村

tsurigane ni tomarite nemuru kotefu kana

Stopping to sleep on the temple bell……a butterfly!


The movement and imagery of this hokku is easy enough to understand, a butterfly flits into the quiet courtyard space of a temple and then lands on the big cast iron hanging bell that is there and falls asleep. Since the movement of the butterfly is expressed so quietly, we immediately relate it to the silence of the courtyard, which in turn amplifies the experience of having it be so silent that a butterfly could actually sleep in this kind of public place. We also are able to fill out this experience by realizing that the butterfly on the bell is only a temporary state of being and that it will be flying off, as all butterflies do, from the bell and out of the temple grounds sometime in the future.


Once we realize the temporal nature of the butterfly, we begin to see the hokku as starting to work as a deeper and symbolic message. Since the season word (kigo) is butterfly (kotefu), and this a butterfly that sleeps, it leads the into immediately recognizing an allusion to the Chinese sage Chuang Tzu’s famous dream of himself being a butterfly. The World Kigo Database has the translation from “The Chuang Tzu”, Chapter 2:

“Chuang Tzu dreamed that he was a butterfly.
All day long, he floated on the breeze
Without a thought of who he was or where he was going.

When he awoke, Chuang Tzu became confused,

“Am I a Man”, he thought,
“who dreamed that I was a butterfly?
Or am I butterfly, dreaming that I am a man?
Perhaps my whole waking life is
but a moment in a butterfly’s dream!.
This is a story of transformation”


When we read this allusion into the hokku, we now understand that it is an allegory for the human condition where every thing is transient and that life is illusionary. The butterfly, like us, is bound to be woke up from sleep. Are we that butterfly, or is that butterfly us?


Besides the Chuang Tzu’s butterfly, there is also another famous passage from Japanese literature that deals with the fleeting nature of human existence and worldly fame and this one has a temple bell as the image for it. That is the opening to the Heike Monogatari where the lines state:

“The sound of the Gion Shōja bells echoes the impermanence of all things; the color of the sāla flowers reveals the truth that the prosperous must decline. The proud do not endure, they are like a dream on a spring night; the mighty fall at last, they are as dust before the wind. — Chapter 1.1, Helen Craig McCullough’s translation”


Although Gion is a famous part of Kyoto, and most Japanese people believe that this is the setting for these lines, the temple bell being described here is really in India at the famous temple of Jetavana where Buddha is said to have given many of his teachings. The legend has it that the temple bells there were rang for those close to death to remind them of the impermanence of their lives and comfort them with the knowledge of knowing they will move into a new life.


It is generally accepted that the bells at a Jetavana Temple were either glass or silver and not the cast iron kind that are found in Japan. However, as noted above, the Japanese have always tended to think of the bells described in the “Heike Monogatari”  as being a hanging cast iron bell which most temples throughout Japan have.  So persuasive has this image of a hanging bell been for them that in 2004 the Japanese donated money to have a Japanese style bell built on grounds of the Jetavana temple ruins in India. Here is a photo of it and, except for the building structure,  it is a replica of a Japanese temple bell:



There is a pretty high chance that Buson also mistakenly thought that the ringing bell at the start of the ‘Heike Monogatari” was the kind he found at the temples in his own country. Once we accept this, we can move on into a further reading of this hokku.


The “Heike Monogatari” is a tale that deals with the fall of the Taira clan that came to power in the late 700s and was then later utterly defeated in the Genpei War which ended in 1185. The tale deals with the acts of the leader of the Taira clan, Kiyomori, who is quite ruthless in carrying out his dream of controlling the court and the politics of Japan, and goes through the wars that followed. Kiyomori summarily kills a lot of people in power and has no qualms about burning temples or persecuting monks. The opening lines of the tale are a Buddhist theme which is prevalent through out the story, that life (and political power) are fleeting and that evil acts in life will have an adverse consequence later in life.


The butterfly is the thing which ties the Taira family into this hokku. Every leading clan in Japan had a family crest (kamon) to identify themselves and their relationships, much the same as coats of arms were once used in Europe. The Taira clan lead by Kiyomori actually used a butterfly as their crest, this impressive swallowtail butterfly design:




Now, we able to read the hokku as an allegory for the Taira clan. Kiyomori with his dream of controlling Japan has come to sleep on the Buddhist bell which will soon ring the truth about how the prosperous and famous will fall into decline. And when the bell rings the butterfly will awake from its slumber and fly away with its dreams driven away like dust in the wind. Which is what happened to the Taira clan when it finally was defeated, it’s leading members killed and it’s lower ranking remnants escaping to the mountainous areas far from center of the power in Kyoto. We can now look at the hokku in a new light and change the translation:


On the temple bell…..the butterfly that stops to sleep!


If we turn the focus around by attaching the action to the butterfly, rather than cutting it into two as the first translation does, we can imply a more ominous tone of foreboding about what the fate of this butterfly might eventually be because it hints at how in the  near future the bell probably will be rung.


The picture that I used at the top of this blog post is of the bell at Renshoji Temple in Fujieda City which has a deep connection to a one of the most famous parts of the “Heike Monogatari,” the scene where Naozane Kumagai kills Kiyomori’s 16 year old nephew Atsumori Taira. Kumagai eventually repented this killing and became a monk of the Pure Land Buddhist sect. He went around preaching and building temples to atone for his life as a samurai and he built this temple in 1185 about 10 years after he reluctantly, and famously, beheaded Atsumori on the beach at Suma.


I happen to work in the building right in front of this temple and I have heard its bell rung from time to time. The next time I do hear it ring, there is going to be a lot more resonance in its tone because this hokku by Buson brings everything full circle for me. Not am I going to think of the Taira clan, I am going to think of Naozane Kumagai as well.


Buson Haiku: Komabune (The Korean Ships)


高麗舟のよらで過ぎゆく霞かな            蕪村

Komabune no yorade sugiyuku kasumi kana

Koma Ship/’s/without stopping/go pass/mist/(exclamation)

The ship from Koma doesn’t stop and passes…… the mist!

Koma is the Japanese shortened reading of the Chinese character for Goryeo, which was the name for the country of Korea from 936 to 1392. After a long 25 year war against the Mongolians the country signed a treaty in 1256 which made Goryeo a tributary ally of the Mongols for 80 years. After the power of the Mongolian empire waned in the 1350s, the country returned to being free of foreign influence and the name of it was changed in 1392 to Joseon by the general who had been the shadow leader of the country since 1388.

Most modern Japanese commentators on this haiku will back off from the reference to Korea and state that it just means any big continental sailing ship from another country that was different from the types of boats the Japanese used. The reason why the reference to Goryeo gets whitewashed out is because having it in the haiku gets in the way of the Shiki idea that everything must be a sketch of life around you. That the change in the country’s name predates Buson’s birth by about 320 years is the antithesis of this.

The commentators who read the reference to Goryeo as being relevant to the haiku tend to see the ship in terms of the country before the Mongols took power of. As Haruo Shirane explains:

“Komabune were the large Korean ships that sailed to Japan during the ancient period, bringing cargo and precious goods from the continent, a practice that had long since been discontinued by Buson’s time.”

So, to write a translation that follows this reading:

The merchant ship 
from Koma 
passes without stoping……
the mist!

Although the countries enjoyed good relationships in,the beginning of the Goryeo era, the political situation between the two countries was warlike for quite a long time after the Mongol lead invasions in the late 13th century. Their relationship wasn’t repaired until the regime change in 1392 to Joeson when the new Korean leaders began to make overtures towards the Japanese Shogunate. This led to the Koreans sending official emissaries for a wide variety of differing reasons into Japan until the late 1800s.

This blog (written in Japanese) has an impressive wood block print of visiting Korean ships in port at Murotsu in Hyogo and looks at this haiku in terms of the emissaries that Joseon occasionally sent into the Japan.



For certain, as shown by the wood block print, the coming of the emissaries was a memorable event. They came to Japan three times during Buson’s life, in 1719, 1748 and 1764, and since this is a Buson haiku from 1670s you have to wonder if this was something that he actually saw first hand. Especially in light of how the official government ships were the only ones he could have seen since for most of his adult life the Japanese foreign policy had closed its borders on the point of death, and trade between the two nations was confined to the island of Tsushima that lay in the strait between them.

To do a translation that looks at the ship being an official government vessel:

The elegant Koma
emissary ships
without stopping…..
the mist! 

The problem most Japanese people now have with this haiku is that they don’t understand what “yorade” mean because it is archaic. The commentators above took it to be an ancient verb construction of the verb “yoru” (to stop in, to drop by) being conjugated by “de” (which is the negative) to make the phrase “yorade” (without stopping).

However, if you check in a Classical Japanese dictionary, there is an entry for the word “yora” (夜ら) which is defined as meaning “yo” (night) plus “ra” (the suffix for the plural) which would translate into “nights.” “De” is an article that indicates space and time the way we do with “of,” “at,” or “in.”  So, since “yora” is written in hiragana, and the Chinese character isn’t used, then the phrase can also be read as “yora de,” which translates as “the nights of,” meaning that the haiku is employing a “kakekotoba” (pivot word)  word play that allows the reader to read this haiku with a double meaning.

A “kakekotoba is:

“A kakekotoba (掛詞?) or pivot word is a rhetorical device used in the Japanese poetic form waka. This trope uses the phonetic reading of a grouping of kanji (Chinese characters) to suggest several interpretations: first on the literal level (e.g. 松, matsu, meaning “pine tree”), then on subsidiary homophonic levels (e.g. 待つ, matsu, meaning “to wait”). Thus it is that many waka have pine trees waiting around for something. The presentation of multiple meanings inherent in a single word allows the poet a fuller range of artistic expression with an economical syllable-count. Such brevity is highly valued in Japanese aesthetics, where maximal meaning and reference are sought in a minimal number of syllables. Kakekotoba are generally written in the Japanese phonetic syllabary, hiragana, so that the ambiguous senses of the word are more immediately apparent.”

Although the wiki entry only states that it was used in “waka,” it was widely used in haikai as well. The reason why it has disappeared in haiku is mainly because of Basho who first studied this style to later abandon it. (Well, not completely, Donald Keene has written about Basho’s development in “World Within Walls” and states that the “Kisagata ya” haiku in “Oku no Hosomichi” (Narrow Road of Oku) is a “kakekotoba” that plays on the word “nebu” to have two readings.(pg.74))

To fill in this second reading of “the nights of,” we must turn towards the time when relationships between Japan and Goryeo were not as cordial. Specifically, when under the control of the Mongols the Koreans tried to invade Japan and there were many nights of many Korean ships In the waters around Japan. The first invasion in 1274 was with mainly Goryeo ships that were built for this purpose, since Mongolia is a land locked country and had no navy. The invasion failed because for some reason the forces that had gained a foothold in Japan were ordered back on the fleet ships that then sailed into a typhoon.

The second invasion was in the spring of 1281. By this time, the Mongolians had also taken control of Southern China and now had two navies to attack Japan from. The plan was for the Korea invasion force and the Chinese invasion force to meet and attack together. However, the fleet from Goryeo reached Japan first and decided to attack without waiting for the Chinese to arrive, but found that the Japanese had fortified their coastline so their attempts to breach it were futile. With the Chinese attacking by day and getting nowhere, the Japanese took to guerrilla warfare on the seas by night:

“In July 1281, the eastern army which carried 25,000 soldiers and 15,000 sailors aboard 900 ships attacked Tsushima and Iki before attempting to land at Hakata Bay without reinforcements. The Mongols launched attacks along the bay for about a week. The Japanese would respond to each attack with night raids. The Japanese ships would carry between 10 – 15 samurai and would close on the Mongol ships under the cover of darkness. When the ships were close enough the ship’s mast would be lowered to act as a bridge and the samurai would close in on the Mongol crew. The samurai excelled in close-quarter fighting and the Korean, Mongol and Chinese were no match for them in such tight quarters. On one occasion 30 samurai swam out to a Mongol vessel, decapitated the entire crew and swam back. Kusano Jiro led an attack in broad daylight and set a ship on fire even though he had his left arm cut off in the process. Kono Michiari also led a daylight raid in which the Mongols were led to believe they were coming to surrender. Instead of surrender Kono and his samurai boarded the vessel and captured a high-ranking general. These raids began unsettling the Mongols.”

It’s easy to imagine that the samurai in their boats were using the mists on the sea to cover their movements against the Goryeo fleet. Reading the phrase as “yoru de” also changes the position of the break in the haiku because it now makes a phrase out of “sugiyuku kasumi” which makes “de” as the point where the cut is:

The nights of 
        the warships from Koma……
the mists that
        go passing by!


Bowdoin University has a great site that digitalized the “Mōko Shūrai Ekotoba” which is a picture scroll that illustrates the two Mongol invasions of Japan. There are plenty of boats in the mist in it.